U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings
  • Advanced Search
  • Journal List
  • HHS Author Manuscripts

Logo of nihpa

A Motivational Theory of Life-Span Development

Jutta heckhausen.

Department of Psychology and Social Behavior, University of California, Irvine

Carsten Wrosch

Department of Psychology and Centre for Research in Human Development, Concordia University, Montreal, Québec, Canada

Richard Schulz

Department of Psychiatry and University Center for Social and Urban Research, University of Pittsburgh

This article had four goals. First, the authors identified a set of general challenges and questions that a life-span theory of development should address. Second, they presented a comprehensive account of their Motivational Theory of Life-Span Development. They integrated the model of optimization in primary and secondary control and the action-phase model of developmental regulation with their original life-span theory of control to present a comprehensive theory of development. Third, they reviewed the relevant empirical literature testing key propositions of the Motivational Theory of Life-Span Development. Finally, because the conceptual reach of their theory goes far beyond the current empirical base, they pointed out areas that deserve further and more focused empirical inquiry.

Most people have a sense of being actively involved in shaping their lives. They follow developmental paths that are coherent in terms of identifying and effectively pursuing long-term goals and, when necessary, disengaging from goals that are no longer attainable. Even when confronted with setbacks, disappointments, and failures, humans have a remarkable capacity to stay on course and maintain a sense of personal agency.

Our approach to the regulation of life-span development focuses on the impressive adaptive capacity of individuals to optimize development across major changes in the life course. The past 15 years of conceptual and empirical work have shown that a central feature of adaptive capacity is the regulation of motivation. An individual’s developmental potential is won or lost by mastering the challenges of regulating motivational processes. This is accomplished by selecting, pursuing, and adapting developmental and personal goals to reflect changes in life-course opportunities, staying ahead of the game by anticipating emergent opportunities for goal pursuits, activating behavioral and motivational strategies of goal engagement, disengaging from goals that have become futile and too costly, and replacing them with more appropriate goals.

In the early 1990s, we set out to capture these phenomena of adaptive regulation of development by proposing a life-span theory of control ( J. Heckhausen & Schulz, 1993 , 1995 ; Schulz & Heckhausen, 1996 ). This theory focused on the role of the individual as an active agent in life-span development, the distinction between primary and secondary control strategies, the proposition that primary control striving holds functional primacy in the motivational system, and the idea of selectivity and compensation as fundamental requirements of optimizing life course development. During the past 15 years, our original life-span theory of control was enriched by advancements in theory and empirical research on goal choice, goal engagement, and goal disengagement. In particular, the Model of Optimization in Primary and Secondary Control ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ) was developed to address how individuals choose goals in accordance with principles of developmental optimization. Moreover, the Action-Phase Model of Developmental Regulation ( J. Heckhausen, 1999 ; J. Heckhausen, Wrosch, & Fleeson, 2001 ; Wrosch & Heckhausen, 1999 ) describes the sequential structure of goal-oriented action cycles involving phases of goal selection, goal engagement, and disengagement in developmental regulation across the life course. The Motivational Theory of Life-Span Development presented in this article integrates the original life-span theory of control with these models and thus provides a comprehensive framework for the study of individual agency in life-span development. In a nutshell, our theory identifies the major challenges faced by individuals throughout the life course and the motivational and self-regulatory processes used to meet these challenges. We view the life course as being organized around a sequential series of action cycles that involve goal selection, goal pursuit, and disengagement from goals. Both optimal and nonoptimal strategies for each phase of this cycle are identified along with key transition points and relevant control strategies.

The goals of this article are fourfold. First, we identify a set of general challenges and questions that a life-span theory of development should address. Second, we present a comprehensive account of our Motivational Theory of Life-Span Development and discuss how the theory meets these challenges. Third, we review the relevant empirical literature, testing 15 key propositions of the Motivational Theory of Life-Span Development. Finally, because the conceptual reach of our theory goes far beyond the current empirical base, we identify several additional areas of inquiry to guide future empirical research.

General Challenges and Questions to Be Addressed by Life-Span Developmental Research

In our original life-span theory of control, we identified key issues that need to be addressed by all life-span theories of development ( J. Heckhausen & Schulz, 1995 ; Schulz & Heckhausen, 1996 ). Here we refine these propositions to lay the foundation for our Motivational Theory of Life-Span Development.

Criteria for Adaptive Development

Any effective theory of life-span development needs to specify which criteria it is using to differentiate desirable and adaptive from undesirable and maladaptive outcomes and patterns of development. Approaches to life-span development and aging vary widely with regard to the kind of criteria they use ( Schulz & Heckhausen, 1996 ). Depending on the focus of the scientific approach, successful development can be gauged through indicators of physiological functioning, such as cardiovascular and pulmonary status ( Rowe & Kahn, 1987 ), cognitive and intellectual performance ( Heine, Lehman, Markus, & Kitayama, 1999 ; Salthouse, 1991 ; Simonton, 1988 ), or achievement in physical ( Schulz & Curnow, 1988 ) or artistic domains ( Ericsson, Krampe, & Tesch-Römer, 1993 ; Lehman, 1953 ; Simonton, 1988 ). The common characteristic of all these criteria is that they reflect broad measurable standards of functioning or performance upon which members of a given society generally agree.

Moreover, these broad indicators can be applied to individuals at different ages using absolute standards (e.g., world record performance in 100-m dash) or relative standards (e.g., best 100-m dash performance for 60-year-olds) that take into account the specific constraints on the individual (e.g., age, disability, lack of training). Finally, such measurable indicators can also help to assess whether a specific individual shows developmental growth or decline relative to his or her own previous performance.

One difficulty with using single domain-specific standards of adaptation and mastery is that individuals usually cannot afford to invest in only one domain without seriously compromising mastery in other domains of life. Most individuals strike a balance by investing effort and time in multiple common life domains, such as education, work, social relations and family, health, and leisure activities. It is the overall mastery across different domains of life and functioning that defines the individual’s overall level of success.

Moreover, one can assess successful adaptation at two levels of analysis: one addressing mastery specific to the individual’s current location in a life-course trajectory and the other addressing the totality of mastery attained during the individual’s life. For example, pursuing a career as a world-class athlete may maximize mastery in a particular domain during the peak performance period in late adolescence and young adulthood but may seriously compromise the ability to master other domains or one’s health at later phases in life. Thus, criteria for adaptive development should be comprehensive in addressing multiple domains of functioning and the totality of mastery across the individual’s life span and should take into account the constraints on the individual that limit goal attainment.

Some researchers in life-span development have argued for more subjective and individualized criteria of psychological experience, such as life satisfaction or psychological well-being ( Baltes & Baltes, 1990 ). One variant of this approach is to conceptualize success in terms of goal attainment , subjectively defined as the realization of desired outcomes and the avoidance of undesired outcomes ( Marsiske, Lang, Baltes, & Baltes, 1995 ). A related variant of this subjective approach proposes that self-consistency is the ultimate criterion of adaptiveness and consequently views downward adjustments of goals and strivings for goal attainment as equivalent means for achieving self-consistency ( Brandtstädter & Rothermund, 2002 ). The common denominator of these more subjective approaches is the notion that adaptiveness is captured not so much by what a person does or accomplishes but, rather, by how a person perceives his or her accomplishments. These subjective approaches offer some appeal for those who follow a phenomenological orientation, but they come with serious drawbacks. First, subjective criteria are individually determined and thus cannot be used for interindividual comparisons of developmental outcomes. Second, they are subject to the rationalization biases individuals often use when they evaluate their own experiences and accomplishments. Third, subjective approaches fail to take advantage of the fact that there is substantial consensus across cultures about what constitutes success in life (e.g., physical, cognitive, intellectual, affective, and creative functioning; social relations; social status; integrity).

To summarize, an effective life-span developmental theory needs to include criteria for adaptive development that can be assessed in ways that facilitate interindividual comparison, prevent distortion by subjective biases, and build on cross-cultural consensus about what constitutes a successful life.

Individual Agency and Developmental Goals

Most developmental scientists would agree that individual agency plays a crucial role in human development across the life span ( Baltes, Lindenberger, & Staudinger, 1998 ; Brandtstädter, 2006 ; J. Heckhausen, 1999 ; Lerner & Busch-Rossnagel, 1981 ). Indeed, the active and goal-oriented role of individuals in their own development is a central proposition of the widely accepted organismic model of development ( Lerner, 2002 ; Reese & Overton, 1970 ; von Bertalanffy, 1968 ). The importance of agency has been further elaborated in models of intentional self-development, which use action theory to conceptualize the individual’s attempts to influence his or her own development (e.g., Brandtstädter, 2006 ; Brandtstädter, Wentura, & Rothermund, 1999 ; Heckhausen, 1999 ).

Humans develop mental representations about desired outcomes of life-course transitions and developmental processes. Often these desired outcomes are strongly influenced by what society has come to identify as a developmental task for a given age period or life-course transition ( Havighurst, 1952 ). These desired outcomes or developmental tasks are adopted by the individual as developmental goals toward which to strive and can thus organize the active attempts of individuals to influence their own development. Many developmental researchers therefore focus on goal-related concepts when investigating individual contributions to life-span development. A variety of different terms have been used to characterize these goals, including personal projects ( Little, 1983 ; Little, Salmela-Aro, & Phillips, 2007 ), life goals ( Nurmi, 1992 , 1993 ), personal goals ( Brunstein, 1993 ; Brunstein, Schultheiss, & Maier, 1999 ; Riediger, Freund, & Baltes, 2005 ; Salmela-Aro, Aunola, & Nurmi, 2007 ; Wadsworth & Ford, 1983 ), personal strivings ( Emmons, 1986 ), personal life tasks ( Cantor & Fleeson, 1991 ), goals of intentional self-development ( Brandtstädter, Wentura, & Rothermund, 1999 ), and possible selves ( Cross & Markus, 1991 ; Markus & Nurius, 1986 ). The empirical research on these goal-related concepts reflects the specific challenges associated with human goal-related striving in the context of the life course.

In general, development-related goal concepts share three characteristics that make them particularly suited for the life-course context. First, developmental goals are directed at developmental processes (e.g., become more independent from my parents) or life-course attainments (e.g., start a career, get married). This implies that the unique action field for developmental goals is the life course with its specific age-graded structure of opportunities and constraints (see the next section). Second, developmental goals comprise desired outcomes at an intermediate level of aggregation (e.g., improve my grades, graduate from college, have a child), between very specific projects (e.g., get an A on the next exam), and broad values (e.g., promote world peace) or motives (e.g., improve my overall mastery). Third (related to the second point), developmental goals typically reach into the intermediate future, 5–10 years ahead, either within the current or next phase of the life course (e.g., within adolescence or from adolescence into early adulthood).

To summarize, an effective life-span developmental theory should view the individual as an active agent in life-span development. Thus, individual agency should be studied by addressing motivational processes involved in goal selection, goal pursuit, and goal disengagement.

Changing Opportunities and Constraints Across the Life Course

Individuals have to adjust to, cope with, and take advantage of the changing opportunities and constraints characteristic of different stages in life. Biological maturation and aging and societal institutions (e.g., education, labor market, retirement) set up a roughly inverted U-shaped curve of control capacity across the life span, with a steep increase during childhood and adolescence, a peak in young adulthood and middle age, and a decline in old age. This general life-course trajectory of first increasing and then decreasing opportunities is overlaid with more domain-specific trajectories of improving and declining opportunities for achieving specific developmental goals. Societal institutions, such as the educational system, vocational career patterns, and welfare systems, structure the life span in terms of critical transitions (e.g., school entry, promotions, retirement) and sequential constraints (e.g., educational qualifications as prerequisites for certain careers). These time-organized opportunity structures present significant regulatory challenges to the individual who must respond in a time- or age-sensitive way. Moreover, the individual needs to come to terms with diminished chances of attaining important life goals, once the opportunities pass by. In summary, any effective theory of life-span development needs to address the way in which life-course variations in opportunities and constraints are met with individuals’ attempts to master their own development.

Selectivity and Compensation as Fundamental Regulatory Challenges

Major approaches to life-span development converge in asserting that the regulatory challenges encountered throughout the life course require that the individual masters two fundamental regulatory challenges: selectivity of resource investment and compensation of failure and loss ( Bäckman & Dixon, 1992 ; Baltes & Baltes, 1990 ; Baltes et al., 1998 ; Brandtstädter, 2006 ; J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ; Salthouse, 1985 ).

Selectivity of goal investment acknowledges the fact that we cannot strive for all goals at once, or even sequentially. Paul and Margret Baltes’s model of selective optimization with compensation championed the idea of selectiveness in life-span development, particularly for successful aging ( Baltes, 1987 ; Baltes & Baltes, 1990 ). The human potential for controlling the environment is multifaceted but resource- and time-limited; as a result, people have to be selective about which goals to pursue and when they pursue them. This implies that they relinquish goals that overstretch or might undermine their capacity to reach specific long-term goals. For example, giving up on postsecondary education may help an athlete’s career in the short run but may compromise his or her potential for effectively influencing his or her environment in the long run. Another more domain-specific example is how individuals exhibit socioemotional selectivity in which social partners they select and maintain at different times of life ( Lang, 2001 ; Lang & Carstensen, 1994 ; Lang & Heckhausen, 2006 ), depending on whether the life phase requires access to new information or socioemotional well-being ( Carstensen, Isaacowitz, & Charles, 1999 ).

Compensation of failure and loss is essential for developmental regulation, because humans experience setbacks in their goal striving not only in old age ( Salthouse, 1985 ) but also normatively across the entire life span ( Bäckman & Dixon, 1992 ; J. Heckhausen, 1999 ). Mastery development is maximized at intermediate levels of difficulty, when failure occurs at about 50% of attempts. Thus, development of mastery cannot thrive unless individuals have effective means of dealing with failure, both in terms of correcting their behavior and in terms of protecting their motivational and emotional resources against the undermining effects of failure (e.g., loss of hope for success, decline in self-esteem, hopelessness). Life-span developmental psychologists have focused on different aspects of compensation, with some primarily addressing attempts to hone action strategies to overcome and undo previous failures (e.g., Bäckman & Dixon, 1992 ) and others focusing on how individuals prevent or counteract negative affective or self-evaluative consequences of failure. For example, the accommodative tendencies, investigated by Brandtstädter et al. (1999) , help the individual adjust goals to what is feasible and protect the individual against self-blame for failure. In sum, an effective life-span developmental theory needs to address processes that help the individual to select appropriate goals in which to invest and to compensate for failures, setbacks, and losses when they occur.

The Motivational Theory of Life-Span Development

In this section, we discuss how the Motivational Theory of Life-Span Development addresses the major challenges raised in the previous section. We subsume under the theoretical umbrella of our Motivational Theory of Life-Span Development the original life-span theory of control ( J. Heckhausen & Schulz, 1993 , 1995 ; Schulz & Heckhausen, 1996 ) and its elaboration in two related process models: the Model of Optimization in Primary and Secondary Control ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ), which addresses the control processes involved in goal engagement and goal disengagement, and the Action-Phase Model of Developmental Regulation ( J. Heckhausen, 1999 ; J. Heckhausen et al., 2001 ; Wrosch & Heckhausen, 1999 ), which addresses the sequential structure of goal engagement and disengagement across the life course. Our original life-span theory of control put forward propositions about primary control as the criterion of adaptive development and about life-span trajectories of primary and secondary control, which are addressed in the first two following sections. Subsequent sections greatly expand the reach and specificity of the original theory by incorporating empirical findings and conceptual developments (i.e., Optimization in Primary and Secondary Control and the Action Phase Model of Developmental Regulation) that have occurred over the past 15 years.

Primary Control Capacity as Criterion of Adaptive Development

Our Motivational Theory of Life-Span Development proposes that the key criterion for adaptive development is the extent to which the individual realizes control of his or her environment (i.e., primary control) across different domains of life and across the life span ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1995 , 1999b ; Schulz & Heckhausen, 1996 ). To further elaborate this proposition, we adapted a conceptual distinction, first made by Rothbaum, Weisz, and Snyder (1982) , between primary and secondary control processes. According to Rothbaum et al., primary control processes are conceptualized as directed at changing the world to bring the environment into line with one’s wishes. In contrast, secondary control processes are defined as changing the self to bring oneself into line with environmental forces. The two processes together are proposed to optimize an individual’s sense of control, even when circumstances constrain the individual’s capacity to control the environment.

Using Rothbaum et al.’s (1982) basic distinction between primary and secondary control, our life-span theory of control specified their functional relations more explicitly and formulated their implications for life-span development. According to our life-span theory of control, the motivational system is set up to maximize primary control capacity across life domains and lifetime ( J. Heckhausen & Schulz, 1993 , 1995 , 1999b ; Schulz & Heckhausen, 1996 , 1997 ). From a functionalist and evolutionary psychology perspective, primary control striving is essential for mastering the challenges associated with maximizing inclusive fitness, such as foraging for food, seeking shelter, competing for mates, and caring for offspring ( J. Heckhausen, 2000b ; J. Heckhausen & Schulz, 1999b ). Moreover, primary control striving is promoted by basic motivational modules that have been favored in mammalian evolution ( J. Heckhausen, 2000b ): a preference for behavior-event over event-event contingencies ( White, 1959 ), a ubiquitous tendency for novelty exploration ( Schneider, 1996 ), and the asymmetry of emotional responses to positive and negative events ( Frijda, 1988 ). The latter pattern of responses reflects stronger and more prolonged aversive affective responses to negative events when compared with the beneficial affective consequences of positive events, a pattern that effectively promotes primary control striving and avoids “resting on one’s laurels.” Thus, behavioral evolution has favored mechanisms of motivational self-regulation that maximize primary control striving.

Primary and secondary control processes work together to maximize the overall primary control capacity of an individual. Primary control capacity varies across domains and age and reflects individuals’ ability to influence important outcomes in their environment. At any given point in the life span, development is adaptive to the extent that it realizes a maximum of primary control, taking into account not only the current ability to control external events but also the future potential for exercising primary control. For example, an expansion of control in one domain, such as gymnastics, would not be optimal if it seriously compromises control in the future because of impaired skeletal growth. The primacy of primary control principle would require a disengagement from goals with such negative side effects for a person’s long-term primary control capacity. In other words, the most adaptive development across the life course is achieved by maximizing primary control in the multiple major domains of functioning (e.g., work, family, health, leisure) and across the different phases of the life span.

The life-span theory of control identifies the function of secondary control more specifically than did Rothbaum et al. (1982) . According to our model, secondary control strategies address internal, most notably motivational, processes to minimize losses in, maintain, and expand existing levels of primary control. Thus, we conceptualized secondary control strategies as auxiliary motivational processes that support short-term or long-term primary control striving, not as alternatives or even processes opposed to primary control.

The proposition that secondary control processes serve primary control striving proved to be an important point of departure for our theory when compared with the earlier work of Rothbaum et al. (1982) . This led us and others (e.g., Bailis, Boerner, Chipper-field, Gitlin, Hall, Light, Isaacovitz, McQuillen, Salmela-Aro, Wahl) following our theoretical framework on to a path quite distinct from investigators who adopted the older view that secondary control processes are solely directed at acceptance, giving up, and fitting in ( Morling & Evered, 2006 ; Morling, Kitayama, & Miyamoto, 2003 ; Skinner, 2007 ; Thompson, Soboloew-Shubin, Galbraith, Schwankovsky, & Cruzen, 1993 ). It is important to note in this context that our conception of primary and secondary control processes was from the beginning focused on control striving and thus motivational phenomena, rather than merely at perceptions of control , a phenomenon of social cognition that used to be the most commonly addressed aspect of control behavior in the 1980s and early 1990s (see review in Skinner, 1996 ).

The life-span theory of control views humans universally as motivated by achieving effects in their environment ( White, 1959 ). We set out to investigate how individuals manage to maintain an active agenda of striving for primary control as they encounter great challenges during their life course in terms of both gains and losses in actual control potential. As reported in the section on life-span trajectories of control striving below, primary control striving remains stable and a dominant motivational source throughout adulthood and into older age ( J. Heckhausen, 1997 ).

Life-Span Trajectories of Primary and Secondary Control

Our life-span theory of control proposed hypothetical life-span trajectories of the availability of primary control and use of secondary control strategies (see Figure 1 ; Schulz & Heckhausen, 1996 ), based on an analysis of control resources at different times during the life course. As primary control capacity increases, plateaus, and then decreases across the life span, individuals keep trying to maximize overall primary control ( J. Heckhausen, 1999 ). According to the life-span theory of control, the striving for primary control is a constant and universal motivational drive throughout the life course. However, as individuals’ capacity for primary control decreases in old age, they typically need to invest more effort in striving for primary control goals and may need to activate secondary control strategies (e.g., anticipate and imagine success, enhance perceptions of personal control) that help them stay committed in spite of the challenges they face. Moreover, as certain primary control goals become unattainable, individuals need to disengage from them in favor of pursuing other more attainable goals. In this process, individuals increasingly resort to secondary control strategies of adjusting expectations, values, and attributions so that losses in primary control are not undermining the individual’s motivational resources for primary control striving in general.

An external file that holds a picture, illustration, etc.
Object name is nihms171079f1.jpg

Hypothetical life-span trajectories for primary control potential and primary and secondary control striving. From Developmental Regulation in Adulthood: Age-Normative and Sociostructural Constraints as Adaptive Challenges , by J. Heckhausen, 1999 , Figure 3.1., p. 72. Copyright 1999 by Cambridge University Press. Adapted with permission.

The Life Course as a Field of Action

Action-oriented approaches, including our own, view the individual as an active producer of his or her own development ( Brandtstädter, 1998 ; Freund & Baltes, 2002b ; J. Heckhausen, 1999 ; Lerner & Busch-Rossnagel, 1981 ). For such an agent in his or her own development, the life course is a field of action that has a time-organized structure of opportunities and constraints ( J. Heckhausen, 1999 ).

Our Motivational Theory of Life-Span Development proposes that the individual’s attempts to regulate his or her own development is organized in cycles of action around the pursuit of developmental goals ( J. Heckhausen, 1999 ). Developmental goals are the organizing motivational units that enable individuals to take an active role in shaping their own life course and development. Developmental goals are similar to other goals in that they are anticipated end states that exert a directional influence on an individual’s behavior.

Not all goals can be pursued at all times of life. In the long-term or macro level of aggregation, biological change and societal age grading of opportunities create a curve of individual control capacity that resembles an inverted U-function. Biological maturation and aging, societal age grading (e.g., going to school, retirement), and social norms about age-appropriate behavior and developmental milestones create a timetable of developmental opportunities, several of which are considered to be normative developmental tasks ( Havighurst, 1953 ). These persist in modern industrial societies, even though for some developmental tasks, particularly regarding the family cycle, normative age-ranges have become somewhat broader (e.g., age of first parenthood), and certain transitions (e.g., moving in with one’s romantic partner, marriage, stable employment) have become decoupled ( Brueckner & Mayer, 2005 ).

In spite of these changes, the human life course still offers an age-graded sequence of increasing and decreasing opportunities to pursue and attain important developmental goals, as illustrated in Figure 2 . As individuals move through the life course, they encounter emerging, peaking, and declining opportunities to strive for certain developmental goals, such as graduating from school, getting married, becoming established in a career, having and bringing up children, or buying a house. These opportunities can cover narrow (e.g., school graduation) or wide (e.g., becoming a grandparent) time windows in the life course. They overlap with each other in conducive (e.g., marriage, first child) or conflicting (e.g., career, first child) ways ( Wiese & Freund, 2000 ) and can form sequentially organized paths (e.g., education, career). As a whole, these trajectories of opportunity for goal striving provide the individual with a timetable that guides goal choice and pursuit.

An external file that holds a picture, illustration, etc.
Object name is nihms171079f2.jpg

Age-graded sequencing of opportunities to realize various developmental goals. From Motivational Psychology of Human Development: Developing Motivation and Motivating Development , by //J. Heckhausen, 2000, Figure 1 , p. 215. Copyright 2000 by Elsevier. Adapted with permission.

It is important to note here that the age-related structuring of the life course itself is subject to historical change ( J. Heckhausen & Schulz, 1999a ). Age boundaries for key life-course phases, such as education and child bearing, have changed dramatically over the past few centuries. In most industrialized countries today, formal education extends well into the late teen years and early 20s, as opposed to the midteens a century ago, and childbearing typically begins and ends at later ages than it did 2 centuries ago. Within the past 150 years, many societies have added an entirely new life-course phase, retirement, as a result of increased longevity and enhanced social mobility. Overall, the trend historically has been toward increased variability and flexibility in life-course trajectories, although this trend has not undone a fundamental structure in the sequencing of life-course events and transitions ( Brueckner & Mayer, 2005 ). Life-course trajectories will continue to evolve as societies and human populations change in the future ( Blossfeld & Huinink, 2000 ; Hagestad & Neugarten, 1985 ; Mayer, 2004 ).

Optimization of Development by Adaptive Goal Choice

In societies with a highly specialized labor force and substantial social mobility, chronological age itself does not automatically propel progression through this timetable of developmental tasks. It is up to the individual to take up the challenge and adopt specific developmental tasks as personal goals ( J. Heckhausen, 1999 ). Only if the individual commits to a specific personal goal for development can developmental tasks be mastered. This also implies that the individual has to determine when the time is right for committing to a certain goal, such as finding one’s romantic partner, having a child, or choosing a career. Thus, a theory of developmental regulation needs to include a higher level regulatory process of goal selection that involves specific heuristics to take into account the available opportunities, time constraints, and long-term consequences of investing in a particular primary control goal. In our model of developmental regulation, this metalevel selection process is referred to as optimization ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ).

In contrast to other life-span developmental models, such as the dual-process model ( Brandtstädter & Rothermund, 2002 ) and the selection optimization and compensation (SOC) model ( Baltes & Baltes, 1990 ; Freund & Baltes, 2000 , 2002a , 2002b ), our Motivational Theory of Life-Span Development does not propose that specific processes of control striving, assimilation, accommodation, selection, or compensation are adaptive per se. Such processes and strategies are “blind” to the fit of a given goal with opportunities, never mind its potential consequences for other goals. Functionality of a control strategy cannot be determined by the strategy as such, independent of the situation to which it is applied. Instead, the functionality of a given control strategy is determined by its match with the opportunities and possible tradeoffs with other primary control domains and long-term con-sequences.

Whether a control strategy is adaptive can only be determined by examining whether it will help optimize an individual’s multidomain and long-term capacity for primary control. Therefore, adaptive control strategies reflect engagement with goals that can be attained realistically in the current developmental ecology and that do not have excessively detrimental consequences for control striving in other domains or for the attainment of future goals. More specifically, primary control striving for a particular goal is adaptive if three requirements are met: (a) congruence of goal and opportunity, (b) consequences for other domains or long-term development are beneficial or at least not detrimental, and (c) a minimum diversity of goals is preserved. Regarding goal–opportunity congruence , individuals need to take into account and use as “adaptive challenges” ( J. Heckhausen, 1999 ) the constraints and opportunities that biological maturation and aging and the societal organization of the life course offer in a given social ecology ( J. Heckhausen & Schulz, 1999a ). The management of interdomain and long-term consequences becomes important when goals in different domains are interrelated in a beneficial or detrimental way. For example, heavy investment in one domain (e.g., career) can deprive other domains (e.g., family) from needed action resources for viable developmental progression. Thus, the choice of and degree of investing in a particular goal must be viewed in the broader context of how this will impact the pursuit of other goals both concurrently and in the future. Finally, goal diversity is needed to avoid exclusive dependence on one domain or goal pursuit. A narrowing down of investment in only one domain can expose the individual to developmental dead ends should the chosen goal become threatened or futile ( Linville, 1987 ). Therefore, a certain level of diversity in goal pursuit needs to be maintained, even in older age. These three issues are addressed by what we have proposed to be the three major heuristics involved in the optimization of goal choice: match goals to opportunities , manage interdomain and long-term consequences , and maintain diversity of goals .

Control Strategies Involved in Goal Engagement and Goal Disengagement

In our model of Optimization in Primary and Secondary Control (OPS model), we originally proposed a classification scheme that was built on the two major regulatory challenges of life-span development: selection and compensation ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ). In a 2 (primary/secondary) × 2 (selection/compensation) matrix, we differentiated between selective primary and selective secondary control strategies and between compensatory primary and compensatory secondary control strategies. In empirical studies that used the OPS model to investigate adaptation to specific life-course transitions ( J. Heckhausen, et al., 2001 ; Wrosch & Heckhausen, 1999 ; Wrosch, Schulz, & Heckhausen, 2002 ), it soon became clear that across these dimensions, control strategies operate together in a goal-engagement mode on the one hand and a goal-disengagement mode on the other hand ( J. Heckhausen, 2003 ; J. Heckhausen & Farruggia, 2003 ). Table 1 provides an overview of goal-engagement-related and goal-disengagement-related control strategies.

Control Processes Involved in Goal Engagement and in Goal Disengagement

Once a developmental goal is chosen by metamotivational processes of optimization, a specific set of control strategies that comprises goal engagement is activated ( J. Heckhausen, 1999 ; Wrosch & Heckhausen, 1999 ). Typically, goal engagement involves selective primary control and selective secondary control. Selective primary control refers to the investment of behavioral resources (i.e., time, effort, skills) into pursuing a goal. Selective secondary control serves to enhance and maintain motivational commitment to a chosen goal, particularly when the goal is challenged by unexpected obstacles or attractive alternatives. Selective secondary control strategies include enhanced valuation of the chosen goal and devaluation of nonchosen alternatives, as well as positive illusions about one’s control potential for achieving the chosen goal. In addition, compensatory primary control may be required when available behavioral resources of the individual are insufficient to attain the goal, and external resources have to be recruited. Specifically, compensatory primary control addresses the recruitment of help or advice from others, the use of technical aids (e.g., assistive devices, such as a wheelchair), or the employment of unusual behavioral means typically not involved in the activity (e.g., lip reading to compensate a hearing disability). Applied to the example of striving for a career promotion, the person who has set this goal for him- or herself will invest more time and effort into work (i.e., selective primary control), imagine the positive consequences and pride that would come with achieving the promotion (i.e., selective secondary control), and seek advice from more senior colleagues on effective strategies to foster career success (i.e., compensatory primary control).

When the individual experiences a loss of control and when the goal becomes unattainable or excessively costly, the individual needs to disengage from the goal ( J. Heckhausen & Schulz, 1993 ; Wrosch, Scheier, Carver, & Schulz, 2003 ). In contrast to the motivational mindset of goal engagement, goal disengagement involves compensatory secondary control strategies. Compensatory secondary control can be attained by deactivating the obsolete goal, thus freeing up resources for the pursuit of other goals that are attainable. In addition, compensatory secondary control includes specific self-protective strategies, such as self-protective causal attribution (avoiding self blame), focusing on successes in other domains, and downward social comparisons, all of which should deflect the potential negative effects of failure experiences on important motivational resources, such as affective balance and self esteem. Converging concepts are proposed by self-regulation and control theory ( Carver & Scheier, 1998 , 2000 ), which argues that as much as commitment and confidence are part and parcel of goal engagement, active disengagement involving the reduction of commitment and deflated confidence are required to relinquish goals. Simply withdrawing effort without breaking up the motivational commitment would have maladaptive consequences. Thus disengagement is an active process of restructuring one’s goals, rather than merely a passive reflection of failure and loss.

Action-Phase Model of Developmental Regulation

How do cycles of goal engagement and disengagement with developmental goals unfold over a lifetime? As individuals move along the age axis of the life span, opportunities to strive for specific goals emerge, peak, decline, and disappear (e.g., graduate from school, establish a long-term partnership, have a child; J. Heckhausen & Farruggia, 2003 ). Striving for primary control requires a repeated adaptation of one’s goal selections and control strivings to the objectively available opportunities and constraints in the given developmental ecology. The patterns of goal engagement and disengagement, along with their respective control strategies, should mirror these changes.

Congruence of changes in opportunities and phases of goal engagement and disengagement

Figure 3 illustrates this adaptive congruence between opportunities and goal engagement and disengagement. The figure displays the rising, peaking, and falling trajectory of opportunities to reach a certain goal (e.g., having a child). The figure also shows the expected trajectory of goal engagement required to attain a goal. The increasing trajectory of opportunities to attain an important developmental goal (see Figure 3 , light grey area) prompts the individual to consider adopting it as a personal goal for development and thus puts the process of optimized goal choice into action (see Figure 3 , first segment on left). In cases where the individual postpones goal selection and goal pursuit to nonoptimal times when opportunities have peaked and started to decline, higher levels of goal engagement (see Figure 3 , dark grey area indicating a high peak) are required to safeguard goal attainment in the face of diminishing opportunities.

An external file that holds a picture, illustration, etc.
Object name is nihms171079f3.jpg

Age-graded opportunity structure and goal engagement for developmental goals. From Motivational Psychology of Human Development: Developing Motivation and Motivating Development , by //J. Heckhausen, 2000, Figure 2 , p. 215. Copyright 2000 by Elsevier. Adapted with permission.

Figure 4 links goal cycles with appropriate control strategies ( J. Heckhausen, 1999 ). During goal choice and before passing the decisional Rubicon ( H. Heckhausen, 1991 ), optimization heuristics of matching opportunities, considering consequences, and maintaining diversity are activated. Once the Rubicon is passed, the person moves into a goal-engagement phase, which involves the investment of selective primary and selective secondary control. As the person gets closer to the point where opportunities become severely constrained (e.g., biological deadline), goal engagement becomes more urgent and intense, which should be reflected in increased use of selective primary and secondary control and compensatory primary control strategies (see Figure 4 , “urgent goal engagement” segment to the left of the “deadline” transition). As opportunities for goal attainment decrease, they may reach a point where goal attainment becomes close to impossible and/or very costly, thus rendering further striving for the goal highly dysfunctional in terms of individual resource allocation. This is the point of a developmental deadline. Once the deadline has been passed without attaining the goal, the individual needs to disengage from the goal and use compensatory secondary strategies to protect his or her motivational resources for future goal pursuits (see Figure 4 , segment to the right of vertical “deadline”).

An external file that holds a picture, illustration, etc.
Object name is nihms171079f4.jpg

Action-phase model of developmental regulation. From Developmental Regulation in Adulthood: Age-Normative and Sociostructural Constraints as Adaptive Challenges , by J. Heckhausen, 1999 , Figure 5.1., p. 114. Copyright 1999 by Cambridge University Press. Adapted with permission.

Developmental deadlines are important markers in this process and guide individuals’ decisions for goal disengagement but also exert an urgency influence before being passed. The informational advantage of anticipating a deadline can be substantial. Without it, individuals could stumble into situations of uncontrollability and futile goal investment that leads to depressive symptomatology ( Klinger, 1977 ; Nesse, 2000 ). That said, developmental deadlines provide a formidable challenge to individuals’ developmental regulation because they require the individual to shift from an urgent and intense engagement with a goal before hitting the deadline to disengagement and self-protection after passing the deadline. Individuals who fail to disengage from the futile goal after passing the deadline run the risk of wasteful investment of resources, frustration, opportunity costs of not pursuing other feasible goals, and depression.

Although we emphasize deadline-related goals in our model, it applies equally well to any situation where goal opportunities shift over time. For example, a student who chooses to pursue a major that in the course of the freshman year proves too difficult for his or her intellectual capacities would be well advised to change the major to a field of study that better matches his or her specific intellectual talents.

The regulatory challenge in these cases lies in identifying when goal pursuit is maladaptive while it is still ongoing and the individual is fully engaged. In such situations, it seems necessary that nonbiased, reality-oriented monitoring processes operate in the background, allowing the individual to disengage from goals that are no longer feasible or desirable. In fact, there is evidence that such monitoring processes occur, influenced by the cognitive and emotional concomitants of difficulty with goal pursuit. For example, individuals who confront goal failure or perceive insufficient progress toward an important goal are likely to experience emotional distress ( Carver & Scheier, 1990 , 1998 ; Higgins, 1987 ; D. Watson, Clark, & Tellegen, 1988 ) and/or a decline in positive affect ( J. Heckhausen, Carmody, Haase, & Poulin, 2008 ; Nesse, 2000 ). Further, the undesirable change in affect arising from difficulty with goal pursuits, can affect goal-directed behaviors and even lead to its termination.

In such circumstances of deteriorating affect during unsuccessful goal engagement, theories of personality functioning and self-regulation have proposed that people typically step outside their goal-pursuing focus and reevaluate the situation (e.g., Carver & Scheier, 1998 ). These monitoring and goal-reevaluation processes have to occur under a volitional mindset, which typically shields an individual against information that could interfere with goal attainment ( Achtziger & Gollwitzer, 2008 ; Gollwitzer, Heckhausen, & Steller, 1990 ). Therefore, individuals may have to switch into a motivational (in other words, a more reality-oriented mindset) to more objectively evaluate the probability of successful goal attainment. Such a shift from a volitional mindset directed at the implementation of goal pursuit to a motivational mindset that deliberates the validity of one’s goal choice may occur when failure in goal progress has become hard to ignore (i.e., multiple failures, high costs) and the associated increase in negative affect or decrease in positive affect reaches a certain threshold ( Carver & Scheier, 1998 ). In addition, a shift from a volitional to a motivational mindset may occur without the experience of goal failure or negative affect. For example, people may consciously decide to reassess at specified intervals (e.g., 6 months after starting goal pursuit) or after certain occasions (e.g., after trying different strategies) the rationality of continued goal pursuit. For example, the student who chooses a very challenging major might plan to reassess the rationality of this decision after the first semester or after testing his or her capacity in a difficult course.

Goal disengagement can also occur as a result of deliberate evaluations of the consequences of continued goal pursuit on a person’s overall development. As discussed earlier, optimization processes cause one to consider the impact of goal pursuit on multiple domains of life and long-term developmental outcomes. Thus, goal engagement is driven not only by self-assessments of progress toward goal attainment but also by the effects of goal pursuit on other important life domains. For example, a person may conclude that achieving a particular goal (e.g., making more time for leisure activity) is well within reach, but the costs of achieving it are too high in terms of their impact on other life domains (family life or career).

Discrete action phases orchestrate control strategies

To conceptualize the adaptive progress of the individual through these changes in goal engagement, a sequentially organized model of action-phases is needed. Such a model exists in general motivational psychology, the Rubicon model of action phases, which divides the action cycle into several phases, each with different functions and accordingly adapted motivational mindsets ( Achtziger & Gollwitzer, 2008 ; Gollwitzer et al., 1990 ; H. Heckhausen, 1991 ; H. Heckhausen & Gollwitzer, 1987 ). Using the Rubicon model as a starting point, we developed an action-phase model of developmental regulation, which expands the Rubicon model in several ways ( J. Heckhausen, 1999 ). First, our model adds another major transition into the action cycle, the transition from predeadline to postdeadline. Second, our model assumes an adaptive congruence of action phases with changes in situational opportunities for goal attainment. Third, our model includes expectations about specific optimization and control strategies involved in the phases of goal choice, goal engagement, urgent goal engagement, and goal disengagement. Fourth, our model includes the postactional phase of either meeting the deadline or failing to meet it. Finally, the model is specifically developed to address long-term goal pursuit in the context of life-span development, but it can also be applied to nondevelopmental action cycles.

The three key ideas of our action-phase model (shared with the Rubicon model) are the following: (a) Shifts between action phases, that is from goal choice to goal engagement and from goal engagement to goal disengagement, are not gradual but discrete and radical. (b) In each action phase, multiple control strategies are orchestrated to maximize the effectiveness with which the motivational function of the respective action phase is realized. (c) Within each action phase a specific motivational mindset shapes characteristics of information processing to optimize the effectiveness of the respective action phase. As part of the motivational mindset, perceptions of control shift also. Specifically, the phases of the action cycle and the control strategies involved in each phase are represented in Figure 4 . The following two sections address the two major transitions in the action-phase model of developmental regulation, from optimized goal choice to goal engagement ( Rubicon ) and from goal engagement to goal disengagement ( deadline ).

Decisional Rubicon: From goal selection to goal engagement

The first shift, in this case from deliberation of goal options to engagement with a chosen goal, occurs when the decision to engage with a certain goal has been made, and thus the decisional Rubicon ( H. Heckhausen, 1991 ) has been crossed. During the phase of optimization (see Figure 4 , left segment) preceding the decision on goal choice, the individual should take into account the availability of opportunities for the various goal options in her/his age-specific developmental ecology ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1999a ). Regarding controllability, personal expectations should be cautious and realistic, rather than enhanced or pessimistic. In this way, the biological and societal conditions prevalent at the age and social position of the individual have a major influence on which goals are selected.

Once the decisional Rubicon is crossed, the motivational system shifts from a mindset of deliberation to a mindset of implementation ( Achtziger & Gollwitzer, 2008 ; Gollwitzer, 1990 ; Gollwitzer et al., 1990 ). The deliberative mindset is relatively impartial, broad, and unbiased, so that decisions are likely to be more realistic and adapted to actual controllability. In contrast, the postdecisional mindset of implementation does not allow for questioning the decision but narrowly focuses on the implementation of the planned action. The deliberative and implemental mindsets contrast with regard to memory for deliberation versus implementation-relevant information, the breadth of attentional focus (i.e., broad before and narrow after the Rubicon), openness to new information, and perceived control ( Gollwitzer, 1990 ). 1 Regarding the latter, control perceptions are realistic in the predecisional phase but enhanced after goal commitment, so that the commitment to goal pursuit is strengthened ( Gollwitzer & Kinney, 1989 ; Taylor & Gollwitzer, 1995 ). Numerous experimental studies have demonstrated such shifts from deliberative to implemental mindsets ( Achtziger & Gollwitzer, 2008 ).

Developmental deadline and constraints: From engagement to disengagement

The second shift in the action-phase model of developmental regulation is the one associated with a radical decline in opportunities for goal attainment, namely the developmental deadline . Developmental deadlines influence behavior not only after they have been passed but also before that point is reached. Individuals approaching a deadline anticipate a steep decline in goal opportunities and feel an ever more urgent need to invest effort to attain the goal before time runs out.

The situation changes radically once the deadline has been passed. After the deadline has been passed without success, further goal engagement becomes dysfunctional. In fact, a radical shift from goal engagement to disengagement is the most adaptive response to deadline-related decline in opportunities. This shift to disengagement is analogous to a lion chasing its prey; at first the lion goes full speed (urgent goal engagement), but when the prey turns out to be too fast and the gap between them widens, the lion will not gradually slow down but rather stop in his tracks and turn around. It is active disengagement in terms of withdrawal of effort and breaking of commitment that achieves this rapid and radical shift from goal engagement to disengagement ( Wrosch, Scheier, Miller, Schulz, & Carver, 2003 ). In addition, human agents hold mental representations of the self and the capacity of one’s own agency, both of which can be compromised by experiences of failure, loss of control, or giving up a goal. Therefore, self-protective strategies of control (e.g., avoiding self-blame by attributing failure to external factors, comparing with less fortunate others) need to be activated to minimize the long-term damage that failure could have on motivational resources (e.g., self-esteem and hope for success in future actions).

Empirical Evidence for Major Propositions of the Motivational Theory of Life-Span Development

The conceptual framework of the Motivational Theory of Life-Span Development as outlined above comprises a set of 15 specific propositions about adaptive developmental regulation that can be investigated empirically. These propositions address and can be grouped into four topics: (a) the adaptiveness of primary control; (b) life-span trajectories of primary and secondary control; (c) optimization of goal choice and appropriate use of control strategies; and (d) action phases of goal choice, goal engagement, goal disengagement, and new goal engagement (also referred to as “reengagement”). For each topic, specific propositions are stated and the relevant evidence is summarized. Each individual study considered is briefly described in Table 2 , which also uses the structure of 15 propositions grouped into four topics.

Empirical Evidence Regarding Theoretical Propositions of Motivational Theory of Life Span Development

Note . MIDUS = Midlife Development in the United States; ADHD = attention-deficit/hyperactivity disorder.

Adaptiveness of Primary Control

The life-span theory of control proposes that the motivational system is set up to maximize primary control across life domains and lifetime ( J. Heckhausen & Schulz, 1993 , 1995 , 1999b ; Schulz & Heckhausen, 1996 , 1997 ). This proposition comprises two hypotheses regarding (a) behavioral preference and (b) objective and subjective benefits of primary control that help maintain the preference.

(1) Preference for primary control striving is universal: Proposition 1 (see Table 2 )

This proposition is well supported in humans as young as neonates ( DeCasper & Carstens, 1981 ; Papousek, 1967 ; J. S. Watson, 1972 ; J. S. Watson & Ramey, 1972 ; White, 1959 ) and for animals of various species ( White, 1959 ). Infants could learn head movements that were associated with external events, such as acoustic signals and milk reinforcement, and even when fully satiated, they were found to continue head movements and greeted the occurrence of the milk bottle with pleasure ( Papousek, 1967 ). Chimpanzees favored objects that could be moved or emitted sounds or light ( Welker, 1956 ); monkeys persisted for hours in trying to solve mechanical puzzles such as complicated door latches ( Harlow, 1953 ); and both children and rats preferred rewards that they produced by their own behavior to the same object that was noncontingent to their behavior ( Singh, 1970 ).

(2) Primary control striving has benefits

The proposed adaptiveness of primary control implies that primary control striving has benefits for the individual both objectively and subjectively (Proposition 2; see Table 2 ). Primary control has benefits in many everyday situations when goals are readily attainable and controllability is high, as well as in more critical situations when the individual’s control capacity is challenged. Several of the studies reported in Table 2 addressed older adults’ control efforts when dealing with functional constraints resulting from health problems ( Wrosch & Schulz, 2008 ; Wrosch et al., 2002 ; Wrosch, Schulz, Miller, Lupien, & Dunne, 2007 ) and particularly from disability such as visual impairment ( Wahl, Becker, & Burmedi, 2004 ) and multiple sclerosis ( Pakenham, 1999 ). The beneficial effects of primary control directed at improving or maintaining health and/or functional capacities is shown for a broad array of outcomes ranging from reductions in depressive symptomatology (e.g., Pakenham, 1999 ; Wrosch et al., 2002 ; Wrosch, Miller, et al., 2007 ) to improved patterns of diurnal cortisol secretion ( Wrosch, Miller, et al., 2007 ), positive affect ( Wahl et al., 2004 ), and chronic and functional health problems ( Fiksenbaum, Greenglass, & Eaton, 2006 ; Wrosch & Schulz, 2008 ) and lower mortality risk ( Gitlin, Hauck, Winter, Dennis, & Schulz, 2006 ). Particularly compelling is a study on primary control enhancing interventions with older adults prone to fall ( Gitlin, Winter, et al., 2006 ) showing major benefits in terms of greatly reduced difficulties with everyday activities and fear of falling and improved self efficacy. Another line of research shows the benefits of primary control striving in the transition to adulthood: Primary control striving benefits both objective outcomes in terms of earning coveted vocational training positions (Haase, Heckhausen, & Köller, 2008) and subjective transition outcomes, such as positive affect (Haase, Heckhausen, & Köller, 2008).

The life span viewed as an action field for the individual involves major changes in the capacity to exert primary control that are based on fundamental biological and social changes in available resources (e.g., strength, vitality, income, social status, social roles). To be effective agents in their own development, individuals need to be aware of these changes of control capacity across the life span (Proposition 3) and adjust their control striving accordingly (Proposition 4).

(3) Adults expect to lose primary control capacity with increasing age

Adults at various ages expect increasing developmental losses and decreasing gains in psychological functioning across adulthood and particularly in advanced older age ( J. Heckhausen, Dixon, & Baltes, 1989 ). These gains and losses at older ages are expected to be less controllable ( J. Heckhausen & Baltes, 1991 ), and older adults perceive developmental change ( Lang & Heckhausen, 2001 ) and life regrets ( Wrosch & Heckhausen, 2002 ) to be less controllable than do young adults. In a large survey of adults ranging widely in age, perceived personal mastery and perceived constraints to control were separately assessed and showed a stable sense of personal mastery across age groups but increasing constraints to control in older adults ( Lachman & Firth, 2004 ).

(4) Primary control striving is stable and secondary control striving increases across adulthood

Findings regarding age differences in primary control striving are mixed, reflecting stable ( J. Heckhausen, 1997 ), increasing ( Wrosch, Heckhausen, & Lachman, 2000 ), and decreasing ( Brandtstädter & Renner, 1990 ) age-related trajectories. General measures of primary control striving appear to reflect either decreases ( Brandtstädter & Renner, 1990 ), stability ( J. Heckhausen, 1997 ), or increases across age ( Wrosch et al., 2000 ). These mixed results may reflect different measurement strategies, with decreases found for an assessment of tenaciousness in goal striving that comprises both positively and negatively worded items ( Brandtstädter & Renner, 1990 ) and increases found for self-reports of persistence in goal striving ( Wrosch et al., 2000 ). The scale developed based on our theory and addressing self-report of behavior involved in persistent goal engagement (e.g., increasing effort when facing obstacles) yielded stable life-course trajectories. Another complicating factor is that aspirations may also vary with age. For example, in one study, downward adjustment of aspirations was coupled with increased persistence ( Wrosch et al., 2000 ). Thus, people may downwardly adjust their goals, which may facilitate more vigorous striving for those goals. Future research should investigate under which conditions in life people perceive their primary control strivings to be more or less persistent and resilient to challenges, as well as how primary control pursuit and goal adjustments work together.

Regarding secondary control striving, the available evidence consistently shows increases with age ( Brandtstädter & Renner, 1990 ; J. Heckhausen, 1997 ; Wrosch, Bauer, & Scheier, 2005 ; Wrosch & Heckhausen, 1999 , 2002 ; Wrosch et al., 2000 ; Wrosch, Scheier, Miller, et al., 2003 ). Older adults report using more goal disengagement, more downward goal adjustment, and more reinterpretations of events and failures that allow the individual to protect the self and its motivational resources.

Optimization of Goal Choice and Use of Control Strategies

The model of optimization in primary and secondary control ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ) proposes a set of three heuristics that individuals should use for optimizing their choice of goals in such a way as to maximize primary control capacity across the life span. First, the chosen goal, and accordingly goal engagement and disengagement, should reflect congruence with opportunities for control; second, the goal choice should consider consequences for other goal pursuits; and third, the choice should help to maintain diversity of goal pursuits ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ; Schulz & Heckhausen, 1996 ). Empirical evidence pertaining to these propositions is unevenly distributed. The two propositions about opportunity congruence of goal engagement and goal disengagement have been widely studied, whereas the other heuristics have received little attention in the empirical literature.

(5) Optimization heuristics have effects on outcomes via their regulatory role for using primary and secondary control strategies

Preliminary evidence from two studies (Haase, Heckhausen, & Wrosch, 2008; J. Heckhausen, Schulz, & Wrosch, 1998 ) indicates that the optimization heuristics of age appropriateness, considering consequences for other goal pursuits, and maintaining goal diversity influence subjective well-being as a function of their effect on specific control strategies involved in goal engagement and goal disengagement. This suggests that the optimization heuristics activated control strategies, which in turn affected outcomes. More research using fine-grained longitudinal studies is needed to examine how optimization strategies forecast adaptive control striving and consequent outcomes.

(6) People choose to engage with a goal when the opportunities for goal attainment are favorable

There is an abundance of studies that address whether individuals choose goals that are congruent with the control opportunities he or she encounters at a given age or life circumstance. A first group of studies examines age-graded differences in choice of developmental or life goals. These studies show that individuals select goals in accordance with age-related changes in control potential across the life span and between different domains of life ( Cross & Markus, 1991 ; Ebner, Freund, & Baltes, 2006 ; J. Heckhausen, 1997 ; J. Heckhausen, et al., 2001 ; Nurmi, 1992 ; Rothermund & Brandtstädter, 2003 ; Salmela-Aro, Nurmi, Saisto, & Halmesmäki, 2001 ; Sheldon & Kasser, 2001 ; Wrosch & Heckhausen, 1999 ). In general, goals aiming at developmental gains are more common among younger adults, whereas older adults increasingly report goals directed at preventing losses ( Ebner et al., 2006 ; J. Heckhausen, 1997 ; Ogilvie, Rose, & Heppen, 2001 ). Investigations of the domain-specificity of goal choices in different adult age groups has well established that midlife adults avoid career-related goals when major gains in this domain are no longer attainable ( Cross & Markus, 1991 ; Heckhausen, 1997 ; Nurmi, 1992 ) and focus on health-related goals when losses in this domain have become an urgent threat to their control capacity ( Cross & Markus, 1991 ; Heckhausen, 1997 ; Rothermund & Brandtstädter, 2003 ). Moreover, individuals at various ages have been shown to closely calibrate their goal choices and aspirations to their current control potential in specific domains, as, for example, when German high-school graduates apply to vocational training positions ( J. Heckhausen & Tomasik, 2002 ; Nagy, Kõller, & Heckhausen, 2005 ), American high-school seniors express more certainty about educational and vocational goals than about family-related and material goals ( Chang, Chen, Greenberger, Dooley, & Heckhausen, 2006 ), and older adults adjust their selective and compensatory primary control engagement to their physical health and advanced old age (Haynes, Heckhausen, Chipperfield, Newall, & Perry, in press; Menec, Chipperfield, & Perry, 1999 ; Rothermund & Brandtstädter, 2003 ; Wahl, Schilling, & Becker, 2007 ).

Finally, several studies have demonstrated that individuals not only choose goals that match their control capacity but also that such choices have benefits for the individual. A case in point is the engagement with educational goals in the post-high-school transition in the United States ( J. Heckhausen & Chang, in press ), which leads to superior developmental outcomes, both in terms of subjective well-being and educational attainments. Convergent evidence comes from a study of middle-tier school (i.e., Realschule ) graduates in Germany striving for vocational training positions (Haase, Heckhausen, & Köller, 2008; Nagy et al., 2005 ). Finally, older adults who are actively engaged in dealing with ongoing and reversible health problems experience fewer health declines and fewer depressive symptoms ( Wrosch & Schulz, 2008 ; Wrosch et al., 2002 ; Wrosch, Schulz, et al., 2007 ). Similarly, Gitlin, Hauck, Dennis, and Schulz (2007) found among African Americans that primary control striving for maintaining everyday activities helps to protect those who struggle with severe functional difficulties from developing depression. The same research group also showed that physical and occupational therapy interventions enhancing the primary control of older adults with functional difficulties significantly improved the participants’ chances of survival ( Gitlin, Hauck, Winter, et al., 2006 ; Gitlin, Winter, et al., 2006 ).

(7) Goal disengagement: People choose to disengage from a goal when the opportunities for goal attainment are unfavorable

Several studies show that individuals disengage from goals that are no longer attainable because of losses in control capacity related to aging, age-related societal opportunities, or illness and disability. Specifically, at older ages during the life span, adults disengage from gain-oriented goals and focus on loss-avoidance goals ( Ebner et al., 2006 ; Heckhausen, 1997 ). Moreover, older adults avoid certain domains of life that have earlier age peaks in opportunity (e.g., child-bearing, partnership, work, finances) when they report personal developmental goals ( Heckhausen, 1997 ; Heckhausen et al., 2001 ; Wrosch & Heckhausen, 1999 ) or ascribe less importance to keeping up with young adult levels of performance ( Rothermund & Brandtstädter, 2003 ). With spinal cord injuries, adults at various ages have been found to degrade the importance of goals in domains that are seriously compromised by the disability (e.g., bearing children, having a career; Weitzenkamp et al., 2000 ). When suffering serious illness or disability ( Menec et al., 1999 ) and at very advanced ages ( Rothermund & Brandtstaädter, 2003 ), older adults even disengage from efforts to avoid further losses in health and everyday functioning.

An abundance of studies shows that such disengagements from unattainable goals benefit the well-being and/or mental and physical health of the individual. Individuals who experience a loss of control due to unfortunate life circumstance, aging, or illness and disability can buffer the negative effects of this loss on subjective well-being, mental health, self-esteem, and perceived personal control by disengaging from relevant goals. When experiencing loss of control (e.g., decreased fertility at midlife, college major is too difficult), individuals at various ages can maintain their subjective well-being (e.g., less burn-out, higher perceived control, less depression) by disengaging from the goals that are rendered unattainable ( Brandtstädter & Rothermund, 1994 ; Carver, La Voie, Kuhl, & Ganellen, 1988 ; de Rijk, Le Blance, Schaufeli, & de Jonge, 1998 ; J. Heckhausen, et al., 2001 ; Thompson et al., 2006 ; Wallace & Bergeman, 1997 ; Wrosch et al., 2005 ; Wrosch & Heckhausen, 1999 ; Wrosch, Miller, Scheier, & Brun de Pontet, 2007 ; Wrosch, Scheier, Miller, et al., 2003 ). For instance, late-midlife adults who disengage from important goals, such as bearing a child or finding a romantic partner, during a time of life associated with steep declines in opportunities for these goals benefit in their subjective well-being and mental health compared with individuals who do not disengage from these goals ( Heckhausen et al., 2001 ; Wrosch & Heckhausen, 1999 ). Analogously, for experiences of control loss that are due to chronic illness and/or disability (e.g., macular degeneration, multiple sclerosis, HIV), individuals who disengage from goals that have become futile because of the disability or illness can protect themselves from mental health problems (e.g., satisfaction with life, mood, depression; Boerner, 2004 ; Evers, Kraaimaat, van Lankveld, Jongen, & al., 2001 ; Rothermund & Brandtstaädter, 2003 ; Thompson, Nanni, & Levine, 1994 ) and even promote better health outcomes ( Evers et al., 2001 ). There is even evidence that a dispositional ability to disengage from unattainable goals can benefit biological functioning and physical health (e.g., cortisol secretion, systemic inflammation, symptoms of illness; Miller & Wrosch, 2007 ; Wrosch, Miller, et al., 2007 ).

(8) When choosing a goal, the beneficial and detrimental consequence for other goals are taken into account

Some studies have started addressing this optimization heuristic. For example, compared with younger adults, older adults chose goals more frequently that facilitate attaining other goals ( Riediger et al., 2005 ). For young and older adults, intergoal facilitation is beneficial for goal engagement, and between-goal interference is detrimental to well-being. In addition, experimental research suggests that individuals activate overriding higher order goals automatically if they are being tempted to engage in goals that are incongruent with their higher order goals ( Fishbach, Friedman, & Kruglanski, 2003 ). Consistent with these findings, other research has demonstrated that an activation of important life goals is associated with an inhibition of alternative life goals, particularly among individuals who are highly committed to their goals ( Shah, Friedman, & Kruglanski, 2002 ). This line of research suggests that individuals possess implicit dispositions toward avoiding short-term temptations in favor of long-term goals, which predict adaptive behavioral responses aimed at achieving self-relevant long-term goals ( Fishbach & Shah, 2006 ).

(9) When choosing a goal, people try to maintain activity in diverse areas of life

This optimization heuristic has so far been largely neglected in empirical research. However, Kumashiro, Rusbult, and Finkel (2008) reported evidence in support of this proposition by showing that individuals seek equilibrium between goals in different key areas of life, namely personal and relational concerns. Their research demonstrated that after having been overly dedicated to one life domain, individuals reduce their motivation to make further progress in that domain and instead pursue goals in alternative domains that have been neglected.

Action Phases of Goal Choice, Goal Engagement, Goal Disengagement, and New Goal Engagement

The action-phase model of developmental regulation specifies how cycles of goal engagement and disengagement sequentially unfold over the life course and in coordination with waxing and waning opportunities to attain important life goals. The model makes specific predictions about discrete shifts from goal choice to goal engagement (Proposition 10) and from goal engagement to goal disengagement (Proposition 12), about the use of primary and secondary control strategies in action phases of goal engagement (Proposition 11) and disengagement (Proposition 13), about the facilitative role of alternative goal pursuits for goal disengagement (Proposition 14), and about functionally adapted mindsets for each phase (Proposition 15). Empirical evidence for these more recent developments of our theory is still scarce, particularly with regard to the shifts between action phases.

(10) When people make a goal choice, their mode of functioning shifts to goal engagement

This phenomenon has so far not been addressed by empirical research in the area of life-span development. However, in motivational research, shifts from choosing (deliberation) to acting (implementation) have been demonstrated extensively within the theoretical framework of the Rubicon model ( Achtziger & Gollwitzer, 2008 ; Beckmann & Gollwitzer, 1987 ; Gollwitzer, 1990 ; H. Heckhausen & Gollwitzer, 1987 ).

(11) Secondary control strategies enhance the effectiveness of primary control strategies during goal engagement

The three studies addressing this issue have suggested that secondary control strategies can help maximize primary control striving. It is interesting to note that this beneficial effect of secondary control strategies on primary control striving can play out in diverse scenarios. First, secondary control strategies can help to turn a success experience into a motivational resource for primary control striving ( Hall, Perry, Ruthig, Hladkyj, & Chipperfield, 2006 ). Second, selective secondary control strategies can buffer the negative effects of major stressful life events on goal engagement ( Poulin & Heckhausen, 2007 ). Finally, even compensatory secondary control strategies can support primary control striving, as in the case of Parkinson’s disease patients who controlled their emotional response to the illness and adjusted their self-concept, which, according to McQuillen, Licht, and Licht’s (2003) findings enabled them to successfully work on keeping illness-related restrictions to their activities at a minimum.

(12) When people find a certain goal pursuit futile or too costly, they shift to goal disengagement

This phenomenon has so far rarely been addressed by empirical research. A notable exception is a study by Babb et al. (in press) that used hypothetical vignettes about peer-related challenges that became increasingly uncontrollable. Older children responded with switching to adjustment, whereas younger children and children with attention-deficit/hyperactivity disorder tended to stick to primary control in spite of its apparent futility.

(13) Self-protective and goal-disengaging compensatory secondary strategies are combined during goal disengagement

Several studies have showed that compensatory secondary control strategies that involve self-protective cognitions have beneficial effects on objective and subjective outcomes. Downward social comparisons and causal attributions avoiding self-blame protected older adults from regret-related despair ( Bauer, Wrosch, & Jobin, 2008 ). Downward social comparison in older adults with low personal-control perceptions were associated with fewer hospitalizations and lower mortality ( Bailis, Chipperfield, & Perry, 2005 ). Attributions that avoid self-blame for outcomes perceived as uncontrollable ( Tykocinski & Steinberg, 2005 ) were found to predict better well-being ( Mendola, Tennen, Affleck, McCann, & Fitzgerald, 1990 ). One study experimentally enhanced self-protective secondary control by instructing subjects to compare themselves with others who are worse off or attribute negative outcomes to causes outside the self. These interventions were effective in reducing regret-related despair, thereby avoiding adverse consequences for physical health ( Wrosch, Bauer, Miller, & Lupien, 2007 ).

(14) Goal disengagement is easier when an alternate goal can be pursued

Under conditions of control loss or severe constraints to the individual’s control, individuals should disengage from a futile goal. Disengagement will be facilitated if an alternative goal is attainable that could profit from the resources freed up by disengagement (Proposition 3; see Table 2 ). This proposition reflects the idea that goal disengagement is particularly adaptive if it frees up resources for alternative primary control pursuits. Experimental and field studies with young ( Aspinwall & Richter, 1999 ; Wrosch, Scheier, Miller, et al., 2003 ) and older adults ( Duke, Leventhal, Brownlee, & Leventhal, 2002 ; Wrosch, Scheier, Miller, et al., 2003 ) indicated that the availability of alternative or substitute goals facilitates disengagement from unattainable or uncontrollable goals. In addition, there is evidence that the selection and pursuit of new goals has beneficial affective and health effects, particularly among people who are engaged in the pursuit of unattainable goals ( Wrosch, Scheier, Carver, & Schulz, 2003 ; Wrosch, Miller, et al., 2007 ). Such beneficial effects of alternative goals extend to the experience of older adults’ life regrets: Among older adults, having multiple goals for the future was associated with reduced levels of regret intensity, whereas for younger adults, having many goals was related to intensified regret experiences ( Wrosch et al., 2005 ).

(15) Information processing is biased to support the function of either goal engagement or goal disengagement

Several studies have tested the hypothesis that information processing is functionally adapted to an action phase of goal engagement versus goal disengagement. Three of these studies addressed goal striving and control behavior before and after passing a developmental deadline and supported the predictions that during phases of goal engagement, information processing (specifically visual fixation and incidental memory) favoring goal pursuit is dominant ( Light & Isaacowitz, 2006 ; Wrosch & Heckhausen, 1999 ) and that during phases of goal disengagement, information favoring goal pursuit is inhibited ( Light & Isaacowitz, 2006 ) and less well recalled ( J. Heckhausen, et al., 2001 ). Related research was conducted in the theoretical context of the dual-process model of self-development ( Brandtstädter, 2006 ; Brandtstädter & Rothermund, 2002 ). Under conditions of uncontrollability, individuals adhere to an accommodative mindset that renders positive, self-protective, and goal-irrelevant information more salient ( Rothermund, 2006 ). In contrast, under conditions of controllable threats, sensitivity to danger ( Brandtstädter, Voss, & Rothermund, 2004 ) and pain signals ( Rothermund, Brandtstädter, Meiniger, & Anton, 2002 ) are enhanced. In addition, research in the general area (i.e., outside of developmental psychology) of self-regulation and mental control shows that goal commitment is associated with cognitive inhibitory processes that protect goal pursuit from distracting influences from alternative goals ( Kuhl, 1985 ), which is beneficial for goal pursuit and attainment ( Kuhl & Weiß, 1994 ; Shah et al., 2002 ).

The empirical evidence strongly supports the ubiquitous preference for primary control when it is available and the benefits from primary control striving. Regarding the theoretically predicted trajectories of primary and secondary control, the evidence also shows that adults at various ages perceive life-span changes in primary control capacity; more specifically, they expect declines with increasing age and particularly in advanced older age. Empirical evidence regarding primary control striving is somewhat complicated by the fact that although control striving remains active, individuals adjust their goals to changing opportunities. In other words, individuals attempt to make the most of their control capacity at any given time in life. Regarding secondary control striving and particularly goal disengagement and self-protection, evidence consistently shows increased use at older ages.

Optimization of goal choice was supported by findings showing that individuals engage in goal pursuit when opportunities are favorable and disengage from goals when opportunities are unfavorable. However, few studies addressed the proposition that goal choice would be informed by considering consequences for other goal pursuits, and only one study to date has shown that individuals attempt to maintain diversity in their goal pursuits. Additional research is needed on how optimization heuristics influence the activation of control strategies.

Finally, regarding the propositions based on the action-phase model of developmental regulation, empirical research has begun to address questions regarding the effective use of control strategies for goal engagement and for goal disengagement, action-phase-specific mindsets, and the processes involved in the shift from one phase to the other (i.e., from goal choice to goal engagement and from goal engagement to goal disengagement). Evidence to date supports many of the propositions of the action-phase model regarding the role of motivational and volitional processes guiding control behavior and during goal engagement and disengagement cycles.

Productive Areas for Research

Although many of the major propositions of our Motivational Theory of Life-Span Development are now supported by empirical research, there remain several additional unresolved questions that should be addressed in future research. We present suggestions for exploring the rich array of research ideas encompassed by our theory.

Optimization, Goal Selection, and Reselection

The regulatory metastrategies involved in optimized goal choice have only rarely been addressed in the empirical literature ( Haase, Heckhausen, & Wrosch, 2009 ). We proposed that major heuristics for making adaptive choices of goals are goal– opportunity matching, management of consequences (or trade-offs) between goals, and goal diversity ( J. Heckhausen, 1999 ; J. Heckhausen & Schulz, 1993 ). These principles of adaptive goal choice are to some extent part of the societal regulation of the life course, in that constraints and incentives are available at the appropriate times. That way, not everyone has to consciously apply the heuristics to regulating his or her developmental future. However, sometimes these three heuristics (i.e., age-appropriate goal–opportunity matching, management of consequences, goal diversity) lead to conflicting goal choices. Under which conditions can we expect individuals to ignore or counteract a given heuristic of optimization? For example, when is it adaptive for an individual to choose goals that are not well supported by opportunity systems (e.g., off-time goals)? Moreover, when can individuals afford to focus on only one goal, counteracting the heuristic of diversity in goal choice? According to our theory, such narrow and exclusive investment can be successful only if the individual has unusually abundant resources for the chosen goal, high individual talent, and a supportive social context ( J. Heckhausen, 1999 ; Schulz & Heckhausen, 1996 ). It would be fascinating to retrace the life histories of goal selection and pursuit in highly specialized experts who strive for world-class levels of performance. A related question concerns when and how highly specialized individuals abandon their primary goal pursuits.

Another related set of questions concerns negative trade-offs between areas of goal investment. Being successful in pursuing specific goals can be seductive and lead an individual to invest too much energy, time, and effort in a limited set of goals to the exclusion of others. How do individuals gauge the costs of lost opportunities? Is the dominant pattern of goal engagement over time one of capitalizing on success and following a canalized path into specialized goal investment? Or alternatively, do people follow the logic of diminished returns, thus investing in formerly neglected goal domains as soon as a certain level of accomplishment is achieved in their primary domain ( Kumashiro et al., 2008 ; Lindenberg, 1996 )?

Research should also address the response to repeated failures in the pursuit of a cherished goal. When do individuals give up on a failure-ridden goal domain and pursue alternative but closely related goals? In this context, the issue of fundamental goal areas becomes critical, within which substitutions are possible. Different research traditions in motivational psychology ( Deci & Ryan, 2000 ; J. Heckhausen & Heckhausen, 2008 ; Ryan & Deci, 2000 ; Skinner & Wellborn, 1994 ) and life-course sociology ( Lindenberg, 1996 ; Steverink & Lindenberg, 2006 ) identify slightly different but converging sets of basic needs or motives that guide behavior. Striving for control and mastery, for positive and meaningful relations with other people, and for influencing others or at least not being dominated by others (autonomy) appear to be commonly accepted as fundamental needs or (implicit) motives for human productivity and well-being ( J. Heckhausen & Heckhausen, 2008 ).

Finally, shifts between goals, particularly between goals from different content domains, are very challenging. Future research could ask which goal commitments require discrete and intentional shifts in commitment or engagement and which involve gradual reorganizations and reevaluations of preferences. It may be difficult to intentionally downgrade the importance of a self-relevant goal and reengage with a different goal, because this might require some kind of self-deception ( Brandtstädter, 2000 ). Perhaps shifts that involve goals central to identity and therefore require an orchestrated investment of resources are more amenable to intentional reengagement.

Development of Optimization and Control Processes in Childhood and Adolescence

Coping and self-regulatory processes are subject to developmental growth during childhood and adolescence ( Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001 ; Compas & Worsham, 1991 ; J. Heckhausen & Heckhausen, 2008 ; J. Heckhausen & Schulz, 1995 ; Rudolph, Dennig, & Weisz, 1995 ). Research in this area has been infused with several theoretical models on how to slice the phenomena of coping and self-regulation into categories of behavior distinguished by their degree of intentionality and engagement versus disengagement ( Compas et al., 2001 ), whether they address the basic human need of competence, relatedness, or autonomy ( Skinner & Wellborn, 1994 ), whether they are directed at the situation, one’s own behavior or one’s emotion ( Eisenberg, Fabes, & Guthrie, 1997 ), or whether they involve primary control or secondary control processes ( J. Heckhausen & Heckhausen, 2008 ; Rudolph et al., 1995 ).

Recent research shows that overall, children and adolescents who are goal engaged and use problem-focused coping attain better psychological adjustment ( Compas et al., 2001 ). In contrast, disengagement and emotion-focused coping is associated with poorer psychological adjustment. However, a few studies suggest that engagement and problem-focused coping is maladaptive with uncontrollable stressors (e.g., parental conflict, sexual abuse) and that disengagement under such circumstances is associated with better adjustment ( Compas, 1987 ; Forsythe & Compas, 1987 ; Rudolph et al., 1995 ). Converging evidence comes from studies on children’s coping when undergoing medical procedures or experiencing illnesses that they could not control ( Rudolph et al., 1995 ; Saile & Huelsebusch, 2006 ; Thurber & Weisz, 1997b ; Weisz, McCabe, & Dennig, 1994 ) or being stuck at summer camp when suffering from homesickness ( Thurber & Weisz, 1997a , 1997b ). Together, these studies suggest that with increasing age during mid childhood and adolescence, youth become increasingly competent in deciding when primary control striving is useful and when it is futile. Moreover, when primary control potential is low (e.g., when you are homesick at an overnight camp), older adolescents are more skilled than younger youths in the use self-protective secondary control strategies.

Future research should investigate how developmental advances in cognitive capacities and emotional self-regulation enable children and adolescents to identify the control potential for a given goal and to activate the relevant control strategies of goal engagement or disengagement accordingly. For example, do the mental processes involved in optimization (goal choice) require anticipation of positive and negative consequences, the representation of counterfactual scenarios (what if I do this, what if I do that?), and thus a cognitive maturity built on formal operations ( Band, 1990 )? Moreover, is the development of control strategies universal, or can we expect significant individual differences? One study of children’s coping with chronic headaches suggests that 10-year-olds already show individual differences in choosing control strategies to match the controllability of challenges ( Saile & Huelsebusch, 2006 ). Those children who failed to adjust their control behavior to the actual degree of controllability in everyday challenges (e.g., studying for an exam; being nonathletic and never chosen for a team) also used maladaptive strategies of coping with their headaches.

Ideally, one would conduct longitudinal studies to track the unfolding of general developmental progress in self-regulation and control strivings. Such longitudinal studies could also investigate the developmental origins and trajectories of individual differences in control-related behavior. For example, individuals may develop patterns of primary control striving that reflect very high or even excessive persistence when facing insurmountable obstacles, whereas others are more amenable to disengage. Similarly, unusually low thresholds for goal disengagement can also develop as a consequence of developmentally inappropriate parental demands on children’s performance ( J. Heckhausen & Heckhausen, 2008 ). A good example for longitudinal research in this area is the study by Wrosch and Miller (2009) , which showed that among adolescents, the capacity for goal disengagement is enhanced after phases of depressive symptoms and that this very capacity to disengage seems to act as protection against later depressive symptomatology.

The secondary control strategies that are directed at either enhancing volitional commitment or compensating for failure and protecting motivational resources pose particular cognitive challenges, because they require that the individual takes a metastance toward his or her own motivational and emotional state of mind and generates means to influence it in ways that maximize motivational resources. Examples are self-protective causal attributions, avoidance of self-blame, self-enhancing social comparison, and devaluing an unattainable goal (“sour grapes”). The level of cognitive sophistication required for such strategies makes them elusive in childhood and “defers” their elaboration to adolescence ( J. Heckhausen & Heckhausen, 2008 ; J. Heckhausen & Schulz, 1995 ). Do they all develop in parallel, or do they supersede each other? Do cultures or families differ with regard to which strategies they prefer (e.g., devalue unattainable goal) and which they shun (downward social comparison)?

Finally, children have to learn to orchestrate primary and secondary control strategies so that a switch from goal choice to goal engagement is made most efficiently; similarly, they must learn to switch from goal engagement to goal disengagement, which requires concerted efforts to deactivate ongoing primary control striving and counteract motivational commitments, as well as handling threats to self-esteem and hopefulness. In sum, the development of optimization and control strategies opens a fascinating field of research that is far from exhausted. In particular, the development of secondary control strategies and the management of interphase transitions between goal engagement and disengagement are ripe for future investigation.

Social Relationships and Developmental Agency

In recent years, researchers have begun to investigate primary and secondary control behavior in the context of social relationships ( Smith et al., 2000 ). Perceived interpersonal control was found to be a key determinant of satisfaction with relationship quality in mother–daughter relationships ( Martini, Grusec, & Bernardini, 2001 ; Smith et al., 2000 ). Specifically tailored control strategies for dealing with interpersonal conflict in later life were found to be effective in securing retreating lines of defense and ranged from protecting relationship harmony to (merely) protecting one’s emotional balance ( Sorkin & Rook, 2004 ).

Social relations can themselves become important instruments for individual agents trying to regulate their own development, particularly when the existing societal context does not prescribe or support a chosen goal. For example, if a youth decides to forgo college and pursue a self-designed career as a rock musician, societal institutions typically do not facilitate this engagement. In this situation, goal pursuit can be facilitated if the individual is able to rely on a social network of like-minded others who can provide support as well as model means and ends.

An individual’s network of social relationships is probably one of the most flexible, dynamic, and at the same time robust contexts that shapes development ( Lang & Heckhausen, 2005 , 2006 ). Some social relationships are defined by the social context (e.g., coworkers, classmates), whereas others are the result of more or less intentional selection of social relationships (e.g., marital partner; Lang, 2001 ). Broader and longer-term adjustments of social networks in accordance with individuals’ changing motives have been reported in several studies, consistent with propositions of socioeomotional selectivity theory ( Carstensen et al., 1999 ). With increasing age, older adults selectively maintain close emotional relationships and discontinue weaker ties that would have more instrumental value ( Lang & Carstensen, 1994 ). Proactive molding of one’s social network by selectively keeping social ties to some select people and dropping ties to others may well be most effective for bringing about transactional influences between the individual agent and his or her social context ( J. Heckhausen & Heckhausen, 2008 ). By selecting social partners who are also committed to one’s preferred goals and by influencing one’s social relationships accordingly, one can set oneself up for a successful trajectory. Alternatively, selecting social partners who adhere to conflicting goals can have detrimental consequences. One exemplar study investigated the career goals and social relationships of Finnish high-school graduates during the transition into work life ( Jokisaari & Nurmi, 2005 ). Social ties with peers of higher socioeconomic status were associated with full-time employment 1 year after graduating from school. Also, social relations that were seen as a hindrance to one’s goals were associated with lower quality jobs 1 year after high school.

We have just begun to explore the ways in which proactive shaping of one’s social relationships influences future developmental ecologies and, thus, a sustained developmental path of the individual ( Lang & Heckhausen, 2005 , 2006 ). These efforts suggest that future research on the motivational processes involved in social relationships will greatly contribute to our understanding of successful development.

Cultural Differences in Reliance on Secondary Control

There has been a longstanding debate about culture-related differences in control behavior ( Azuma, 1984 ; Gould, 1999 ; J. Heckhausen & Schulz, 1999b ; Schulz & Heckhausen, 1999 ; Weisz, Rothbaum, & Blackburn, 1984 ). In particular, the proposition regarding the functional primacy of primary control as a universal characteristic of human (and beyond that, vertebrate) behavior, was rejected by some researchers in the field of culture-comparative psychology ( Morling & Evered, 2006 ). However, the first cross-cultural comparison using a Turkish translation of the OPS scales indicated similar patterns of endorsement for control strategies between Turkish and European adults at various age levels ( Ucanok, 2002 ). More recently, a study using a Chinese translation of the OPS scales with large samples of mainland Chinese students and young adults showed that among mainland Chinese students, primary control striving for academic goals was strongly endorsed and did not drop significantly, even after experiencing a major failure or setback at the university entrance exam ( Wong, Li, & Shen, 2006 ). Accepting failure did not appear to appeal to these Chinese youths. On the contrary, compensatory secondary control strategies and, in particular, goal disengagement were used to a very small extent compared with samples from Western industrial countries studied in other research.

The argument about the cultural relativity of primary control striving boils down to the proposition that individuals from interdependent cultures are more oriented toward others in their community or immediate social group when choosing goals ( Markus & Kitayama, 2003 ; Morling & Evered, 2006 ). This does not conflict with the Motivational Theory of Life-Span Development, which makes no assumptions about the individualistic versus collectivistic generation of goals for primary control. The issue of how goals are selected is one of optimization. A promising question for further research is whether interdependent cultures use additional, more community-oriented heuristics to select goals.

A related issue is whether in a certain culture some specific threats to primary control are viewed as accessible to primary control, as opposed to secondary control in some other cultures. In a study comparing preferred control strategies in Thai and American children, Thai children were found to prefer secondary control when adult authority figures were involved or when being separated from a friend ( McCarty et al., 1999 ). American children, by contrast, favored secondary control in case of physical injury. Thus, there is not simply a main effect of culture on preferred control strategy, but culture and stressor characteristics interact to determine preferences for primary or secondary control.

Our theory does propose cultural differences in the way goals are pursued and disengaged from ( Schulz & Heckhausen, 1999 ), particularly with regard to secondary control strategies, both selective and compensatory, that involve the self. With a lesser focus on independent and self-centered aspects of agency ( Markus & Kitayama, 2003 ), East Asian and other cultural groups around the globe may be less dependent on using secondary control strategies for keeping self-esteem and self-concept at high levels. Evidence supporting this idea comes from a study showing less use of self-protective secondary control strategies among Japanese and East Asian Canadians compared with European Canadians ( Tweed, White, & Lehman, 2004 ) and from Wong et al.’s (2006) study of mainland Chinese people showing resilience in primary control striving, even after major setbacks and little use of self-protective strategies.

We need to learn more about the dynamics of different primary and secondary control strategies in different cultural contexts and how different degrees of interdependent versus independent socialization affect usage of specific control strategies ( Ashman, Shiomura, & Levy, 2006 ; Cheng, 2000 ). For example, goal disengagement in interdependent cultures may require that others who are involved in goal pursuit are persuaded of the necessity to disengage, just as it requires self-protection for individuals socialized in independent cultures.

Individual Agency, Social Change, and Migration

The modern world with its rapid changes, increased interdependence of national economies, easy access to international travel, and stark contrasts between different societies’ control potential brings about new challenges and opportunities for individual agency. For basic research, these globalization-related societal developments afford opportunities to study the interface of individual and society in a dynamic adjustment process. There are two major sets of questions resulting from processes of increased international interdependence and exchange. First, what are the effects of social change on individual agency and control striving within specific national states and their societies? Second, how does the control potential in different societies affect individuals’ decisions to leave a given societal setting and seek a more favorable one? Regarding the first question, international life-course sociological research programs have identified the consequences of globalization processes as rendering life-course planning and particularly career-planning more difficult, because long-term career paths are becoming de-standardized, 2 less predictable, and thus long-term consequences of individual decisions have become less transparent ( Blossfeld et al., 2007 ; Buchholz et al., in press ). However, the degree to which these changes in globalization affect different parts of the population in different countries depends on the subgroup (e.g., young adults, women, preretirement employees) and the welfare policies of a given society. In general, youths and young adults who have not yet established their status in the labor market are more severely affected by the loss in predictability of life courses ( Blossfeld, Mills, Klijzing, & Kurz, 2005 ). Women who have interrupted their careers for family care (e.g., caring for a child or ill parent) also experience major discontinuity and uncertainty ( Blossfeld & Hofmeister, 2006 ). In contrast to these groups, men with established vocational careers in midlife are least affected ( Blossfeld, Mills, & Bernardi, 2006 ). For people close to or in retirement, the impact of the globalized economy has led to early retirement plans and unemployment imposed by employers and instigated by general economic crisis, thus reducing primary control capacity of the individual. The severity of these consequences for older employees depends on the retirement provisions of the particular welfare state ( Blossfeld, Buchholz, & Hofäcker, 2006 ). In fact, for all subgroups of society, the severity and specifics of the globalization effects are filtered by the specific characteristics of the national labor market, the educational and retirement system, and the role of the family. Thus, in some countries (e.g., Scandinavian countries) the negative effects of increased uncertainty in the labor market are buffered, and in others (e.g., Great Britain, the United States) the impact is direct and mostly unmitigated by state-run welfare systems or family networks ( Hofäcker, Buchholz, & Blossfeld, in press ).

What are the consequences of globalization for adaptive control striving for individual agents? If career paths become more unstable, there may be more opportunities for upward mobility of individuals and therewith an increase in primary control capacity. However, a situation with less societal structuring of opportunities also means that individuals have to rely more on their personal and social capital, and that is unevenly distributed across the social strata ( J. Heckhausen, in press ). Higher educational attainment in particular should play an even greater role in a more thoroughly globalized economy in determining the potential for attaining or maintaining high social status. Disadvantaged groups with low personal and social capital, such as youths, older adults, and women with interrupted careers, are more vulnerable to becoming marginalized and relegated to precarious forms of employment ( Bynner & Parsons, 2002 ).

It is difficult to predict how social class affects self-regulatory processes in the context of globalization-related social change. On the one hand, under circumstances of social marginalization and particularly low levels of individual resources (e.g., long-term unemployment), it is essential that an individual identifies those goals that are attainable and aggressively pursues them. On the other hand, it can be argued that a minimum of social resources is needed to pursue any developmental goal and that severe resource constraints essentially relegate the individual to pursue short-term survival goals. Among individuals with sufficient resources to regulate their developmental trajectories, one would predict that a keen ability to analyze opportunities to optimize goal choices is particularly important during periods of social change. Under conditions of social change, it is essential to master optimization of goal choice in terms of recognizing negative shifts in opportunities and adjusting goal engagement and disengagement accordingly. This may become even more important for individuals with relatively few resources, such as those at the lower end of the social ladder ( Tomasik, Silbereisen, & Heckhausen, in press ).

Regarding the second set of questions, individuals can decide to leave their current social and geographical setting and migrate to a country with different opportunities for individual agency and upward mobility. As discussed above, even in a globalized world economy, countries and their societal systems differ with regard to the degree of primary control they offer to members of different social groups. Recent research on subjective well-being across the globe has revealed that the degree of free individual choice is a dominant factor in determining the degree of perceived happiness in a country. Specifically, results from representative national surveys carried out between 1981 and 2007 in 52 countries show that happiness is associated with the perception of increased free choice in a given country, and that, in turn, is closely linked to positive economic development, democratization, and increasing social tolerance in a given country ( Inglehart, Foa, Peterson, & Welzel, 2008 ). It would be interesting to investigate whether streams of migration follow country differentials in the extent to which an individual can exert free choices and has control over the short- and long-term outcomes of his or her actions, including the influence on his or her own development and life course.

Evidence-Based Interventions

Our theory is well suited to serve as a conceptual foundation for intervention. The challenges of optimally adapting one’s control behavior to the changes in control opportunities across the life span are tremendous, and it is not surprising that many individuals fall short of optimized control behavior at least some of the time. Thus, intervention programs have the potential of helping individuals optimize control behaviors to the specific challenges posed by a particular stressor.

Specifically, guidance and training may be particularly needed with regard to the following aspects of control striving and developmental regulation: optimized goal choice, orchestrated goal engagement (with selective secondary control), willingness to accept and request help from others (compensatory primary control), disengagement from unattainable goals, and compensatory secondary control strategies directed at protecting motivational resources (e.g., self-protective causal attribution).

Research has begun to develop and test intervention strategies for some of these control-related challenges. For example, Weisz, Southam-Gerow, Gordis, and Connor-Smith (2003) developed a primary and secondary control enhancement training to treat mild to moderate child depression and found that children undergoing treatment were significantly improved to the point of being in the normal range of depressive symptoms both immediately after the treatment and 9 months later.

Regarding youths with academic problems, work is underway to investigate the effectiveness of interventions addressing goal setting and the use of primary and secondary control strategies. Educational psychologists working with control-theoretical models have identified particularly vulnerable college students who have little flexibility to adjust goals and use compensatory secondary control strategies but hold high ambitions and strong primary control strivings ( Hall et al., 2006 ). Such students profit significantly, in terms of their motivation for schoolwork and their grades, from training aimed at promoting compensatory secondary control, combined with attributional retraining (i.e., attribute failure to insufficient effort instead of lacking ability; Hall, Perry, Chipperfield, Clifton, & Haynes, 2006 ). Moreover, to combat failure and dropping out among college students, this group of researchers has developed and successfully used intervention techniques that focus on retraining the causal attributions for failure such that effort investment is fostered ( Perry, Hechter, Menec, & Weinberg, 1993 ). Ongoing longitudinal work expands this approach to interventions targeting unrealistic goal setting in college students by instructing them on how to disengage from unattainable academic ambitions and reengage with academic goals that match the students’ abilities and interests ( J. Heckhausen & Hall, 2006 ).

An important field for control-related interventions is the management of disability and caregiving among older adults. Gitlin et al. have developed multidisciplinary interventions (e.g., occupational therapy, physiotherapy) to prevent falls among frail elderly that are individually tailored to the abilities and circumstances of a given individual ( Gitlin, Hauck, et al., 2006 ; Gitlin, Winter, et al., 2006 ). These interventions work by mobilizing and improving the remaining physical strengths (i.e., selective primary control) and by instructing the older person to use technical aids and the assistance of others (i.e., compensatory primary control). Many caregiving interventions are based on enhancing caregivers’ ability to exert primary control over stressors, such as patient disruptive behaviors, by teaching them rudimentary behavior-modification skills, as well as enhancing their primary and secondary control by coaching them when to seek help from others or accept the fact that some stressors, such as the suffering of the patient, are fundamentally uncontrollable. Because of the multidimensional nature of caregiving challenges, the most effective intervention programs typically enhance both primary and secondary control in multiple domains ( Belle et al., 2006 ). To date, the primary emphasis in intervention studies with caregivers has been on enhancing primary control, with relatively little attention being paid to strategies that involve teaching caregivers which goals are unattainable and giving them the means for disengaging from those goals without feeling guilty. The relatively modest effects reported in the literature may in part be due to the overemphasis on primary control; significant additional benefit may be achieved by focusing as well on training secondary control strategies.

Summary and Conclusion

The Motivational Theory of Life-Span Development focuses on the impressive adaptive capacity of individuals to optimize development across major changes in the life course. Conceptual and empirical work in the past 15 years has shown that this adaptive capacity relies on self-regulation of motivational processes. The challenges individuals face as they develop from infants to adolescents, to adults, and into older age are challenges of selecting, adapting, and pursuing developmental and personal goals to reflect changing life-course opportunities. These motivational self-regulatory skills involve anticipating emergent opportunities for goal pursuit, activating behavioral and motivational strategies of goal engagement, disengaging from goals that have become futile and/or too costly, and replacing them with more feasible and timely goals.

A life-span developmental theory should address the following general challenges and questions: (a) Criteria of adaptive development should be assessed in ways that facilitate interindividual comparison, prevent distortion by subjective biases, and build on cross-cultural consensus about what constitutes a successful life; (b) investigate how the individual as an active agent in development selects and pursues goals and disengages from them; (c) examine the relation between life-course variations in opportunities and individuals’ engagement and disengagement with developmental and personal goals; (d) study the heuristics that help the individual to select appropriate goals to invest in, and to compensate for failures, setbacks, and losses when they occur.

Our Motivational Theory of Life-Span Development addresses each of these four challenges and enabled us to derive 15 specific and empirically testable propositions about motivation and control-related behavior and cognition that can be grouped into four major topics: (a) adaptiveness of primary control; (b) life-span trajectories of primary and secondary control; (c) optimization of goal choice and accordant use of control strategies; and (d) action phases of goal choice, goal engagement, goal disengagement, and goal reengagement.

Although many of the major propositions of our Motivational Theory of Life-Span Development are now supported by empirical research, there remain several additional questions that should be addressed in future research. Among these research challenges are the following: How do individuals get from one goal cycle to the next (e.g., substitute goal, alternative domain), and what role do optimization heuristics play in this regard? How do goal selection and control processes develop in childhood and adolescence? What is the role of control processes in social relationships and interpersonal interactions? How can individual agents support their primary control pursuits by selecting and shaping their social networks? Are there differences across different cultures in the use of heuristics of goal choice and the employment of secondary control strategies, particularly as they pertain to self-reinforcement and self-protection? What are the effects of social change on individual agency and control striving within specific countries and societies? What is the role of control potential in individuals’ decisions to migrate from one country and society to the other? What are effective intervention programs that combine training in the use of primary and secondary control strategies among populations such as depressed children, struggling college students, and overburdened caregivers for older adults?

The life-span theory of control originally proposed in 1995 has developed and elaborated a comprehensive Motivational Theory of Life-Span Development that comprises a set of specific testable propositions. This conceptual framework has guided much empirical research during the last 15 years, and many of its propositions have received substantial empirical support. However, some propositions remain to be tested, and an abundance of related research questions await empirical inquiry.


Preparation of this article was in part supported by National Institute of Nursing Research Grants NR08272 and NR09573, National Institute on Aging Grants AG15321 and AG026010, National Institute of Mental Health Grant MH071944, National Center on Minority Health and Health Disparities Grant MD000207, National Heart, Lung and Blood Institute Grants HL076852 and HL076858, National Science Foundation Grant EEEC-0540856 to Richard Schulz, a grant from the Canadian Institutes of Health Research, and a grant from the Social Sciences and Humanities Research Council of Canada to Carsten Wrosch.

1 Brandtstädter and colleagues, in their dual-process model, proposed a convergent distinction between assimilative and accommodative mindsets involving automatic modes of information processing that are functionally adapted to the assimilative orientation to goal pursuit versus the accommodative orientation to goal adjustment and disengagement ( Brandtstädter & Rothermund, 2002 ; Brandtstädter, et al., 1999 ; Rothermund, 1998 ).

2 Brueckner and Mayer (2005) , however, argued that a strongly standardized life course was the exception rather than the rule in modern societies anyway and that destandardization of life-course patterns is overstated by many life-course sociologists and actually mostly restricted to the sequencing of family relative to education and career events.

Contributor Information

Jutta Heckhausen, Department of Psychology and Social Behavior, University of California, Irvine.

Carsten Wrosch, Department of Psychology and Centre for Research in Human Development, Concordia University, Montreal, Québec, Canada.

Richard Schulz, Department of Psychiatry and University Center for Social and Urban Research, University of Pittsburgh.

Please note that Internet Explorer version 8.x is not supported as of January 1, 2016. Please refer to this support page for more information.


Acta Psychologica

Spelling out some unaddressed conceptual and methodological challenges in empirical lifespan research.

The importance of taking a lifespan approach to describe and understand human development has long been acknowledged (e.g., Baltes, 1987). Nevertheless, theoretical or empirical research that actually encompasses the entire lifespan, that is, from early childhood to old age, is rare. This is not surprising given the challenges such an approach entails. Many of these challenges (e.g., establishing measurement invariance between age groups) have been addressed in the previous literature, but others have not yet been sufficiently considered. The main purpose of this article is to present several examples of such largely unaddressed conceptual and methodological challenges and reflect upon possible ways to address them. We discuss the usefulness of a lifespan approach and the generalization of the challenges to other research comparing different groups, such as gender, culture, or species.

Cited by (0)

Alexandra M. Freund and Moritz M. Daum share the senior authorship.

People also looked at

Study protocol article, development and health of adults formerly placed in infant care institutions – study protocol of the lifestories project.

empirical research article on lifespan development

A growing volume of research from global data demonstrates that institutional care under conditions of deprivation is profoundly damaging to children, particularly during the critical early years of development. However, how these individuals develop over a life course remains unclear. This study uses data from a survey on the health and development of 420 children mostly under the age of three, placed in 12 infant care institutions between 1958 and 1961 in Zurich, Switzerland. The children exhibited significant delays in cognitive, social, and motor development in the first years of life. Moreover, a follow-up of a subsample of 143 children about 10 years later revealed persistent difficulties, including depression, school related-problems, and stereotypies. Between 2019 and 2021, these formerly institutionalized study participants were located through the Swiss population registry and invited to participate once again in the research project. Now in their early sixties, they are studied for their health, further development, and life-course trajectories. A mixed-methods approach using questionnaires, neuropsychological assessments, and narrative biographical interviews was implemented by a multidisciplinary team. Combining prospective and retrospective data with standardized quantitative and biographical qualitative data allows a rich reconstruction of life histories. The availability of a community sample from the same geographic location, the 1954–1961 cohort of the Zurich Longitudinal Studies, described in detail in a paper in this issue ( Wehrle et al., 2020 ), enables comparison with an unaffected cohort. This article describes the study design and study participants in detail and discusses the potential and limitations of a comparison with a community sample. It outlines a set of challenges and solutions encountered in the process of a lifespan longitudinal study from early childhood into the cusp of old age with a potentially vulnerable sample and summarizes the lessons learned along the way.


In Switzerland, placing infants in institutions was not uncommon in the first half of the 20th century ( Ryffel, 2013 ). The main reasons for having a child placed in an institution were either being an unmarried or underaged mother or having a migrant background, foremost Italian migrant worker status (German: Gastarbeiter ) ( Meierhofer and Keller, 1974 ). Having a child as a young, unmarried mother was, from the point of view of the authorities and society, slovenly (liederlich) and was to be “disciplined” ( Ramsauer, 2000 ; Lengwiler and Praz, 2018 ; Unabhängige Expertenkommission Administrative Versorgungen, 2019 ). Migrant workers were subjected to serious prejudice and residence permit restrictions and were forced to work full time with long working hours to stay in Switzerland ( D’Amato, 2012 ; Joris, 2012 ). Generally, infants were placed into institutions at a very young age, before the age of two weeks, due to the lack of paid maternity leave ( Huber, 1995 ). At that time, the infant was seen as a simple reflex-driven being ( Meierhofer, 1958 ), and a belief was prevalent that “there will be no harm to infants if they are cared for by strangers” ( Meierhofer and Keller, 1974 ). Hard-earned success in reducing child mortality had made preventing the spread of germs a priority, so an institutional practice of “isolation” was the norm, involving as little physical contact as possible, feeding according to a rigid plan, and strict hygiene ( Ryffel, 2013 ). Care practices were generally characterized by strict routines that did not take into consideration individual needs and an intense wariness of spoiling children ( Gebhardt, 2009 ). This created the conditions of chronic deprivation found to be responsible for the profound negative effects on development described in more recent work ( Nelson et al., 2014 ; Berens and Nelson, 2015 ).

A growing volume of international research shows that children placed in institutional care as it is commonly implemented are typically deprived of a supportive, intensive, one-to-one relationship with a primary caregiver. Such conditions of deprivation are profoundly damaging to children, particularly during the critical early years of development ( Schore, 2001 ). Children who were placed into institutions shortly after birth and subjected to severe sensory, emotional, and sometimes physical neglect show a dramatic decrease in brain activity compared to children who were never institutionalized ( Center on the Developing Child, 2007 ). They are more likely to suffer from growth delay, frequent infections, and hearing and vision problems. Furthermore, motor development is often delayed and stereotypical behaviors such as body rocking and head banging occur ( Browne, 2009 ; Berens and Nelson, 2015 ; Sherr et al., 2017 ). Finally, a number of studies have found negative effects on cognitive and social development ( Schore, 2001 ; Johnson et al., 2006 ). Although no data is currently available on the effects of institutionalization across the entire lifespan, a number of large studies have demonstrated the potential long-term negative effects of other adverse events and circumstances during childhood, such as child abuse and neglect, household substance abuse, and mental illness, on development and health into adulthood ( Felitti et al., 1998 ; Werner, 2013 ). The impact of such events has also been shown to be especially severe during the critical developmental stages in early childhood, with possible permanent effects on the morphology of the brain ( Shonkoff and Phillips, 2000 ).

However, not all individuals experience negative health outcomes in relation to stress. In his seminal work, Antonovsky (1979) coined the term “salutogenesis” after observing “how people manage stress and stay well.” Similarly, findings from a 40-year longitudinal study by Werner (2013) showed that one third of all high-risk children exposed to adverse experiences early on displayed resilience that allowed them to develop into caring, competent, and confident adults. Particular protective factors and biographical events helped to balance out risk factors at critical periods in their development in a dynamic interaction of personal factors, environmental factors, and biographical developments. Further, evidence has started to emerge of what mitigates the negative impact of institutionalization, such as age at entry, stability of care, size of institution and children/staff ratio ( Berens and Nelson, 2015 ; Sherr et al., 2017 ). Care circumstances and practices with children in institutions can vary greatly, with some of the most severe conditions of deprivation studied as part of the Bucharest Early Intervention Study with children in Romania ( Nelson et al., 2014 ). This study was also able to show that when children are moved from institutional care into family based care, they have a chance of restoring brain development, highlighting the plasticity of the developing brain ( Nelson et al., 2014 ; Bick et al., 2015 ).

The potential impact of this early institutionalization across the lifespan is unknown. Furthermore, whereas data is available on the effects of exposure to adversity in early childhood into early and mid-adulthood, the effects of adverse experiences into late adulthood remain as yet virtually unexplored. With a unique combination of historical and newly collected 60-year long-term follow-up data, the overall aim of the LifeStories project (German: Lebensgeschichten ) is to examine the personal developmental trajectories of individuals that were affected by placement in institutions as infants in the late 1950s and early 1960s in Switzerland. The study builds on the pioneering work of Dr. Marie Meierhofer on the delayed development of infants placed in institutions for care at the end of the 1950s ( Meierhofer and Keller, 1974 ). By simultaneously using data from an unaffected comparison sample of children growing up in families, the 1954–1961 cohort of the Zurich Longitudinal Studies of the University Children’s Hospital Zurich (ZLS, N = 445) —one of the most significant data sets on child development globally ( Ulijaszek et al., 1998 )—the project will shed light on the impact of certain institutional care practices during infancy across the lifespan.

Methods and Analysis

This population-based study uses a combined prospective and retrospective, mixed-methods approach over a 60-year period to investigate how the lives of individuals who had been placed in an institution as infants in the late 1950s in Switzerland developed subsequently. The availability of data from the 1954–1961 cohort of the Zurich Longitudinal Studies enables comparison with an unaffected group of children growing up in families at the same time and in the same geographic region. The overall study design is depicted in Figure 1 . This article describes the cohort of children placed in institutions as infants. The ZLS are described in the paper in this issue ( Wehrle et al., 2020 ).


Figure 1. Study design.

The following sections describe the assessment at three time points of the cohort of infants placed in institutions.

Wave 1: 1958–1961

As a baseline, this project uses data from the study conducted by Dr. Meierhofer between 1958 and 1961. She conducted a survey on health and development in 12 infant and toddler care institutions in Zurich, Switzerland. Data collection took place over 16 months. During this period, around 630 children were placed in these 12 institutions. Figure 2 provides an example of an infant room in one of these institutions to illustrate the living conditions at that time. Children who fulfilled the following eligibility criteria were included in the study: children were at least three months old at the time of the examination, they were not older than six months at the first time of in-care-placement, their institutional stay was never interrupted for more than three months, and they did not have any diagnosed medical disorder. In addition, children were excluded if they suffered from any acute infectious disease at the time of the examination or responded with distress to the test situation. This resulted in a sample of N = 429. 1 .


Figure 2. Example of one of the rooms infants were kept in at one of the 12 institutions ( Meierhofer and Keller, 1974 ).

Two types of data were collected during Wave 1: person-centered data for each child and institution-centered data for each institution in which children were placed. Data was collected as a mixed-method assessment and included quantitative and qualitative data.

Person-centered data

For all children, demographic data, the psychosocial situation of the family, nature and frequency of current contact with the family, and information on children’s institutional stay (e.g., age at in-care placement) were recorded. Because in most instances it was not possible to obtain this information directly from the parents, Dr. Meierhofer used the administrative records available (e.g., guardianship records) and information provided by staff members of the care institutions. These data sources were also used to record children’s health status (e.g., previous diseases, birth and pregnancy history). Some of this information was recorded in standardized, self-developed study templates; other data was documented in narrative notes and underwent quantitative coding afterward. To assess children’s development, a standardized developmental test was used [ Échelle de développement , Brunet-Lézine for all children up to 2.7 years ( Brunet and Lézine, 1955 ), Terman-Merrill developmental test for children up to 4.6 years ( Terman and Merrill, 1937 ), and Schweizertest for children up to 7.6 years ( Biäsch and Fischer, 1969 )]. The individual test result was used to calculate an overall developmental quotient (DQ) for each child. Furthermore, height and weight were measured. In addition, children’s behavior and interaction with staff members, other children, and the examiner were assessed by observations during daily routines in the institutions and during the test situation. In general, the observational data was collected and documented in a qualitative manner, and some of the data was subsequently quantitatively coded. In addition, original photographs, slides, and film has been preserved.

Institution-centered data

Information on childcare circumstances and practices in the different institutions was collected, such as interaction time, child-staff-ratio, and the educational background of staff members. In addition to information provided by the heads of the institutions, observations were recorded by the researchers during daily routines (see Table 1 for an overview of the instruments used at Wave 1).


Table 1. Overview of the assessment instruments (Wave 1 and 2).

Today, most of the data from Wave 1 is stored in the Federal Archive in Aarau. The research team has been granted permission to access and analyze this historical data for scientific purposes related to the historic reappraisal of care practices before 1981, on condition that it is used in anonymized form, according to article 11c of the Swiss Federal Act on the Reappraisal of Compulsory Social Measures and Placements before 1981 (German: Bundesgesetz über die Aufarbeitung der fürsorgerischen Zwangsmassnahmen und Fremdplatzierungen vor 1981 ) and the Act on Public Information, Data Protection and Archives (German: Gesetz über die Information der Öffentlichkeit, den Datenschutz und das Archivwesen; IDAG ) and the corresponding by-law (VIDAG).

Unfortunately, even though additional data from Wave 1 was retrieved from various private archives as well as the archive of the Marie Meierhofer Children’s Institute, some of the data of this initial assessment remains missing. However, identifying data has been preserved for 98% of the original cohort: for 420 of the 429 children. Aggregated developmental data (e.g., DQ) is available for the majority of the children up to 3 years at the time of Wave 1 ( n = 322).

To make this data accessible, all documents were retrieved from the archives as scans or digital photographs. Because this data is only available in analog form (mostly hand-written, some of it typed up with manual typewriters), data had to be entered manually into electronic form. A couple of issues posed a challenge to systematic data entry: firstly, information for the individual children was not available in a systematically compiled form but was distributed over many different datasheets. Secondly, the information was not recorded consistently across all of the 12 institutions but was collected using many slightly different datasheets (see Figure 3 for an example of the available data). To deal with this and to reduce errors during data entry, standardized input templates were created for quantitative data ( Limesurvey GmgH, 2020 ). Data was then imported into R ( R Core Team, 2018 ) and merged for further processing and analysis. Qualitative data was typed up in Microsoft Word. Photographs, audio files, and film have been digitized for preservation.


Figure 3. Example of person-centered data (Wave 1).

In her analyses of the data at the time, Dr. Meierhofer found that the children placed in institutions had significant delays in cognitive, social, and motor development compared to the children growing up in families studied by the ZLS. Differences in children’s development within the group of institutionalized children could not be explained by families’ socio-economic status or the infants’ contact with their family members. However, differences in the quality of care (i.e., interaction time and child–staff ratio) between the 12 institutions included in the study accounted for the variance in children’s development. Meierhofer and Keller (1974) concluded that the developmental delays of the children were primarily caused by the poor conditions of care in the institutions.

Wave 2: 1971–1973

To assess the developmental progress and health status of the children, Dr. Meierhofer and her team conducted a follow-up study of a subsample of the children between 1971 and 1973, then aged between 13 and 15 years (Meierhofer and Hüttenmoser, unpublished).

Of the original cohort, Dr. Meierhofer only considered children born between 1957 and 1959, who were up to age three years at the time of Wave 1, for the follow-up study ( N = 354). The address for 82 of the eligible children was not found at the time, and so it was not possible to contact them. An additional 13 children were excluded as they had moved to other parts of Switzerland and the distance from these children was deemed too great for their inclusion in the study. For 64 children, parents or legal guardians actively declined consent for participation. Another 24 did not respond to the invitation to participate (presumed passive decline). Two of the children had died. An additional ten children were excluded for other reasons (not further specified). Of the remaining 159 children, 16 participated in a preliminary study between 1969 and 1971 ( Meyer-Schell, 1971 ) 2 and were then excluded from further study. The final sample of the full follow-up study therefore consisted of N = 143 children. The distribution of gender, nationality, and marital status of the mother within the group of children examined was, according to Meierhofer and Hüttenmoser (unpublished) approximately comparable to that of the overall sample. In the further course of the study, 17 of these children with suspected epilepsy or other neurological disorders were excluded from further analyses (Meierhofer and Hüttenmoser, unpublished).

Data collection during Wave 2 was carried out with a multi-method and multi-informant approach: quantitative and qualitative data were collected using administrative records, standardized tests and questionnaires, and interviews with parents or other primary caregivers, teachers, and the children themselves.

To obtain general information on the family background, children’s care histories (e.g., number of different placements, reasons for placement changes), and their academic career, the study team used administrative data available, for example, from guardianship records. If necessary, parents or other caregivers were asked to provide further information. Most of the information was qualitative in nature and documented using standardized templates developed by the study team 3 .

Standardized testing was conducted using a battery of neuropsychological tests and questionnaires: Children’s cognitive abilities were assessed with the WIP, a short version of the Hamburg Wechsler Intelligence test for children ( Dahl, 1968 , 1972 ). Their mental health was assessed using the Kinder-Angst-Test (KAT), ( Thurner and Tewes, 1969 ), a standardized questionnaire on children’s anxiety. Furthermore, two projective tests were conducted: the Rorschach-Test ( Bohm, 1965 ) to assess personality and the Foto-Hand-Test ( Belschner et al., 1971 ) to assess aggressive behavior. The test battery also included the Sohnaufsatz ( Ungricht, 1955 ), in which children had to write a short essay that was later analyzed for both content and graphology, and the Baumtest ( Koch, 1957 ), which was intended to provide information on the child’s character and affective development. In addition, children’s height and weight were measured. Furthermore, data on pubertal development and general health was collected through information provided by the parents or caregivers. In addition, the social status of the children among their classmates was determined with a sociogram ( Bastin, 1967 ): the children themselves and five of their classmates were asked to indicate with which children they would most like to spend time and with whom they would least like to spend time. The number of children’s mutual choices was evaluated.

In addition, semi-structured interviews were conducted with primary caregivers, children themselves, and their teachers. For the interviews with the parents, key themes were predefined, such as pregnancy and birth history, care histories, children’s development (motor skills, language, sleep, tidiness), social relationships, and education. Furthermore, interviews with the children contained questions about school, leisure activities, friendships, and their ideas about their future. The teacher survey focused on academic performance and social contact with classmates. If necessary, interviews with migrant workers were conducted in Italian and subsequently translated. The original audio-recordings and some of the transcripts have been preserved. The qualitative information collected during the interviews was scored according to Thalmann’s symptom-burden scale ( Thalmann, 1971 ) and used to assess children’s behavior. In addition, teachers and researchers recorded information on the children’s personality and learning and academic behavior (for example reliability, discipline, independence, pace of work) on a 5-point standardized scale ( Polaritätenprofil ). Table 1 provides an overview of all instruments used at Wave 2.

At the time of Wave 2, data from the ZLS was not ready for comparison. Dr. Meierhofer and her team therefore selected the instruments for the follow-up-study so that data of a normative sample and, if possible, even comparative data from Switzerland was available that allowed basic comparisons (Meierhofer and Hüttenmoser, unpublished).

In her analysis, Dr. Meierhofer found that at the time of Wave 2, children who were placed in institutions as infants showed increased depression, school related-problems (e.g., significantly higher grade retention rate than the comparison group), and stereotypies (Meierhofer and Hüttenmoser, unpublished). Due to a fierce nature/nurture debate with one of her colleagues at the time, these results were never published ( Wyss-Wanner, 2000 ) and exist only in an original research report in the archives of the Marie Meierhofer Children’s Institute (Meierhofer and Hüttenmoser, unpublished).

In contrast to the Wave 1 data, the raw Wave 2 data has been fully preserved. Part of the data is stored in paper form at the Federal Archive in Aarau. It has been digitized in the same way as the Wave 1 data by first retrieving the data from the archive as digital photographs and then entering the data manually into data entry masks created with LimeSurvey ( Limesurvey GmbH, 2020 ). The remaining data is available in paper form and on microfilm in the archive of the Marie Meierhofer Children’s Institute and has now also been digitized for further processing. Figure 4 provides an example of the available data. Just as for the Wave 1 data described above, data entry masks were created in LimeSurvey ( Limesurvey GmbH, 2020 ), and further data processing were carried out with R ( R Core Team, 2018 ).


Figure 4. Example of person-centered data (Wave 2).

Qualitative data will be typed up in Microsoft Word either from available audio files or from original transcripts.

Wave 3: 2019–2021

In a newly revived effort funded under the umbrella of the Swiss National Research Program 76, which is focused on scientific investigation into the compulsory social measures and placements before 1981, all individuals that took part in the study by Dr. Meierhofer at Wave 1 were once again located and contacted. These individuals were about 60 years when taking part in this third assessment of health and development (Wave 3).

Study preparations

Prior to Wave 3, two preparatory studies were conducted. They are separate, small studies but are briefly summarized here.

Exploratory study

Between 2012 and 2013, an exploratory study was implemented. Semi-structured narrative interviews were conducted with 16 individuals that had taken part in Waves 1 and 2. As part of the semi-structured interviews, participants talked about topics such as family, institution, foster or adoptive family, relationships, health, education, work life, hobbies, and aging ( Ryffel and Simoni, 2016 ). Furthermore, findings suggested that despite clear indications of resilience in some individuals, many generally struggled to find a sense of belonging and coherence in their lives. This seemed to be the case more often for individuals placed in institutions as a result of having been born to unmarried and/or underaged mothers than born to migrant workers. Children of migrant workers tended to reassured themselves that they were loved by their parents and that it was external circumstances that led to the placement. They remember it as a measure that “made sense” ( Ryffel and Simoni, 2016 ; Sand and Gruber, 2017 ).

The experience of this exploratory study informed the search strategy, contact procedures, ethical considerations, and research questions for Wave 3 (for details see corresponding sections below). It also reaffirmed the need for a comprehensive long-term follow-up assessment, as it revealed how diverse the life trajectories can be within this cohort. It also showed that individuals are willing to participate in a research study and talk about their biographical trajectories.

Participatory research preparation

To prepare for Wave 3, in 2019, we conducted focused interviews with four individuals that had been placed in institutions as children to elicit feedback on procedures, documents, and assessment instruments to be used for contacting the cohort. This was in response to the request of those affected by the compulsory social measures and placements before 1981 to be included in research related to the reappraisal and reconciliation process ( UEK Administrative Versorgungen, 2019 ; Unabhängige Expertenkommission Administrative Versorgungen, 2019 ). Interviewees’ feedback helped to make documents more understandable and identified wording that might cause insecurities or trigger negative reactions. They also indicated shortcomings in some of the questionnaire items that they thought were potentially misleading or not appropriate for the situation of the participants ( Lannen et al., 2020 ). In addition, they made significant contributions to how best to approach and work with the cohort. This participatory preparation showed that the inclusion of formerly institutionalized individuals in research is feasible and provides substantial benefits to the research quality on historic compulsory social measures and placements ( Lannen et al., 2020 ).


Eligibility criteria were defined separately for (a) locating individuals and (b) contacting them.

In order to be eligible for the search, individuals had to have taken part in Dr. Meierhofer’s study at Wave 1 and to have been assigned a study number for which identifying data was available. In addition, a minimum of three data points needed to be available for the search (full name, date of birth, and a location of residence from some point in their lives). A number of individuals were found to have moved abroad (25%, n = 107). If they were found to be residing in a country other than their country of origin, we were able to search them through the embassy or the Bureau of Foreign Affairs in charge. If a city they have moved to abroad is known, they can be searched for through the local population registry, if one exists. No search strategy exists for individuals who moved to their country of origin without any indication of location. In this case, individuals abroad are ineligible for the search.

A person who was not found and those deceased were obviously deemed ineligible for contact. Those that were found, were only eligible for contact, if there was no indication that they might have been adopted without their knowledge (happened early in life, was followed by a name change). This was based on the ethical concern of uncovering a potentially unknown adoption and causing distress to the individual. If there was an indication in the historical files that the individual knew about the adoption, or that the adoption happened within the family (e.g., by the new partner of the mother), individuals remained eligible for the study. In addition, individuals also became ineligible for contact if they had a data protection barrier with the municipalities (see below section “Locating individuals”). Finally, individuals who had actively declined study participation when contacted for the exploratory study ( Ryffel and Simoni, 2016 ) were also deemed ineligible.

Locating individuals

The search process took place between October 2018 and March 2021.

Due to the sensitive nature of the request, the process of locating individuals after several decades was set up in a way that minimized the risk of false identification of individuals. Therefore, even though very labor intensive, the search for each individual’s identity needed to be officially confirmed through the population registry. Luckily, this was possible, as in Switzerland, every individual is formally registered with the municipality where he or she resides. Individuals who relocate must give notice of departure with the old municipality and formally register with the new municipality, so individuals can be tracked through the system over time (municipal population registry). In addition, events (such as birth, marriage, divorce, or death) are registered in the civil population registry ( Zivilstandsregister ) in a person’s commune of origin ( Heimatort ) 4 .

In Switzerland, legislation differs slightly between the cantons 5 , but generally, municipalities will provide full name, address, and dates of arrival in and departure from the municipality from the population registry to private individuals and organizations acting in the public interest without requiring a reason for the request 6 . If a “credible argument” is made, the municipality that the person moved to and from, date of birth, gender, marital status, and commune of origin ( Heimatort ) can also be released 7 . Before the information is released, the municipality is to verify that the request is made in the public interest 8 ; if this is not case, the municipality is not allowed to release the information. Individuals that have actively instated a data protection barrier are excluded 9 . The municipalities also provide information on deceased individuals should the applicant make explicit an interest for the information 10 . Some municipalities requested more detailed information on the public interest of the request. In these cases, detailed information on the study and on data protection was provided. The study was described as a general survey on health and development and the fact that the individuals in question were placed in institutions as infants was not disclosed to protect their privacy.

A successful search for an individual through municipal population registries requires three data points: the full name, date of birth, and a municipality in which the individual has lived at some point. For a majority of individuals that took part in Wave 1, information on municipalities was available through the study archives. For some, we were able to gain access to municipalities of residence through the Zurich’s City and Federal Archive (archived municipal registration files before 1976, archived supplementary files of birth registries, guardianship records, infant care institution files etc.). The municipal authorities released the current address if the individuals still lived in that municipality or released the name of the municipality the person moved to. The address information request was then iterated through as many municipalities as necessary until the current address was found.

For some cases, instead of a municipality of residence, the person’s commune of origin ( Heimatort ) was available through the archives. A formal research request was submitted to the office of civil registry in Zurich and granted under Art. 60 of the Zivilstandsverordnung vom 28.04.2004 (ZStV; SR 21 1.112.2) to inquire whether a place of residence can be found in the files for individuals still alive or a date of death for those deceased. Upon the precedent of Zurich, civil registries in other locations provided access to the information. For these individuals, the last known address through civil registries was thus identified and then the above-mentioned process of locating individuals through municipalities repeated for these individuals. An additional research request granted access to the information on deceased individuals whether death was of natural or unnatural (accident, crime, or suicide) causes. This information was made available by the civil registries in the municipalities where individuals resided at the time of death.

For a few individuals, their municipality of residence was found through moneyhouse.ch, an online platform providing information drawing among others from the cantonal commercial registers (since full name and date of birth was available through this portal) and then verified with the municipality.

Overall, the procedure for contacting individuals was very resource intensive and is still ongoing. The search was reiterated through up to 10 municipalities until a person’s current address was identified. More than 300 municipalities were contacted, as well as more than 40 offices of civil registries in Switzerland alone. In total, more than 2000 emails were sent.

So far, 86% ( n = 268 of total 313) of individuals residing in Switzerland were found, out of those, 78% ( N = 208) were eligible for contact (not deceased or ineligible).

So far, 25% ( n = 107) of individuals were found to have moved abroad. If a person had moved to a country other than their home country, we placed an inquiry with the embassy or the Bureau of Foreign Affairs. If the person was registered by their country as living abroad, he or she could be located. If a person was not registered but the municipality the person moved to was known, rather than just the country, then analogous to the search in Switzerland, individuals were searched for through the population registry (if some form of population registry existed in that country). This search strategy via municipalities was applied for all individuals who moved abroad regardless of their nationality or home country. Using these strategies, we have identified 47%, n = 50 individuals eligible for search abroad. Out of those, we have been able to locate and contact 24% ( n = 12) so far.

Figure 5 shows the search path.


Figure 5. Search process to locate individuals.

Contact Procedures

The contact procedure for this study was built on a number of key principles: first, it takes a stepwise approach, so that those individuals that do not wish to learn about any study details can opt out swiftly (ethical protocol, see section “Ethics”). Second, contact procedures were also staggered to ensure sufficient resources were available within the research team for personal contact with the cohort and data collection. Third, contact procedures were designed for a broad range of life trajectories, building on Antonovsky’s salutogenetic approach ( Antonovsky, 1987 ). This is expressed in the wording of documents and behavior of the researchers. Fourth, even though some participants might not remember that they had participated in the study in the past, overall, the study invitation has been framed as an invitation to remain in the study. Finally, the procedure has been set up as an opt-out procedure: at each step, the next point of contact through the research team will be announced, and individuals have to actively let the research team know if they want to disengage from the process. Passive decline has been clearly operationalized as no more contact after a final reminder for participation by mail if a participant cannot be reached by phone after three attempts.

Furthermore, the Swiss Federal Act on the Reappraisal of Compulsory Social Measures and Placements before 1981 (German: Bundesgesetz über die Aufarbeitung der fürsorgerischen Zwangsmassnahmen und Fremdplatzierungen vor 1981 ) stipulates that individuals have the right to view their archived files. Accordingly, the study team has created a leaflet with detailed information on how individuals can access any archived information, either related to their institutional placement or the study.

As an initial form of contact, a short letter with basic information about the study and the individual’s name and year of birth found in the archives was sent to each prospective participant. It also included a request to let the researchers know by phone, email, or returning a slip (a stamped return envelope was included) if they did not want to participate. In that case, they were not contacted again. Two weeks later, anyone who had not actively declined was sent a second letter. This time, more detailed information was provided about the study, including a detailed leaflet that outlined its history, aims, and procedure. It reiterated that participation is voluntary and that they can withdraw at any time. It also outlined data protection measures. In addition, the leaflet specified the components of the study (questionnaire, neuropsychological assessment, biographical narrative interview; for details see section “Data collection”), including expected duration and reimbursement. Individuals received CHF 80. – each for filling in the questionnaire and participating in the neuropsychological assessment, and CHF 40. – for participating in the biographical narrative interview. In addition, they received reimbursement for travel expenses. Again, if anyone indicated that they did not want to participate further (active decline), they were not contacted again. Otherwise, senior staff members called all the individuals for whom we were able to obtain a phone number to invite them to participate in the study and go over the consent form initially over the phone. Notes of the conversation were recorded. Those individuals for whom phone numbers were not available were invited to provide one with a return slip. It was also possible to indicate preferences for communication (mail only, for example, for those who do not want to talk on the phone but would like to participate in the study).

The same researcher remained available throughout the study to ensure consistency and to build trust. If an individual agreed to participate, the questionnaire was sent together with the consent form, and appointments for neuropsychological assessment and interviews were arranged. Help with completing questionnaires was offered when necessary over the phone as a standardized interview. A reminder call was made the day before the appointment. If necessary (e.g., due to poor health), the researcher offered to travel to the participants’ home for data collection.

If the full questionnaire was not returned after 3–4 weeks, the researcher checked in with the participant either by phone or letter to remind the participant to fill in the questionnaire and to see if any assistance was needed. Those individuals who could not be reached by phone were sent a final reminder letter, including the short version of the questionnaire, a consent form, and a slip to indicate active decline or interest in further participation. Sending a final reminder letter with a questionnaire is based on the work of the Zürcher Längsschnittstudie Von der Schulzeit bis ins mittlere Erwachsenenalter ( Schallberger and Spiess Huldi, 2001 ), which led to a substantial number of additional questionnaires returned.

For those individuals that were contacted as part of the exploratory study ( Ryffel and Simoni, 2016 ), letters were tailored to the individuals’ responses.

The recruitment process is depicted in Figure 6 .


Figure 6. Recruitment process.

Data collection

Data collection started in September of 2019, is still ongoing, and is expected to last throughout Q1 of 2021.

A mixed-methods approach was chosen to maximize depth and breadth of data and to enable the accommodation of data collection to individuals’ possibilities and preferences where necessary. Specifically, three types of data were collected: questionnaire data, data from neuropsychological assessments, and data from biographical narrative interviews. The full battery of questionnaire data covered physical and mental health, social abilities and demographics, information on work and family life, retrospectively captured education and professional background, critical life events, and transitions. It included constructs such as sense of coherence, self-efficacy, and the ability to fulfill one’s basic psychological needs, which are all factors believed to be important moderators for adjustment and fulfilled, happy lives. All constructs and the corresponding operationalization and instruments used to assess them are listed in Table 2 . The full questionnaire battery took about 120 min to complete. A pilot data collection phase had indicated that the nature and length of questionnaire was appropriate for the target population. For those unable or unwilling to fill in the full questionnaire battery, a short version of the questionnaire was provided, covering physical and mental health outcomes and demographic information (an estimated 60 min to complete).


Table 2. Overview of the assessment instruments (Wave 3).

In addition, the researchers invited participants to come to the Marie Meierhofer Children’s Institute in Zurich for neuropsychological assessment (estimated duration: 120–150 min). A full list of assessment instruments is listed in Table 2 .

Questionnaire and test data were digitalized by entering it into digital questionnaires programmed with LimeSurvey ( Limesurvey GmbH, 2020 ). This procedure facilitated data entry and reduced the potential for errors. Inter-rater reliability checks will be performed for test scoring. In addition, data entry was double checked by a second researcher for both test data and questionnaire data. Data was exported from LimeSurvey using a comma-separated format (.csv) and cleaned with R ( R Core Team, 2018 ).

Preliminary power analyses indicated that for a t -test for independent samples with an alpha-error of 0.05 and a power of 0.80, at least 102 participants are required to detect medium effect sizes, while the detection of large effect sizes requires 42 participants.

Finally, participants were invited for biographical narrative interviews ( Rosenthal, 1995 ), focused on the individual narrative that participants present when asked to talk about their lives. These interviews were conducted without a predefined time limit, and the participants are asked to explore the topics and share the experiences that are relevant to them as they wished. The interviews were conducted at the location of the participants’ choice. Interviews were recorded with an audio recorder and later transcribed and anonymized ( Bohnsack et al., 2013 ). The audio files were then deleted to ensure data protection.

The time and nature of death of deceased participants was determined from the records of the municipal and civil registries.

Comparison With Zurich Longitudinal Studies

In the 1950s, Dr. Meierhofer was in close contact with Prof. Guido Fanconi, the medical director of the University Children’s Hospital who initiated the ZLS study in 1954. Dr. Meierhofer aligned the assessment instruments used to study the infants in the institutions with those used in the ZLS ( Wehrle et al., 2020 ). Dr. Meierhofer studied the children at two time points: when children were between birth and three years of age (Wave 1) ( Meierhofer and Keller, 1974 ), and a subsample of the children about 10 years later (Meierhofer and Hüttenmoser, unpublished). Children in the ZLS, were continuously assessed until age 18 ( Wehrle et al., 2020 ). In a parallel effort, both cohorts are currently being located and assessed again. Once again, the study instruments have been closely aligned, thus allowing a 60-year longitudinal design with a comparison group.

Adjustments Related to the Pandemic of COVID-19

About midway through data collection for Wave 3, the COVID-19 pandemic reached Switzerland, and measures to contain the spread of the virus were implemented. Neuropsychological assessments and interviews on site were suspended for three months between mid-March and mid-June 2020 and then recommenced with protection measures. Data collection for questionnaire data continued throughout, and now includes a questionnaire to assess the impact of the measures to contain the pandemic on a range of outcomes: work and finances, home and social life, daily routines, and mental and physical well-being. This additional questionnaire was also sent to participants who had already completed data collection. Furthermore, this same questionnaire was sent to participants of the ZLS (for details see this issue Wehrle et al., 2020 ). This will allow us to check for any possible bias in data collected before or after the onset of the pandemic as well as whether reactions differed between the two cohorts to the measures implemented by the government.

Data Analysis

Dr. Meierhofer and her team analyzed the historical data with the tools and the statistical techniques available at that time. For instance, tables and graphs were drawn by hand.

The historical data are now being re-analyzed with modern statistical methods to increase the reliability and validity of the historical results. Specifically, for data from Wave 1, we will seek to replicate Dr. Meierhofer’s finding that institutional placement and especially care conditions are significant predictors of the differences in children’s developmental outcomes. Our primary aim with the data of Wave 2 is to compare it with the data now available from the ZLS and to finally make the data available through publication.

Data will allow longitudinal analyses over two and three time points, both within group for children placed in institutions and between groups with children placed in institutions and children who grew up in families. The within-group analyses will examine the role of a set of potential predictors of inter-individual differences in growth in health and development. The between-group analyses will compare the two cohorts on the same outcomes.

We will apply structural equation modeling and multivariate regressions using a combination of variable- and person-centered statistical methods. These methods will allow us to account for measurement error (latent variable modeling), the multilevel structure of the data (multilevel modeling or correction of standard errors where multilevel modeling cannot be applied), and multiple covariates. A further important aspect of data analysis will be the implementation of multiple imputation methods to address missing data in general and selective drop-out in particular. A preliminary review of the raw data also indicates that historical data contains some errors that took place when transferring data between sheets or when rounding to the decimal. These errors will be corrected systematically.

For data from each wave, we will compute the deviation score of each participant of the cohort placed in institutions from the ZLS cohort mean. The resulting deviation score at Wave 3 will then be predicted by the deviation scores for Waves 1 and 2 by regression modeling. In addition, the influence of the individual variables on the longitudinal trajectory of health and development will be calculated using random-intercept autoregressive models ( Hamaker et al., 2015 ) and latent growth models ( Sticca and Perren, 2015 ). Multivariate associations among multiple constructs of interest will be examined using multivariate random-intercept cross-lagged models ( Hamaker et al., 2015 ) and parallel process models ( Sticca and Perren, 2015 ), depending on the research question at hand.

Qualitative data will be coded and descriptive analyses run and depicted in graphs where useful. Longer handwritten narratives and texts from Waves 1 and 2 will be subjected to content analysis ( Mayring, 2000 ). Selected historical qualitative materials and narrative interviews conducted at Wave 3 will also undergo reconstructive and sequential analysis according to Rosenthal’s ( Rosenthal, 2015 ) method. The aim of this abductive analysis method is to identify the latent content of the text, make generalizations, and come to conclusions via in-depth analyses of individual cases ( Rosenthal, 2015 ).

Qualitative and quantitative data will be triangulated and historical data contextualized in the historic discourse on education and upbringing, structural violence, welfare practices, discrimination of certain family models, and compatibility of employment and family.

A number of measures identified ethical issues and mitigated risks for all three waves.

An independent ethics expert reviewed both historical studies (Wave 1: 1958–1961 and Wave 2 1971–1973), drawing on primary historical data and reports and publications. The review concluded that although research ethics practices have developed over time, Dr. Meierhofer adhered to the key ethical principles that hold today, chiefly the principle of not harming the subjects. The review even concluded that the children that took part in the study may have benefited; they certainly received some extra interaction time and attention as part of the study, in contrast to the deprivation prevalent in the institutions at that time. Assessment took place through observation and conversation and was non-invasive. At Wave 1, consent was provided by the head of the care institutions and at Wave 2 by the children’s legal guardians. Without such consent, children were not included in the study. Data was anonymized for analysis and publication ( Brauer, 2019a ).

For Wave 3, a number of ethical issues were identified. These included, for example, the risk of disclosing a previously unknown or not remembered institutional placement, the risk of disclosure of an institutional placement to next-of-kin or other third party, distress caused by learning about previously unknown inclusion in a study, discussing potentially distressing events from the past as part of data collection, and distress caused by accessing archived materials.

A comprehensive ethical framework and ethics protocol were developed that detailed every step of interaction with the individuals of the cohort with the aim of mitigating potential risks. In addition to detailed informed consent, voluntary participation, the option of withdrawing consent at any time, and protection of personal data, measures included a step-wise approach to contacting individuals with increasing amounts of information on the study, an option to claim mistaken identity, easy opt-out procedures, contact only with senior researchers, regular, standardized screening for distress, psychological support available on site for study participants, and psychological supervision available to researchers.

Furthermore, a thorough consent procedure was implemented, including seeking consent to link new and historical data. Participation was voluntary, and consent could be withdrawn at any time. Data will only be analyzed and published in an anonymized form.

These ethical considerations and ways to mitigate distress were developed and reviewed in consultation with an external, independent ethicist ( Brauer, 2019b ). Furthermore, findings from the exploratory study outlined in section “Exploratory Study” ( Ryffel and Simoni, 2016 ) and the focus interviews conducted as part of the participatory study preparation ( Lannen et al., 2020 ) (see section “Participatory research preparation”) also shaped the ethical framework of the study.

Wave 3 was reviewed and approved by the Ethics Committee of the Faculty of Philosophy at the University of Zurich (Approval Number 19.4.7).

A number of longitudinal studies over several decades on human development have been conducted in different parts of the world (for an exemplary overview see this issue Wehrle et al., 2020 ) and emerging data from robust longitudinal studies of children placed in institutions is becoming available ( Nelson et al., 2014 ). However, the 60-year span of the LifeStories project presents the longest follow-up of children who spent their early years in an institution. Central to the feasibility of this study are a number of significant opportunities and challenges, with several lessons emerging from each of them.

Historical Data

Comprehensive long-term longitudinal studies over several decades are rare, because they require continuous project leadership across several generations of researchers and an institutional home for the data to guarantee data access and data preservation over time. Other challenges of long-term longitudinal studies include the advancement of science overall: concepts and methodologies may change fundamentally after the initiation of the study. For instance, instruments used several decades ago might no longer be in use, thus making direct comparisons across the life span difficult. Some of these challenges are described and discussed in detail in the paper in this issue ( Wehrle et al., 2020 ). For this study, even though it was dormant for over 40 years, access to data was possible through the continuity of institutional involvement and became accessible to the next generation of researchers at the Marie Meierhofer Children’s Institute. However, the fact that the data was not preserved in its entirety and had to be reassembled from a variety of documents—some organized by institution, some by subject, and some by outcome—posed a significant challenge. Using modern tools such as LimeSurvey templates and having another researcher enter data a second time mitigated the risk of data entry errors.

Locating Individuals

A second challenge was to locate individuals after many decades while reducing possible false identification. However, being able to locate individuals through population registry is promising and has been successfully implemented, primarily by studies in the Nordic countries, where a central population registries exist ( Kreicbergs et al., 2004 ; Lannen et al., 2008 , 2010 ; Surkan et al., 2008 ; Lysholm and Lindahl, 2019 ). In Switzerland, the process is not centrally organized, and tracking individuals through the system over time proved to be very time consuming. Nonetheless, it was possible to find 86% ( n = 268) of individuals as long as they still lived in Switzerland even decades after the previous contact. In fact, it proved possible to find individuals that Dr. Meierhofer was unable to locate at the time of Wave 2. It is plausible, but difficult to fully verify, that the reasons why individuals were not found in Switzerland included both death and a name change after an adoption.

Due to the substantial proportion of children of migrant workers ( Meierhofer and Keller, 1974 ) in the cohort, almost one quarter of the individuals moved abroad. Finding these individuals is central to addressing some of the key research questions of the study, as the development trajectory of this subgroup of individuals might differ significantly due to familial backgrounds and reasons for placements. Tracing individuals who had moved abroad proved more difficult, but so far, 24% ( n = 12) of eligible individuals living abroad were found.

Contacting and Recruiting Study Participants

In recent years, a growing number of studies have examined the experience of formerly institutionalized individuals in Switzerland ( Bombach et al., 2017 , 2018b ; Ammann et al., 2019 ; UEK Administrative Versorgungen, 2019 ). However, the primary approach of these studies was an opt-in approach in response to a call for contemporary witnesses to share their experiences. In contrast, because this was a pre-existing, predefined cohort, the study team approached a set cohort of individuals and invited them to continue their participation in the study after many decades. Although it is the availability of data from this cohort that provides the opportunity to conduct this longitudinal study, the study presents a number of challenges. Some individuals will be unaware of their former institutional placement, as they might not remember it due to their age at the time or due to repression of memories ( Freyd, 1996 ; Ryffel and Simoni, 2016 ). This also might be a period of their lives that was not discussed in the family because it was considered stigma or taboo ( Sack and Ebbinghaus, 2012 ; Ryffel and Simoni, 2016 ; Lengwiler, 2018 ). Hence, contacting them might disclose this information. A number of measures were put in place to mitigate some of the risks when contacting the individuals, including a comprehensive ethical framework and ethics protocol and psychological support available on site. Furthermore, all contact with the individuals was made solely by senior staff members. This proved essential when having to respond to diverse biographical themes triggered during contact procedures. For example, as a result of their past experiences with authorities, concerns about being purposely misled or judged by the researcher surfaced regularly ( Lannen et al., 2020 ). Working with senior researchers was also key to responding appropriately to any questions, concerns, or potential distress that arose during contact. Earning the trust of the participants was essential. This was achieved chiefly by consistency; the same researcher stayed in contact with the individual throughout the study. It was also achieved by the researchers’ efforts to be empathetic no matter what situations came up and to be “humanly available” to the participants ( Lannen et al., 2020 ). Trust was further built by the ability within the research team to deploy researchers with the same linguistic backgrounds as the study participants (Swiss-German, German, Italian, Spanish).

Another important element proved to be the participatory approach to study preparation that included individuals affected by institutionalization as children. Their input significantly shaped the narrative, the approach of the study, and the researchers’ skills ( Lannen et al., 2020 ). It also provided an opportunity to respond to the request of those affected to be included in research related to the historic reappraisal process of compulsory social measures and placements before 1981 in Switzerland ( UEK Administrative Versorgungen, 2019 ).

Finally, the project built on Antonovsky (1987) and Werner (2013) findings and a belief that the life trajectory of an individual is determined by many factors of internal and external nature. When asked to provide a recommendation on what would be helpful to formerly institutionalized children in overcoming their experience, Gahleitner, as part of her expert mandate in the round-table discussions related to institutional placement in Germany, stated: “A salutogenetic approach (…) allows insights that are usually lost when the focus lies on a search for deficits” 11 ( Gahleitner, 2009 ).

This was expressed in how letters to the participants were formulated and how researchers interacted with them. It was also expressed in the outcome measures included in data collection and an attempt to move away from a solely deficit-oriented narrative in relation to their starting conditions. While prepared for any distress expressed by the participants as a result of past events, the team was also able to acknowledge evidence of resilience and a sense of coherence ( Antonovsky, 1987 ). Overall, the approach attempts to overcome the risk of categorizing the individuals in the group of individuals that started unfavorably and developed poorly (those in infant care) compared to a community sample of individuals that grew up at home and therefore are considered to be the norm (ZLS). The study is embedded in a framework of respect and empathy for each individual’s life story. It is open to a multitude of life trajectories between and within each study arm without prejudging the outcomes. At the same time, the framework honors the sense of injustice felt by many of the individuals subjected to welfare practices implemented before 1981. It is an attempt to honor their experiences without passing judgment, and also seek the potential resilience and strength that may have grown from each individual’s unique experience.

Methodological Approach

A number of factors fundamentally shaped the methodological approach of the study. These included the feedback on instruments from individuals affected by institutional placement as children during study preparation. Moreover, the research team applied a multidisciplinary and multimethod approach. Leveraging backgrounds in developmental psychology, pediatrics, educational sciences, sociology, and anthropology allowed the team to use validated instruments for assessment of health as well as psychological constructs and combine them with, for example, reconstructive methods. Although this approach required constant discussions and feedback loops in order to negotiate common terminology and framing, it enabled a holistic view of the interplay between individual development and a particular societal and educational context. Furthermore, the expertise combined in quantitative and qualitative approaches enabled the collection of both standardized and narrative data and thus reflect the complexity of the human experience as completely as possible.

However, this study also faced a number of key methodological challenges:

First, the availability of a community sample of individuals that grew up in the same geographic location at the same time ( Wehrle et al., 2020 ) enables us to distinguish historical variables from biological, personal, and social ones and extract generalizable statements and recommendations. However, children placed in institutions in the late 1950s were exposed to multiple risks: both the societal, legal, and familial circumstances that put the children at risk of institutionalization in the first place and the deprivation the children experienced as part of the institutional placement ( Meierhofer and Keller, 1974 ). This is relevant, because other studies have found evidence of a dose–response effect in that the more adverse events a child is exposed to, the more profound and lasting the effect on the individual will be ( Felitti et al., 1998 ; Finkelhor et al., 2007 ; Huebner et al., 2016 ). As part of this study, it will be important to distinguish pre-existing risks as much as possible from the impact of institutionalization and possible compensatory effects arising from relationship conditions with parents or better institution quality by including variables in our models that account for possible additive and interaction effects. Fortunately for us, children were generally assigned to institutions independently of variables such as health, development, or potential covariates such as family background. Placement was decided by geographic location or space availability, hence mirroring a quasi-experimental design of children with similar distribution of backgrounds and preconditions in each institution and thus increasing the probability that differences between development outcomes in the different institutions are due to circumstances in the care facility.

Second, individuals in this cohort are organized by institution in 12 clusters. This is a challenge, as 12 clusters is at the very low end, if not below the required amount of clusters from a statistical point of view. The other challenge arises when comparing data with the ZLS, as this multilevel structure of data is only present in one cohort. In addition, some information only exists for the entire institution even though it can be assumed that conditions within the institutions varied between units within a single institution. Combined expertise in the project team, the project board and through collaboration with specific experts will enable us to address this methodological challenge through multilevel modeling or correction of standard errors where multilevel modeling cannot be applied.

Third, there are a number of challenges inherent to longitudinal studies that run for several research generations. These challenges are described in detail in the paper of this issue ( Wehrle et al., 2020 ), and include dealing with archived data and their sometimes spotty documentation, the predefined nature of the cohort or balancing continuity of measurements while keeping up with latest standards.

Fourth, the more time passes, the greater grows the chance of bias as a result of the possible impact of the events on mortality for those more severely affected, as described in seminal literature ( Felitti et al., 1998 ). Furthermore, the results of the preliminary study ( Ryffel and Simoni, 2016 ) also suggest that a self-selection bias might have operated toward participants with more adaptive long-term outcomes. In addition, women and the children of single mothers were over-represented compared to migrant families. Analyzing whether participation in the study is selective in relation to key variables from Wave 1 and 2 is essential.

Research Questions

Mastering these challenges will allow us to address a number of research questions.

Historical data:

(1) Circumstances of placements:

What information can be found on why infants were institutionalized as part of historic welfare practices, and how is this represented in the literature on relevant discourses from that time? What were the conditions and routines like within the institutions?

(2) Health and development of children

Can findings from the historical data be replicated? What additional findings related to the impact of infant institutionalization can be brought to light from the historical data? What additional and relevant insights can be found by comparing Wave 2 to the comparison group of the ZLS? How do familial and institutional contexts relate to children’s health outcomes and abilities?

What deficits identified in Wave 1 and Wave 2 can still be detected in individuals 60 years after they were institutionalized as infants?

Newly collected and longitudinal data:

How have the lives of individuals that spent their infancy in institutions evolved? What is their morbidity (physical and mental) and mortality? What are their cognitive, social, and motor abilities today? What are their educational paths and socio-economic conditions? What deficits identified in Waves 1 and 2 can still be detected in individuals 60 years after they were institutionalized as infants? What specific vulnerabilities and/or strengths have emerged during the life course? How have some of the care circumstances and practices related to institutional placement affected long-term outcomes? What are the key individual variables and life events that moderate the relationship between early institutionalization and long-term outcomes? How do individuals talk about their life trajectories? What themes are relevant to them and how do they make sense of what happened?

Outlook and Conclusion

Overall, the study will allow us to better understand the effects of being placed in infant institutions under conditions of deprivation, will improve our knowledge of possible development trajectories, and will contribute to the reconciliation process and historical reappraisal process of compulsory social measures and placements before 1981 in Switzerland. It will reveal important insights for children placed in institutions today and allow us to better understand how children with different starting conditions can be supported in developing healthily. Finally, the present study will provide a unique contribution to our understanding of the interplay between individual and environmental promotive and protective factors for development over decades in individuals at risk.

A number of opportunities exist when considering potential outlooks for the study. The individuals of the LifeStories project are now at the cusp of old age. Some evidence has recently emerged demonstrating that exposure to adverse events in childhood is linked to premature biological aging in adulthood ( Shalev et al., 2013 ; Puterman et al., 2016 ). However, this data was collected retrospectively. Longitudinal data is needed on how adverse events may affect biological aging.

Consent from all participants of Wave 3 was sought to contact them again in the future for additional prospective assessments. Thus, collecting biological and neuroimaging data will be important to further understanding the specific aging processes in this vulnerable cohort.

In addition, collecting information from additional perspectives will allow us to complement the efforts related to the historic reappraisal process of compulsory social measures and placements before 1981 in Switzerland. To better understand the circumstances that led to infant care placements at the time, the perspective of the parents of formerly institutionalized infants would be essential and is time sensitive due to their advanced age. This also applies to former employees of the institutions. In addition, interviews with children of those placed in institutions are relevant, as emerging evidence shows the effect such measures may have from a generational perspective ( Bombach et al., 2018a , b ; UEK Administrative Versorgungen, 2019 ).

Making the details of the study process available as part of a study protocol allows the scientific community to learn what works to successfully conduct longitudinal studies and shed light onto the secrets of long-term adaptation processes across the lifespan.


The results bear significant potential for scientific, practical, and social impact at national and international levels, and results will be disseminated accordingly.

Understanding the impact that measures related to compulsory social measures and placements before 1981 in Switzerland had on the children during infancy and childhood and the consequences 60 years later will contribute to the historical reappraisal and political reconciliation of such measures in Switzerland. Accessing historical records will enable a data-driven approach to understanding the causes, characteristics, and mechanisms that led to and surrounded infant care practices before 1981. It will allow society to make sense of how practices came about and enable it to uncover the zeitgeist and societal norms and values that framed the actions. The study will illuminate a piece of Swiss institutional history, provide a basis to reflect on today’s practice critically, and might sensitize the society to the possible existence of blind spots in today’s practices in Switzerland and around the world.

Due to its longitudinal design, the study is highly relevant and one of the few long-term follow-up studies of its kind. Through the existence of a comparison group from the same era ( Wehrle et al., 2020 ), historical variables can be distinguished from biological, personal, and social ones, and generalizable statements and recommendations can be extracted that are relevant for today. This will improve our understanding of the detection and handling of difficult early life conditions and our support of the development of resiliency over a life-course perspective.

It will also improve our understanding of the long-term consequences of deprivation caused by institutionalization of young children. The study will reveal factors mitigating institutional care and protective processes for favorable development trajectories as a result of resiliency processes. It will also provide information on how the professional and policy community can best support children and young people in care and their families across the globe.

To this end, scientific publications focused on the key research questions will be published in peer-reviewed journals related to medicine, psychology and educational and other social sciences, and results will be presented at international scientific conferences of these disciplines. In addition, a significant effort will be made to make results accessible to a non-scientific expert audience, society and the survivors of compulsory social measures and placements before 1981.

Evidence briefs will serve as a key tool for disseminating results to a non-scientific audience. A set of evidence briefs will be published in German with a local focus related to the reappraisal and rehabilitation of compulsory social measures and placements before 1981. Additional briefs that focus on the results of international relevance will be published in German and English and target professionals and key organizations involved in out-of-family placements and a multidisciplinary audience of professionals and policy-makers working with individuals at all stages of life. Further, we will work toward a book publication with some of the historical images and materials available. In addition, roundtable discussions will be held with relevant professionals. Results will be presented at local conferences for a non-scientific audience. Efforts will be made to develop a curriculum that includes our findings in university-level training (psychology, social works, educational sciences) and with children in primary and secondary school settings.

All publications will be made available free of charge to participants if desired.

Ethics Statement

The studies involving human participants were reviewed and approved by Ethics Committee of the Faculty of Philosophy at the University of Zurich. The patients/participants provided their written informed consent to participate in this study.

Author Contributions

PL is the principal investigator of the study. She drafted the Abstract, Introduction, several sections of the description of Wave 3 (Study Preparations, Eligibility, Comparison with ZLS, Adjustments to COVID-19, and Ethics), the Discussion, and the Dissemination section. She edited all other sections of the manuscript. HSa is completing her doctoral work as part of this study. She prepared the historical data and drafted the section on Design and Analysis for Wave 1 and 2. She also contributed significantly to the section on locating individuals. She provided input to the Discussion section and the section on data collection of Wave 3. FS is a senior researcher in the project and drafted the section on data collection and analysis of Wave 3. He provided input to the discussion section, in particular the section on methodological challenges. IRG is a research assistant in the project. He drafted the section on locating individuals and provided input to the discussion section and the section on contact procedures. CB is a senior researcher in the project. She drafted the section on contact procedures and edited the manuscript. In particular, she provided input to the sections on Study Preparation, Ethics, Discussion, and Dissemination. HSi is a co-investigator of the study. She provided input to the Introduction and Discussion sections and edited a final version of the manuscript. FW is a senior researcher in the Zurich Longitudinal Studies (ZLS). She provided input to the sections about the ZLS and the Discussion section. OJ is a co-investigator of the study, provided input to the sections about the ZLS and the Introduction and Discussion sections, and edited the final version of the manuscript. He is the principal investigator of the Zurich Longitudinal Studies. All the authors contributed to the article and approved the submitted version.

The project was funded by the Swiss National Science Foundation, under the umbrella of the National Research Programme 76 Welfare and Coercion; it was co-funded by the Maiores Foundation. Additional funds to expand the interview part of the study were provided by the Lotteriefonds of Zurich and the City of Zurich. A contribution to the publication was made by the Grueniger Foundation.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer MM declared a shared affiliation, though no other collaboration with the authors to the handling editor.


First and foremost, we would like to thank the study participants for taking the time to participate in the study and for sharing their stories with us. We are grateful for the benevolence of the former care institutions, who have supported the study and our staff. Our gratitude goes to all the municipalities and civil registries who have supported us in our efforts to locate individuals. Specifically, we would like to thank Anja Huber at the Zurich City Archive, Markus Stoll from the Civil Registry in Zurich, Jeannine Rauschert at the Federal Archive in Aarau, and Verena Rothenbühler at the Federal Archive in Zurich. We are grateful to Marco Hüttenmoser for granting access to the data and for being available for numerous questions. Gaby Ryffel also supported us by being available for various issues that came up. We would like to thank the students who have supported the study: Corina Conrad, Alexandra Dübendorfer, Astrid Estermann, Naomi Gnägi, Nina Graf, Lea Kolly, Nicole Kradolfer, Linda Moresi, Manuela Rüdiger, Andrea Stamm, and Christina Widmer. Our thanks go to our MMI secretariat staff and our psychology staff for supporting the study. We would also like to thank our expert project board for their support (Matthias Allemand, Andrea Büchler, Moritz Daum, Miriam Gebhardt, Florence Martin, Pasqualina Perrig-Chiello, Delia Pop, Cornelia Rumo, Diana Wider, and Hansjörg Znoj) and the staff of the Zurich Longitudinal Studies Jon Caflisch, Dominique Eichelberger, Giulia Haller, and Tanja Kakebeeke. Our thanks also go to Simon Milligan for editing the manuscript. We are thankful to the funders that made this study possible.

Ammann, R., Huonker, T., and Schmid, J. (2019). Gesichter der Administrativen Versorgung. Porträts von Betroffenen. Vol. VOL. 1 Unabhängige Expertenkommission Administrative Versorgung. Zürich: Chronos.

Google Scholar

Antonovsky, A. (1979). Health, Stress, and Coping. San Francisco, CA: Jossey-Bass, 278.

Antonovsky, A. (1987). Unraveling the Mystery of Health: How People Manage Stress and Stay Well. A Joint Publication in the Jossey-Bass Social and Behavioral Science Series and the Jossey-Bass Health Series , 1st Edn. San Francisco, CA: Jossey-Bass, 218.

Balzer, C., Berger, J.-M., Caprez, G., Gonser, A., Gutbrod, K., and Keller, M. (2011). Materialen und Normwerte für die Neuropsychologische Diagnostik (MNND). Rheinfelden: Verlag Normdaten.

Bartholomew, K., and Horowitz, L. M. (1991). Attachment styles among young adults: a test of a four-category model. J. Pers. Soc. Psychol. 61, 226–244. doi: 10.1037/0022-3514.61.2.226

PubMed Abstract | CrossRef Full Text | Google Scholar

Bastin, G. (1967). Die Soziometrischen Methoden. Bern: Huber.

Beierlein, C., Kovaleva, A., Kemper, C. J., and Rammstedt, B. (2012). ASKU – Allgemeine Selbstwirksamkeit Kurzskala. Available online at: https://psycharchives.zpid.de/handle/20.500.12034/431 (accessed December 17, 2020).

Belschner, W., Lischke, G., and Selg, H. (1971). Foto-Hand-Test (FHT) zur Erfassung der Aggressivität. Freiburg: Alber.

Berens, A. E., and Nelson, C. A. (2015). The science of early adversity: is there a role for large institutions in the care of vulnerable children? Lancet 386, 388–398. doi: 10.1016/S0140-6736(14)61131-4

CrossRef Full Text | Google Scholar

Biäsch, H., and Fischer, H. (1969). Testreihen zur Prüfung von Schweizer Kindern vom 4.-15. Altersjahr. 2. Bern: Huber.

Bick, J., Zhu, T., Stamoulis, C., Fox, N. A., Zeanah, C., and Nelson, C. A. (2015). Effect of early institutionalization and foster care on long-term white matter development: a randomized clinical trial. JAMA Pediatr. 169, 211–219. doi: 10.1001/jamapediatrics.2014.3212

Bohm, E. (1965). Lehrbuch der Rorschach-Psychodiagnostik. Bern: Huber.

Bohnsack, R., Nentwig-Gesemann, I., and Nohl, A.-M.(eds). (2013). Die dokumentarische Methode und ihre Forschungspraxis. Grundlagen Qualitativer Sozialforschung. 3. Wiesbaden: Springer VS. doi: 10.1007/978-3-531-19895-8

Bombach, C., Gabriel, T., and Keller, S. (2018a). “Legitimieren” und “integrieren” – Die Auswirkungen von Heimerfahrungen auf den weiteren Lebensverlauf,” in Fremdplatziert Heimerziehung in der Schweiz, 1940–1990 , eds G. Hauss, T. Gabriel, and M. Lengwiler (Zürich: Chronos), 253–272.

Bombach, C., Gabriel, T., and Keller, S. (2018b). “Vulnerabilität und Anerkennung. Erzählte Biografie nach Heimplatzierungen zwischen 1950 und 1990,” in Zwischen Erinnerung und Aufarbeitung Fürsorgerische Zwangsmassnahmen an Minderjährigen in der Schweiz im 20 Jahrhundert , eds B. Ziegler, G. Hauss, and M. Lengwiler (Zürich: Chronos), 83–109.

Bombach, C., Gabriel, T., Keller, S., Ramsauer, N., and Marx, A. S. (2017). Zusammen Alleine: Alltag in Winterthurer Kinder- und Jugendheimen 1950–1990. Zürich: Chronos, 224.

Borkenau, P., and Ostendorf, F. (2008). NEO-FFI. NEO-Fünf-Faktoren-Inventar nach Costa und McCrae. 2., neu normierte und vollständig überarbeitete Auflage. Göttingen: Hogrefe.

Brandt, J., and Benedict, R. H. B. (2001). Hopkins Verbal Learning Test – Revised. Administration Manual. Lutz, FL: Psychological Assessment Resources.

Brauer, S. (2019b). Gutachten zum Geplanten NFP 76-Forschungsprojekt “The Impact of Infant Institutionalization in Switzerland – a 60-year follow up” des Marie Meierhofer Instituts für das Kind. Zurich: Brauer & Strub.

Brauer, S. (2019a). Ethische Einschätzung bezüglich der Einbindung von Probanden in den beiden Marie Meierhofer-Studien zur Entwicklung von Heimkindern von 1958–1961 und 1971–1973. Zurich: Brauer & Strub.

Browne, K. (2009). The Risk of Harm to Young Children in Institutional Care. London: Save the children.

Brunet, O., and Lézine, I. (1955). Échelle de Développement Psychomoteur de la Première Enfance. Issy-les-Moulineaux: Editions scientifiques et psychotechniques.

Buysee, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., and Kupfer, D. J. (1989). The pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res. 2, 193–213. doi: 10.1016/0165-1781(89)90047-4

Center on the Developing Child (2007). The Impact of Early Adversity on Child Development. Available online at: www.developingchild.harvard.edu (accessed December 17, 2020).

Dahl, G. (1968). Übereinstimmungsvalidität des HAWIE und Entwicklung Einer Reduzierten Testform. Meisenheim (am Glan): Hain.

Dahl, G. (1972). Reduzierter Wechsler-Intelligenztest. WIP. Meisenheim (am Glan): Hain.

Duhm, E., and Hansen, J. P. F. T. (1957). Der Rosenzweig-P-F Test. Deutsche Bearbeitung der Rosenzweig “Picture Frustration Study”, Form für Kinder. Göttingen: Verlag für Psychologie Dr. C. J. Hogrefe.

D’Amato, G. (2012). “Die durchleuchtete, unsichtbare Arbeitskraft. Die italienische Einwanderung in die Schweiz in den 50er Jahren,” in Schöner Leben, Mehr Haben Die 50er Jahre in der Schweiz im Geiste des Konsums , ed. T. Pfunder (Zürich: Limmat), 237–252.

Felitti, V. J., Anda, R. F., Nordenberg, D., Williamson, D. F., Spitz, A. M., Edwards, V., et al. (1998). Relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults. The adverse childhood experiences (ACE) study. Am. J. Prev. Med. 14, 245–258. doi: 10.1016/S0749-3797(98)00017-8

Finkelhor, D., Ormrod, R. K., and Turner, H. A. (2007). Poly-victimization: a neglected component in child victimization. Child Abuse Negl. 31, 7–26. doi: 10.1016/j.chiabu.2006.06.008

Franke, G. H. (2017). BSCL. Brief-Symptom-Checklist. Deutsche Version. Göttingen: Hogrefe.

Freyd, J. J. (1996). Betrayal Trauma: The Logic of Forgetting Childhood Abuse. Cambridge, MA: Harvard University Press, 232.

Fydrich, T., Sommer, G., and Brähler, E. (2007). F-SozU. Fragebogen zur Sozialen Unterstützung. Göttingen: Hogrefe.

Gahleitner, S. B. (2009). Was Hilft Ehemaligen Heimkindern bei der Bewältigung ihrer Komplexen Traumatisierung? Expertise im AUFTRAG des Runden Tisches Heimerziehung. Berlin: Silke Birgitta Gahleitner.

Gebhardt, M. (2009). Die Angst vor dem Kindlichen Tyrannen?: eine Geschichte der Erziehung im 20. Jahrhundert. München: Deutsche Verlags-Anstalt (DVA).

Glaesmer, H., Grande, G., Braehler, E., and Roth, M. (2011). The german version of the satisfaction with life scale (SWLS): psychometric properties, validity, and population-based norms. Eur. J. Psychol. Assess. 27, 127–132. doi: 10.1027/1015-5759/a000058

Goodenough, F. L. (1926). Measurement of Intelligence by Drawings. Chicago: World Book Co.

Grosse Holtforth, M., and Grawe, K. (2003). Der Inkongruenzfragebogen (INK). Z. Für Klin. Psychol. Psychother. 32, 315–323. doi: 10.1026/0084-5345.32.4.315

Hamaker, E. L., Kuiper, R. M., and Grasman, R. P. P. P. (2015). A critique of the cross-lagged panel model. Psychol Methods 20, 102–116. doi: 10.1037/a0038889

Huber, D. (1995). Ein halbes Jahrhundert in Erwartung. Zur Geschichte der in der Schweiz bis heute fehlenden Mutterschaftsversicherungen. Olympe Fem. Arbeitshefte Zur Polit. 3, 92–99.

Huebner, G., Boothby, N., Aber, L., Darmstadt, G., Diaz, A., Masten, A. S., et al. (2016). Beyond Survival: The Case for Investing in Young Children Globally. Washington, DC: National Academy of Medicine.

Johnson, R., Browne, K., and Hamilton-Giachritsis, C. (2006). Young children in institutional care at risk of harm. Trauma Violence Abuse 7, 34–60. doi: 10.1177/1524838005283696

Joris, E. (2012). “Dezenter sex appeal – eklatante Diskriminierung. Weiblichkeits- und Männlichkeitsmythen in Zeiten von Textilrevolution und kaltem Krieg,” in Schöner Leben, Mehr Haben Die 50er Jahre in der Schweiz im Geiste des Konsums , ed. T. Pfunder (Zürich: Limmat), 105–119.

Kakebeeke, T., Knaier, E., Chaouch, A., Caflisch, J., Rousson, V., and Largo, R. H. (2018). Neuromotor development in children. Part 4: new norms from 3 to 18 years. Dev. Med. Child Neurol. 60, 810–819. doi: 10.1111/dmcn.13793

Koch, K. (1957). Der Baumtest. Der Baumzeichenversuch als Psychodiagnostisches Hilfsmittel. Bern: Huber.

Kreicbergs, U., Valdimarsdóttir, U., Steineck, G., and Henter, J.-I. (2004). A population-based nationwide study of parents’ perceptions of a questionnaire on their child’s death due to cancer. Lancet 364, 787–789. doi: 10.1016/S0140-6736(04)16939-0

Lannen, P., Bombach, C., and Jenni, O. G. (2020). Using participatory methods to develop and implement research on historic compulsory social measures and placements in Switzerland. Int. J. Child Youth Fam. Stud. 11, 96–120. doi: 10.18357/ijcyfs114202019940

Lannen, P., Wolfe, J., Mack, J., Onelov, E., Nyberg, U., and Kreicbergs, U. (2010). Absorbing information about a child’s incurable cancer. Oncology 78:259. doi: 10.1159/000315732

Lannen, P., Wolfe, J., Prigerson, H. G., Onelov, E., and Kreicbergs, U. C. (2008). Unresolved grief in a national sample of bereaved parents: impaired mental and physical health 4 to 9 years later. J. Clin. Oncol. 26, 5870–5876. doi: 10.1200/JCO.2007.14.6738

Lengwiler, M. (2018). “Aufarbeitung und Entschädigung traumatisierender Fremdplatzierungen. Die Schweiz im internationalen Vergleich,” in Zwischen Erinnerung und Aufarbeitung Fürsorgerische Zwangsmassnahmen an Minderjährigen in der Schweiz im 20 Jahrhundert , eds B. Ziegler, G. Hauss, and M. Lengwiler (Zürich: Verlag), 159–176.

Lengwiler, M., and Praz, A.-F. (2018). “Kinder- und Jugendfürsorge in der Schweiz. Entstehung, Implementierung und Entwicklung (1900-1980),” in Fremdplatziert Heimerziehung in der Schweiz, 1940-1990 , eds G. Hauss, T. Gabriel, and M. Lengwiler (Zürich: Chronos), 29–52.

Limesurvey GmbH (2020). LimeSurvey: An Open Source Survey Tool [Internet]. Hamburg: LimeSurvey GmbH.

Loch, N., Hiller, W., and Witthöft, M. (2011). Der cognitive emotion regulation questionnaire (CERQ) erste teststatistische Überprüfung einer deutschen Adaption. Z. Für Klin. Psychol. Psychother. 40, 94–106. doi: 10.1026/1616-3443/a000079

Lysholm, J., and Lindahl, B. (2019). Strong development of research based on national quality registries in Sweden. Ups J. Med. Sci. 124, 9–11. doi: 10.1080/03009734.2018.1520761

Mayring, P. (2000). “Qualitative Inhaltsanalyse,” in Qualitative Forschung Ein Handbuch , eds U. Flick, E. von Kardorff, and I. Steinke (Reinbeck: Rowohlt).

Meierhofer, M. (1958). Pädagogische Probleme der Säuglings – und Kleinkinderpflege. Pro Juventute 10, 1–3.

Meierhofer, M., and Keller, W. (1974). Frustration im Frühen Kindesalter. 3. Aufl. Bern: Hans Huber.

Meyer-Schell, I. (1971). Nachuntersuchung von Sechzehn Schulkindern, die ihre frühe Kindheit in Einem Heim Verbrachten [Lizenzarbeit]. Zürich: Universität Zürich.

Morfeld, M., Kirchberger, I., and Bullinger, M. (2011). SF-36. Fragebogen zum Gesundheitszustand. 2., Ergänzte und Überarbeitete Auflage. Göttingen: Hogrefe.

Nelson, C. A., Fox, N. A., and Zeanah, C. H. (2014). Romania’s Abandoned Children: Deprivation, Brain Development, and the Struggle for Recovery. Cambridge, MA: Harvard University Press, 402. doi: 10.4159/harvard.9780674726079

Neumann, E. (2002). “KBEK – Kurzmessung der Bindung an die Eltern in der Kindheit,” in Leibniz-Zentrum für Psychologische Information und Dokumentation (ZPID) (Hrsg.), Elektronisches Testarchiv. Trier: ZPID.

Paulus, C. (2011). Der Saarbrücker Persönlichkeitsfragebogen zur Messung von Empathie (SPF-IRI). Available online at: http://bildungswissenschaften.uni-saarland.de/personal/paulus/homepage/empathie.html (accessed December 17, 2020).

Petermann, F., and Petermann, U. (2013). WAIS-IV. Wechsler Adult Intelligence Scale – Fourth Edition. Deutschsprachige Adaptation der WAIS-IV von D. Wechsler. Frankfurt/Main: Pearson Assessment.

Puterman, E., Gemmill, A., Karasek, D., Weir, D., Adler, N. E., Prather, A. A., et al. (2016). Lifespan adversity and later adulthood telomere length in the nationally representative US health and retirement study. Proc. Natl. Acad. Sci. U.S.A. 113, E6335–E6342. doi: 10.1073/pnas.1525602113

R Core Team (2018). R: A Language and Environment for Statistical Computing [Internet]. Vienna: R Foundation for Statistical Computing.

Ramsauer, N. (2000). “Verwahrlost” Kindswegnahmen und die Entstehung der Jugendfürsorge im schweizerischen Sozialstaat 1900-1945. Zürich: Chronos.

Rey, A. (1941). L’examen psychologique dans les cas d’encephalopathie traumatique. Arch. Psychol. 28, 286–340.

Rosenthal, G. (1995). Erlebte und Erzählte Lebensgeschichte. Gestalt und Struktur biographischer Selbstbeschreibungen. Frankfurt: Campus Verlag.

Rosenthal, G. (2015). “Interpretative Sozialforschung. Eine Einführung. 5., aktualisierte und ergänzte Auflage,” in Grundlagentexte Soziologie , eds M. Diewald and K. Hurrelmann (Weinheim: Beltz Juventa).

Ryffel, G. (2013). Resilienz und Defizit. Entwicklung nach Einem Frustrierenden Start im Säuglingsheim. Zürich: Marie Meierhofer Institut für das Kind.

Ryffel, G., and Simoni, H. (2016). Forschungsdokumentation zum Projekt Lebensverläufe nach Strukturell Bedingter Frustration der Frühesten Entwicklungsbedürfnisse unter dem Aspekt der Resilienz. Retrospektive Teilstudie des Projektes Leben mit Brüchen in der Frühen Beziehungsumwelt. Zürich: Marie Meierhofer Institut für das Kind.

Sack, M., and Ebbinghaus, R. (2012). “Was hilft ehemaligen Heimkindern der DDR bei der Bewältigung ihrer komplexen Traumatisierung?,” in Aufarbeitung der Heimerziehung in der DDR Expertisen , ed. Beauftragter der Bundesregierung für die Neuen Bundesländer (Berlin: Arbeitsgemeinschaft für Kinder und Jugendhilfe – AGJ), 299–395.

Sand, H., and Gruber, L. (2017). Leben in Brüchen: Thematische Inhaltsanalyse der Interviews. (Projektarbeit im Rahmen des Masterstudiengangs Frühe Kindheit, Pädagogische Hochschule Thurgau). Zürich: Marie Meierhofer Institut für das Kind.

Schallberger, U., and Spiess Huldi, C. (2001). Die Zürcher Längsschnittstudie von der Schulzeit bis zum mittleren Erwachsenenalter. Z Für Soziol. Erzieh Sozial. 21, 80–89.

Schmidt, S., Muehlan, H., and Brähler, E. (2016). AAS-R. Revised Adult Attachment Scale. Deutsche Version. Göttingen: Hogrefe.

Schore, A. N. (2001). The effects of early relational trauma on right brain development, affect regulation, and infant mental health. Infant Ment. Health J. 22, 201–269. doi: 10.1002/1097-0355(200101/04)22:1<201::AID-IMHJ8>3.0.CO;2-9

Schumacher, J., Leppert, K., Gunzelmann, T., Strauß, B., and Brähler, E. (2004). Die Resilienzskala. Ein Fragebogen zur Erfassung der psychischen Widerstandsfähigkeit als Personmerkmal. Z. Für Klin. Psychol. Psychiatr. Psychother. 53, 16–39.

Schumacher, J., Wilz, G., Gunzelmann, T., and Brähler, E. (2000). Die sense of coherence scale von antonovsky. Teststatistische Überprüfung in einer repräsentativen Bevölkerungsstichprobe und Konstruktion einer Kurzskala. Psychother. Psychosom. Med. Psychol. 50, 472–482. doi: 10.1055/s-2000-9207

Shalev, I., Entringer, S., Wadhwa, P. D., Wolkowitz, O. M., Puterman, E., Lin, J., et al. (2013). Stress and telomere biology: a lifespan perspective. Psychoneuroendocrinology 38, 1835–1842. doi: 10.1016/j.psyneuen.2013.03.010

Sherr, L., Roberts, K. J., and Gandhi, N. (2017). Child violence experiences in institutionalised/orphanage care. Psychol. Health Med. 22, 31–57. doi: 10.1080/13548506.2016.1271951

Shonkoff, J. P., and Phillips, D.A. (eds). (2000). From Neurons to Neighborhoods: the Science of Early Child Development. Washington, DC: National Academies Press.

Sticca, F., and Perren, S. (2015). The chicken and the egg: Longitudinal associations between moral deficiencies and bullying: a parallel process latent growth model. Merrill Palmer Q. 61, 85–100. doi: 10.13110/merrpalmquar1982.61.1.0085

Strauss, E., Sherman, E. M. S., and Spreen, O. (2006). A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary. 3. New York, NY: Oxford University Press.

Surkan, P. J., Steineck, G., and Kreicbergs, U. (2008). Perceptions of a mental health questionnaire: the ethics of using population-based controls. J. Med. Ethics 34, 545–547. doi: 10.1136/jme.2007.022418

Terman, L. M., and Merrill, M. A. (1937). Measuring Intelligence. New York, NY: Houghton Mifflin.

Terman, L. M., and Merrill, M. A. (1965). Stanford-Binet Intelligenztest. S-I-T. Deutsche Bearbeitung von Prof. Dr. Heinz-Rolf Lückert. Handanweisung. Göttingen: Verlag für Psychologie Dr. C. J. Hogrefe.

Thalmann, H. C. (1971). Verhaltensstörungen bei Kindern im Grundschulalter. Stuttgart: Klett.

Thurner, F., and Tewes, U. (1969). Kinder-Angst-Test (KAT). Göttingen: Hogrefe.

Tombaugh, T. N. (2004). Trail making Test A and B: normative data stratified by age and education. Arch. Clin. Neuropsychol. 19, 203–214. doi: 10.1016/S0887-6177(03)00039-8

UEK Administrative Versorgungen (2019). UEK Administrative Versorgungen – Unabhängige Expertenkommission zur Wissenschaftlichen Aufarbeitung der Administrativen Versorgungen [Internet]. Available online at: http://www.uek-administrative-versorgungen.ch/de/Startseite.1.html (accessed September 4, 2017).

Ulijaszek, S. J., Johnston, F. E., Preece, M. A., and Tanner, J. (1998). The Cambridge Encyclopedia of Human Growth and Development. Cambridge: Cambridge University Press, 509.

Unabhängige Expertenkommission Administrative Versorgungen (2019). Organisierte Willkür. Administrative Versorgungen in der Schweiz 1930-1981. Schlussbericht. Vol. VOL. 10 A. Zürich: Chronos.

Ungricht, J. (1955). Der Sohnaufsatz als psychodiagnostisches Hilfsmittel. Schweiz Z. Für Psychol. Ihre Anwendungen. 14, 27–45.

von Staabs, G. (1964). Der Scenotest. Beitrag zur Erfassung unbewusster Problematik und charakterologischer Struktur in Diagnostik und Therapie , 4th Edn. Bern: Hans Huber.

Wehrle, F. M., Caflisch, J. A., Eichelberger, D., Kakebeeke, T. H., Latal, B., and Jenni, O. G. (2020). Health and development across the lifespan – the Zurich longitudinal studies. Front. Hum. Neurosci. Res. Top. Longitud. Aging Res. Cogn. Behav. Neurosci.

Werner, E. E. (2013). “What can we learn about resilience from large-scale longitudinal studies?,” in Handbook of Resilience in Children , eds S. Goldstein and R. B. Brooks (Boston, MA: Springer), 87–102. doi: 10.1007/978-1-4614-3661-4_6

Wyss-Wanner, M. (2000). Ein Leben für Kinder: Leben und Werk von Marie Meierhofer, 1909-1998. (Zürcher medizingeschichtliche Abhandlungen). Zürich: Juris Druck + Verlag.

Keywords : longitudinal study, lifespan development, institutional care, adverse childhood experiences, early childhood, child development, compulsory social measures and placements

Citation: Lannen P, Sand H, Sticca F, Ruiz Gallego I, Bombach C, Simoni H, Wehrle FM and Jenni OG (2021) Development and Health of Adults Formerly Placed in Infant Care Institutions – Study Protocol of the LifeStories Project. Front. Hum. Neurosci. 14:611691. doi: 10.3389/fnhum.2020.611691

Received: 29 September 2020; Accepted: 03 December 2020; Published: 20 January 2021.

Reviewed by:

Copyright © 2021 Lannen, Sand, Sticca, Ruiz Gallego, Bombach, Simoni, Wehrle and Jenni. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Patricia Lannen, [email protected]

This article is part of the Research Topic

Longitudinal Aging Research: Cognition, Behavior and Neuroscience

JavaScript is currently disabled, this site works much better if you enable JavaScript in your browser.

*indicates required fields

Karger Publishers

Human Development

Learning as an important privilege: a life span perspective with implications for successful aging.

Author affiliations a UC Riverside, Riverside, CA, USA b University of British Columbia, Vancouver, BC, Canada c Johns Hopkins University, Baltimore, MD, USA d University of the Pacific, Stockton, CA, USA

Keywords: Life span learning Adult development Functional independence Cognitive development

Related Articles for " "

Log in to MyKarger to check if you already have access to this content.

Buy a Karger Article Bundle (KAB) and profit from a discount !

If you would like to redeem your KAB credit, please log in .

Rent via DeepDyve

Start free trial

Subcription rates

Article / Publication Details

Abstract of Article

Received: May 20, 2020 Accepted: October 29, 2020 Published online: March 02, 2021 Issue release date: April 2021

Number of Print Pages: 14 Number of Figures: 0 Number of Tables: 0

ISSN: 0018-716X (Print) eISSN: 1423-0054 (Online)

For additional information: https://www.karger.com/HDE

Research has demonstrated the cognitive and mental health benefits of learning new skills and content across the life span, enhancing knowledge as well as cognitive performance. We argue that the importance of this learning – which is not available equally to all – goes beyond the cognitive and mental health benefits. Learning is important for not only the maintenance, but also enhancement of functional independence in a dynamic environment, such as changes induced by the COVID-19 pandemic and technological advances. Learning difficult skills and content is a privilege because the opportunities for learning are neither guaranteed nor universal, and it requires personal and social engagement, time, motivation, and societal support. This paper highlights the importance of considering learning new skills and content as an important privilege across the life span and argues that this privilege becomes increasingly exclusionary as individuals age, when social and infrastructural support for learning decreases. We highlight research on the potential positive and negative impacts of retirement, when accessibility to learning opportunities may vary, and research on learning barriers due to low expectations and limited resources from poverty. We conclude that addressing barriers to lifelong learning would advance theories on life span cognitive development and raise the bar for successful aging. In doing so, our society might imagine and achieve previously unrealized gains in life span cognitive development, through late adulthood.

© 2021 S. Karger AG, Basel

Copyright / Drug Dosage / Disclaimer


2021, Vol.65, No. 1



Privacy Policy | Terms of Use | Imprint | Cookies

Karger International: COM | CN

© 2023 S. Karger AG, Basel

Sign up for MyKarger

Manage with MyKarger your orders simply and fast, save your favorite articles in your reading list, edit your newsletter profile and benefit from attractive discounts.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Log in to MyKarger?

We found an already existing MyKarger account with this e-mail address:

Complete Your Name

Thank you for registering.

Your MyKarger account has been created.

Reset Your Password

To reset your password, enter your e-mail address or your user ID you registered with. You will be sent an e-mail containing a link to reset your password.

Check Your E-Mails

A link to reset your password has been sent to your e-mail address. Follow the instructions and try to log in again.

Stay Up to Date Banner

SIPS logo

Conceptualizing Goals and Goal Focus

Goal focus across the lifespan, towards a lifespan developmental account of goal focus, current challenges, future directions, conclusions, contributions, acknowledgements, funding information, competing interests, data accessibility statement, for whom is the path the goal a lifespan perspective on the development of goal focus.

Correspondence concerning this article may be addressed to either author at University of Zurich, Department of Psychology, Binzmuehlestrasse 14, Box 11, 8050 Zurich, Switzerland, [email protected]

[email protected]

[email protected] Alexandra M. Freund and Moritz M. Daum share the senior authorship

Lea Moersdorf, Moritz M. Daum, Alexandra M. Freund; For Whom Is the Path the Goal? A Lifespan Perspective on the Development of Goal Focus. Collabra: Psychology 5 January 2022; 8 (1): 31603. doi: https://doi.org/10.1525/collabra.31603

Download citation file:

Goals are an intensely studied concept in various research areas within psychology. They can be defined as cognitive representations of means-ends relations. The relative focus on the means or the ends (i.e., goal focus) can vary between persons and over time. Taking a lifespan perspective, we use the existing developmental, social-cognitive, and motivational literature to portray how goal focus might develop across the entire lifespan. For this purpose, we take findings on the perception of goal-directed behavior in infancy, the development of (self-)representations and goal pursuit in adolescence, and of goals across adulthood into account. We propose that goal focus changes across the lifespan due to age-related cognitive and motivational development, and that the relative impact of cognitive and motivational processes on goal focus varies across the lifespan. We conclude by integrating different approaches and findings from a lifespan perspective.

Setting and pursuing goals are fundamental processes that guide human behavior over time and across situations (e.g., Kruglanski, 1996 ). Goals substantially contribute to subjective well-being and serve as a source of meaning in life (e.g., Emmons, 1996 ). They constitute a major topic in psychological science, for instance in motivational psychology (e.g., Austin & Vancouver, 1996; Kruglanski et al., 2002 ), social psychology (e.g., Fishbach & Ferguson, 2007 ), industrial and organizational psychology (e.g., Locke & Latham, 2002 ), health psychology (e.g., Lüscher et al., 2017 ), and educational psychology (e.g., Cain & Dweck, 1995; Zimmerman & Kitsantas, 1997 ). Goals have also been a research topic in developmental psychology including infant and child development (e.g., Carpenter et al., 2005; Daum et al., 2008; Gampe et al., 2016; Hunnius & Bekkering, 2014; Woodward, 1998 ), adolescent development (e.g., Gestsdottir & Lerner, 2008; Salmela-Aro, 2009 ) as well as adult development and aging (e.g., Freund et al., 2019; Hamm et al., 2016 ). However, the two developmental research strands – goal research in infancy and childhood on the one hand, and goal research in adulthood and aging on the other – come from different perspectives and involve different theoretical as well as empirical approaches. Consequently, there is a dearth of goal research that encompasses the entire lifespan, including consistent terminology and an overarching theoretical framework (as is true for many research topics; for notable exceptions in action or goal research see e.g., McGuigan et al., 2011; Wermelinger et al., 2019 ).

We maintain that a lifespan perspective on goal focus, that is, the relative salience of the means and ends of goal pursuit, is theoretically and methodologically fruitful for multiple reasons: First, from an action-theoretical perspective, goals direct human behavior throughout the lifespan, and constitute an important way in which humans actively shape their development. Second, given the impact of goals on cognition, emotion, and behavior, understanding what kinds of goals humans pursue and how they represent them in different phases of the lifespan constitutes a promising venue to investigate developmental changes in cognition, emotion, and behavior. Third, the distinction of means and ends (which is at the heart of the construct of goal focus) is also central in the literatures concerning different phases of the lifespan, such as action perception in infancy (e.g., Carpenter et al., 2005 ), developing identity-relevant goals in adolescence (e.g., Salmela-Aro, 2009 ), and motivational development in adulthood (e.g., Mustafić & Freund, 2012b ). Integrating these research strands in a lifespan approach into one overall framework encourages exchanges and critical discussions among developmental researchers but also with motivation scientists. This might lead to a better understanding of developmental processes across the lifespan as well as of the goal construct itself. For instance, discussing questions such as “What constitutes a goal?” from different perspectives can sharpen the distinction of goals from wishes or actions. Finally, adding a large age range to the study of goal focus introduces greater variability in certain goal dimensions, such as their content, abstractness, or temporal scope, as well as in development-related processes, such as abstract thinking. This variability can help to pinpoint under which conditions theoretical assumptions about the underlying processes and functions of goal focus might hold. For instance, goal focus research constrained to adulthood might neglect the role of cognitive processes impacting goal focus during childhood because these processes show little variability across adulthood. In sum, a lifespan perspective on goal focus promises a more comprehensive view of goals and their development as well as a more thorough understanding of developmental processes that influence goal focus.

In this paper, we propose such a lifespan perspective on goal focus. To this end, we integrate findings from infancy to adulthood and into old age into one theoretical perspective on the factors influencing goal focus across the lifespan. With this, we aim to stimulate both theoretical discussions and empirical research on goals across the lifespan.

The existing developmental literature on goal focus is limited and pertains primarily to changes across adulthood. Consequently, it is currently not possible to offer a systematic review or meta-analysis of empirical studies on this topic. Therefore, we build our theoretical model on the basis of research that distinguishes between a focus on the means and outcomes of goal pursuit in different age phases. For infancy and early childhood, we consider the literature on the perception of goal-directed behavior (henceforth called action perception). Within this literature, we focus particularly on research that allows to differentiate between the perception of actions and their end states. Furthermore, we include research on broader developmental processes that may help explain changes in goal focus during childhood and adolescence, such as the development of self-representations, which are indicative of the development of cognitive representations more generally. For the adult segment of the lifespan, we review the theoretical and empirical work on developmental changes in goal focus across adulthood.

To establish common ground between the considered literatures, we first provide a definition of goals and then delineate the concept of goal focus, the main concept of this article. Afterwards, we turn to our lifespan perspective on goal focus.

Goals can be defined as cognitive representations of the association of means and ends. In other words, goals include both, the ends (or end states) one wants to attain and the means (or actions) to attain them (Kruglanski et al., 2002) . Although this definition is widely accepted in the motivational literature, it differs notably from the one used mostly in infancy and early childhood research, where the term goal is used more ambiguously and often without an explicit definition (for an exception, see Loucks et al., 2017 ). In most research on early childhood, goals refer to the concrete and observable “endpoint” or “outcome” of an action such as the grasping or the manipulation of an object, or in some cases a certain object itself (e.g., Carpenter et al., 2005; Elsner & Pfeifer, 2012 ). Often, it remains unclear whether the term ‘goal’ refers to the endpoint itself or the representation of the endpoint, and whether means are also included in this representation. This ambiguous use of the term ‘goal’ is likely due to the limited cognitive capacities of infants and young children (e.g., they cannot yet verbalize desired end states and their relations to specific means); consequently, research paradigms target primarily the perception and production of simple object-directed manual actions.

In the present paper, we use the term goal as defined in the motivational literature, namely as a cognitive representation of a desired (or dreaded) outcome a person can approach (or avoid) through specific means (e.g., specifically chosen actions; Kruglanski, 1996 ). In this definition, goals can vary with regard to their abstractness and temporal scope (e.g., Austin & Vancouver, 1996; Carver & Scheier, 1990 ), with some goals referring to temporally closer states than others (e.g., “I want to pass the exam tomorrow” vs. “I want to receive my PhD in the next four years”), and some goals being more concrete than others (e.g., “I want to run a marathon in less than 3 hours and 30 minutes” vs. “I want to lead a happy life”). Thus, also very concrete and temporally close goals, such as grasping a toy in order to play with it, are included in this definition. Therefore, this definition provides enough flexibility to refer to the kinds of goals infants and young children perceive (and pursue) as well as to the increasingly abstract goals as children develop into adults. We use this definition as a lens through which we view the existing theoretical and empirical literature. Instead of keeping each article’s individual wording, we translate the thoughts and findings into a common language based on this definition to facilitate integration.

We use the term outcome to denote the intended ends or consequences of goal pursuit, that is, why a person pursues a certain goal. Note that this can be different from (or only a part of) the actual outcome (e, g., when things go wrong, or a person does not anticipate the full range of consequences of their behavior). We concentrate on intended outcomes here, because goals would be mere behavior-effect associations if they were stripped off all intentionality. With means , we describe how to reach a certain outcome, comprising concrete actions, 1 such as grasping an object or studying for an exam, as well as more abstract means such as “being friendly.” Whereas means provide guidelines for action, outcomes serve to give direction and meaning (Freund & Hennecke, 2015). Typically, means are more concrete relative to their outcomes, which is related to the hierarchical organization of goals: The outcome of a lower order (more concrete) goal can constitute the means of a higher order (more abstract) goal (e.g., Carver & Scheier, 1990, 1998; Kruglanski et al., 2002 ). For example, the outcome of completing a PhD may be the means to get a job; getting a job, in turn, may be the means that enables one to live independently; which is another step on the way to the potential ultimate goal of achieving a happy life. It is therefore essential in goal research to specify the status of means and outcomes in reference to a given goal in a larger goal system. 2 When referring to the position of a goal in the goal hierarchy, we will speak of goal complexity: A goal is more complex, the further up it is in the hierarchy of a goal system (i.e., the more levels of subgoals are below it and therefore the more links to subgoals or concrete actions it encompasses).

Goals also vary in their temporal extension (e.g., Austin & Vancouver, 1996 ). While some goals may span only minutes or hours, other goals may span years or even decades. The time frame can refer to means as well as outcomes: means may be applied for shorter or longer time periods, and an outcome may be achieved in the near or distant future. However, relative to the means, the outcome of a given goal is temporally (at least slightly) more distant, due to their cause-effect relation. Taking a developmental approach, we assume that there are differences in the temporal extension and complexity of children’s compared to adults’ goals: While adults are able to represent simple, temporally close goals as well as complex, temporally distant goals, young children’s understanding of time (e.g., Friedman, 2005 ) and their ability to form abstract representations is still limited (e.g., Harter, 1999 ), resulting in a limited ability to represent temporally distant and complex goals.

Imagine Larry who wants to become friends with Paula. One strategy could be that Larry is particularly attentive to her. In this case, he focuses on the means, that is, the way how to befriend Paula. However, Larry could also focus on what the friendship with Paula means to him, for example, having a trusted companion. In that case, Larry primarily focuses on why to befriend Paula, that is, the consequences or the outcome of the friendship with Paula. These two kinds of goal focus are referred to as focus on the means (i.e., process focus) and focus on the ends (i.e., outcome focus). Adopting a process focus implies that the how (the means or process) of goal pursuit is more salient, whereas the why (the end or outcome) is more salient in an outcome focus (Freund & Hennecke, 2015) . Note, that when a person focuses on the means of goal pursuit at a given point in time this does not imply that the associated outcome is irrelevant or not represented – it is just less salient than the means (else it would be a representation of behavior). Conversely, adopting an outcome focus does not imply that the associated means to achieve it are irrelevant or not represented (else it would be more of a representation of a wish or dream). In other words, goal focus is defined by which aspect of goal pursuit is more   salient to a person at a specific point in time. Further, goal focus refers to the relative salience of the means or the outcomes within a given goal and not a general preference for higher-order (abstract) or lower-order (concrete) goals.

Goals often comprise multiple means that are linked to the same outcome (i.e., equifinality) or multiple outcomes that are linked to the same means (i.e., multifinality). For instance, Larry might consider multiple means such as “giving Paula little presents”, and “telling her secrets” to achieve the desired outcome of having a trusted companion. Similarly, confiding secrets in Paula might serve multiple outcomes for Larry, such as establishing Paula’s trust and also to regulate his emotions associated with the secret. Depending on the goal, some means and/or outcomes might seem to be more typical (or less typical) than others, and there are likely individual differences in this perception. For instance, some people might perceive “having a trusted companion” to be the most typical description of (or even identical to) the goal “to befriend Paula”. Others might perceive “telling her secrets” to be the best description.

We address goal focus as a relevant goal dimension for two reasons: First, to better understand how children construe goals and, thereby, how they learn about the associations of means and outcomes, it is essential to distinguish between these two aspects of a given goal and investigate how their relative salience changes across childhood. Given the enormous strides in cognitive development (especially regarding social-cognitive abilities and the formation of abstract representations) from infancy to adolescence, we hypothesize that goals as cognitive representations change accordingly regarding their relatively more concrete means and their relatively more abstract outcomes. As of yet, research on changes in goal focus has exclusively centered on the development across adulthood. Hence, a comprehensive description of how goal focus changes across the entire lifespan does not yet exist. Second, there is growing evidence that goal focus has predictive power for the success of goal pursuit as well as subjective well-being, for instance in dieting (Freund & Hennecke, 2012) , exercising (Freund et al., 2010; Kaftan & Freund, 2020) , and academic achievements in adults ( Krause & Freund, 2016 ; for an overview, see Kaftan & Freund, 2018 ). Consequently, goal focus might also be a meaningful predictor of successful goal pursuit in children and adolescents.

How does goal focus change across the lifespan? Coming back to our example of Larry and Paula, the main question of this paper is how the focus on the means or the outcomes of pursuing the friendship with Paula differs depending on whether Larry is 5, 10, 20, 40, or 70 years old. As we take a lifespan perspective, we consider developmental trends that might underlie these age-related differences in goal focus. Of course, there are also inter- and intraindividual variations in goal focus within age groups (e.g., depending on the specific goal and phase of goal pursuit or a person’s cultural background; e.g., Freund et al., 2019; Miyamoto et al., 2013 ). However, here we focus exclusively on developmental changes. In particular, we propose that the relative importance of cognitive and motivational factors for goal focus differs across the lifespan: We hypothesize that aspects of cognitive development primarily drive changes in goal focus from infancy throughout childhood to adolescence. This includes basic cognitive abilities (e.g., processing speed, memory capacity, forming action-effect associations), advanced cognitive capacities (e.g., understanding of time, formation of abstract representations, ability to represent the future), and social-cognitive abilities (e.g., perspective taking, normative learning). In contrast, we posit that motivational development primarily drives changes in goal focus throughout adulthood and into old age. This includes, for example, changes in people’s orientation towards gains and losses as well as changes in the temporal distance of goals (for an overview, see Figure 1 ).


Note, that early approaches to the development of effectance motivation also stress the importance of having an effect in the environment and the associated feeling of efficacy as a motivational factor (e.g., “self-efficacy,” e.g., Bandura, 1993 ; “effectance,” White, 1959 ). We do not challenge this view, but assume that this motivational factor stays more or less constant across the lifespan (although not necessarily across situations/domains, Bandura, 1993 ) and generally affects social, cognitive, affective, and motivational processes (see the principle of the primacy of primary control suggested by J. Heckhausen & Schulz, 1995 ). Similarly, the above-mentioned cognitive abilities constitute an important basis for representing goals and therefore also goal focus. Because we assume that with adulthood these abilities are fully fledged (see Figure 1 ), they should not impact goal focus in a particular manner across adulthood. To illustrate this, the bottom part of Figure 1 acknowledges a certain impact of motivational development in early childhood and a certain impact of cognitive development in adulthood.

Additionally, societal and social factors might impact goal focus. For example, the structure and routines of the respective society might impact goal focus in certain life phases, such as during formal education (e.g., an increasing focus on tangible outcomes such as grades; we consider this in the section on goals inside and outside the school context). Also, social influences (e.g., through the affiliation motive) might contribute to a focus on the means or outcomes. For instance, in the context of normative behavior, the exact way how a certain ritual is done might become especially salient to demonstrate belonging to a certain group of people. So, in such a context, the concrete means might be more salient than the outcome. However, there is no clear evidence how social influences might systematically contribute to changes in goal focus across the entire lifespan. Therefore, we only mention them where we perceive an added value.

In the following, we will describe the research not chronologically moving from infancy to old age, but instead we start with adulthood. We do so because there exists some research that particularly investigated how goal focus changes across adulthood, whereas the evidence for the other age groups is far more limited. Also, the terms “goal” and “goal focus” are more clearly defined in the literature in adulthood. After the section on goal focus across adulthood, we turn to infancy and early childhood, and then to adolescence. We conclude with an integration of the approaches and findings from a lifespan perspective.

Goal Focus Across Adulthood

Theoretical considerations.

One of the most dramatic changes across adulthood is the decrease of (subjectively) available resources. 3 Whereas developmental gains are prominent in younger age groups (e.g., increases in available money or social capital across young and into middle adulthood), developmental losses become more and more ubiquitous with increasing age, particularly in old age (e.g., decreases in physical strength or life time; Baltes et al., 2006; Mustafić & Freund, 2012a ). Moreover, young adults may not yet have had the opportunity to acquire a large number of resources (but only start to accumulate them), whereas older adults are more likely to have accumulated resources such as money or social capital. Therefore, younger adults can be expected to be highly motivated to accumulate resources that can then be invested into goal pursuit. In contrast, with increasing age and, in many cases, more accumulated resources, adults are likely to be more motivated to maintain their resources and shield them against losses (Gong & Freund, 2020) . In fact, correlative and experimental evidence supports the hypothesis that goal orientation towards gains, maintenance, and avoidance of loss shifts across adulthood: Compared to older adults, younger adults hold and select primarily gain oriented goals (e.g., “I want to improve my physical fitness”), whereas middle-aged and older adults are more maintenance oriented and loss-avoidant in their goals (e.g., “I want to stay physically fit”/“I do not want to lose my physical fitness”; Ebner et al., 2006 ). Moreover, younger adults are more persistent in pursuing tasks that are aimed at gains, whereas older adults pursue a task longer when it is oriented towards bringing back losses (Freund, 2006) . As argued and empirically supported by Mustafić and Freund (2012b) , the motivational orientation towards maintenance and the avoidance of losses are related to a stronger focus on the means, because these goals do not specify an endpoint towards which to strive. For instance, wanting to stay healthy is not achieved at any particular point in time and can therefore not be abandoned as some other goals are after their completion, but instead requires continued goal pursuit. Maintenance or loss-avoidance goals are typically extended over longer periods of time and require a sustainable way to pursue them (e.g., working out regularly, healthy diet), which, in turn, necessitates a stronger monitoring of the means of goal pursuit. Even though people might check on the desired outcome (i.e., the status quo) from time to time, there is little need for close monitoring of the outcome because no change is expected (instead, “continue what I have been doing” should be in the center). Thus, because of an increasing maintenance orientation across adulthood, focusing on the means becomes also more likely. The opposite is the case for gain-related goals, which are tied to a change of the actual towards the desired state. When focusing on the achievement of gains (e.g., getting a degree), there is often an endpoint signaling when it has been achieved (e.g., graduation). To monitor progress in goal pursuit (i.e., a reduction of the actual-desired discrepancy), the actual and desired state have to be compared regularly (e.g., number of required classes successfully completed; Carver & Scheier, 1990 ), which should render the outcome more salient than the means (Freund et al., 2019) . Taken together, shifts in goal orientation across adulthood (as indicated in the top part of Figure 1 ) contribute to younger adults’ stronger outcome focus and older adults’ stronger focus on the means.

Future time perspective is another construct that might contribute to a developmental shift from outcome focus to a focus on the means in adulthood. Future time appears more extended in younger adults and becomes increasingly shorter with age (Lang & Carstensen, 2002) . As proposed by Socioemotional Selectivity Theory (SST; Carstensen et al., 1999 ), people’s social goals relate to their perception of future time (Carstensen et al., 1999; Lang & Carstensen, 2002) . According to construal level theory (e.g., Trope & Liberman, 2003 ), people represent events that lie in the distant future (or are generally psychologically more distant) more abstractly (i.e., high-level construals), and events that lie in the near future (or are psychologically closer) more concretely (i.e., low-level construals). With regard to actions, these low-level construals are represented in terms of “ how ” aspects, whereas high-level construals are represented in terms of “ why ” aspects of actions ( Trope & Liberman, 2003 ; see also Vallacher & Wegner, 1987 ). Thus, construal level theory predicts that, because of a shorter future time perspective, older adults are more likely to focus on the means of goal pursuit. Younger adults, having an extended future time perspective, should be more likely to hold goals that lie in the distant future, represent them abstractly, and hence focus on the outcomes (Freund et al., 2019) .

There are also reasons to argue for the opposite developmental trajectory of goal focus based on research on skill acquisition. According to Zimmerman and Kitsantas (1997) , goal focus depends on skill level, such that people are likely to adopt a focus on the means during skill acquisition and to switch to an outcome focus once they have mastered the skills. One might argue that older adults have acquired many of the skills they need for goal pursuit and should therefore be able to concentrate on the outcomes they want to achieve. In contrast, younger adults might more likely still need to acquire many of the skills and means necessary for goal pursuit (e.g., to establish a career) and therefore focus more on the means than on the outcomes.

Apart from these motivational changes across adulthood, cognition also undergoes significant changes, such as a decline in fluid abilities (e.g., Li et al., 2004 ). However, the cognitive changes are not likely to impact goal focus for two reasons: First, factors that contribute to how one represents goals are mainly related to crystallized abilities, such as knowledge, which typically do not decline substantially across adulthood until the very end of life. Second, even if such changes as cognitive slowing were to affect the representation of goals, there is no reason to assume that it would do so differentially for means or outcomes. 4

Empirical Evidence for Age-Related Changes in Goal Focus Across Adulthood

Research that investigated age-related changes in goal focus is limited to few studies (e.g., Freund et al., 2010; Mustafić & Freund, 2012b ), which we will report in more detail. The overarching goal of these studies was to find out whether there are age-related differences in goal focus between younger and older adults. More precisely, Freund et al. (2010) tested the opposing hypotheses outlined above whether goal focus shifts from the outcomes towards the means with age (as predicted by differences in goal orientation and future time perspective) or vice versa (as predicted by differences in skill level). For their first study, the authors developed a questionnaire to assess preference for goal focus. It contained different goals that were described in terms of the means or outcomes of goal pursuit (e.g., example goal: to quit smoking, example means: not buy cigarettes, example outcome: reduce toxins in body). Participants indicated which descriptions fitted the goal best in their opinion. Whereas younger adults chose more outcome- than means-related descriptions, older adults did not show a clear preferential pattern. Importantly, however, older adults chose more means-related descriptions relative to younger adults. Confirming this pattern using a different paradigm in a second study, older adults were more likely than younger adults to select a task that focused on the means rather than the outcomes of a given goal.

The third study tested the hypotheses in everyday life with younger and older adults who all shared the goal to start exercising regularly. Goal focus was operationalized as people’s motives to exercise: Items indicating the enjoyment and social aspects of engaging in exercising as reasons served as indicator of a focus on the means, while items concerning the outcomes of health, weight control, and attractiveness indexed an outcome focus. As expected, older adults were more likely to report a focus on the means than an outcome focus. Moreover, younger adults were more outcome-focused than older adults. However, contrary to expectations, both age groups reported similar levels of focusing on the means. A similar pattern of results was found by Mustafić and Freund (2012b) . In this study, participants aged between 18 and 82 years listed their personal goals and rated them regarding the salience of the means and the outcomes. Age was significantly positively related to focusing on the means but not the outcome. This pattern of age-related differences in goal focus across adulthood has also been found in studies that were primarily concerned with the adaptiveness of goal focus (Kaftan & Freund, 2020) . Taken together, these studies provide evidence that goal focus varies across adulthood. Although the results differ slightly across studies, the general pattern suggests that older adults focus more on the means and younger adults more on the outcome.

In summary, theoretical considerations as well as empirical findings indicate an increasing focus on the means (and decreasing focus on the outcomes) across adulthood. We propose that motivational processes are the major driving forces for this change. More specifically, we assume an increasing focus on maintenance and loss-prevention goals as a driving process (Mustafić & Freund, 2012b) , as well as an increasing pursuit of goals that lie in the closer future due to the limited future time older adults experience (Freund et al., 2019; see Figure 1 ). The proposed influence of skill acquisition does not seem to play a major role in age-related differences of goal focus, at least not in a general fashion as proposed before. Older adults might also need to acquire skills for new goals, and consequently focus on the means.

As there is no theoretical or empirical work on the concept of goal focus in infancy, childhood, and adolescence, we refer to the relevant literature on the development of action perception in infancy and early childhood (preschool age), and broader developmental processes in school-aged children and adolescents (e.g., the development of goals and other representations, decision-making and risk-taking) in the next sections. 5

Goal Focus in Infancy and Early Childhood

In this section, we aim to answer the question how goal focus develops before reaching adulthood and turn to research from infancy to early childhood. To this end, we first summarize some major findings on the perception of goal-directed behavior, that is, action perception. This research informs us from which age on infants are sensitive to the goal-directedness of (others’) actions and lets us conclude that they have a basic form of action-outcome-representations, conforming to our definition of goals. We consider the ability to form action-outcome-representations as an important precondition for goal focus and hence briefly present research on this. Then, we present findings that can inform us about infants’ and young children’s goal focus.

Perception of Goals and Goal-Directedness in Infancy and Early Childhood

How do infants and children perceive actions as being directed towards certain outcomes? How do they learn about the goal-directedness of actions? According to ideomotor theory (Greenwald, 1970; James, 1890) , perceiving and knowing the effects (i.e., the outcome) of an action is important for action planning and control. In infancy, initially arbitrary movements become associated with the outcome they produce, so that bidirectional action-outcome representations are formed, with “action” and “outcome” referring to the intended movement and anticipated effect (for a review, see Shin et al., 2010 ). Therefore, activating the representation of an outcome will activate the motor pattern as well and enable intentional movements (i.e., actions). Although ideomotor theory cannot explain how infants learn about others’ intentions, it explains how their own basic goals (i.e., cognitive representations linking actions/means and outcomes) emerge through experience. The link between the representations of own and others’ actions is discussed, for example, in the common coding approach (Prinz, 1997) . Here, action perception and production are assumed to be closely intertwined, in that “there are certain products of perception on the one hand and certain antecedents of action on the other that share a common representational domain” (Prinz, 1997, p. 152) . This idea has been empirically supported (e.g., Gerson et al., 2015; Reid et al., 2019 ). Together with ideomotor theory, common coding helps to explain how infants learn about others’ goals through the bidirectional association of sensory and motor representations. For example, once infants have an action-outcome representation of, for instance, their own grasp, seeing a person’s grasping hand will also activate their action code of grasping, and further the associated outcome representation (e.g., attaining a certain object). Support for this comes from behavioral and neurophysiological research. Eye-tracking data shows that the prediction of the outcome of an observed action is related to the subsequent imitation of the same action if shared motor representations are involved (Gampe et al., 2016) . Further, 7-month-olds show motor system activation when observing an adult’s goal-directed behavior, which in turn predicts the infant’s likelihood of imitating the behavior (Filippi et al., 2016) .

Multiple studies support the notion that infants do not only behave in a goal-directed way from early on but are also able to perceive others’ behavior as goal-directed. For instance, infants as young as 10 months anticipate the outcome of observed actions (as indicated by predictive gaze shifts, i.e., their gaze arrives at the target position before the observed agent does) both when performing the action themselves, but also when observing others, though to a lesser extent ( Rosander & von Hofsten, 2011 ; for a similar finding in adults, see Flanagan & Johansson, 2003 ). The frequency of predictive gaze when observing others increases with age (Gampe et al., 2016) , but also depends on the salience of the outcome (Henrichs et al., 2012) , and the dynamics of the action (Daum et al., 2016) . Furthermore, infants’ action prediction improves with their own motor experience (Krogh-Jespersen & Woodward, 2018) , and action production is related to subsequent action prediction (Cannon et al., 2012) . Apart from predictive-gaze studies, results of a violation-of-expectation looking time paradigm suggest that children from 6 to 12 months of age expect an agent to perform the most efficient movement when performing an action (e.g., Gergely & Csibra, 2003 ). Beyond this, 10-month-olds seem to take the costs of actions into account when evaluating outcomes (Liu et al., 2017) . Further evidence for children’s early sensitivity to others’ goals and intentions comes from imitation paradigms: 14- to 18-month-olds imitate 6 intentional movements of an agent at a much higher rate than accidental ones (Carpenter et al., 1998) and 18-month-olds reproduce the intended outcome of an observed action, even when observing a failed attempt (e.g., Meltzoff, 1995 ). Together these results indicate that infants are cognitively able to perceive the goal-directedness of actions as well as predict their outcomes and therefore have at least a rudimentary form of action-outcome representation. The abilities to perceive others’ behavior as goal-directed and to correctly predict action outcomes increase with age and experience but are also task-dependent. Also, actions are more easily perceived as goal-directed when resulting in a salient action effect (e.g., Henrichs et al., 2012; Jovanovic et al., 2007 ). Infants’ sensitivity to others’ goals is an important premise for the findings in the next section, because the paradigms used there largely rely on the observation of others’ actions.

Empirical Indications for Goal Focus in Infancy and Early Childhood

Whereas the above-mentioned studies inform us about infants’ cognitive ability to perceive actions as goal-directed, and to form action-outcome representations, most of them do not allow to draw any conclusions regarding infants’ goal focus (i.e., whether they focus more on the means or outcomes). For instance, one might argue that children’s production of a successful action when observing a failed attempt serves as an indicator for an outcome focus in early childhood. Instead of reproducing the exact same (unsuccessful) means, children diverge from the movement pattern to bring about the intended outcome. However, it is also possible that children do not encode the unsuccessful movement pattern (e.g., pulling on an object and then slipping) but only the target movement (e.g., pulling), which can then be imitated together with the intended outcome. This would not allow any interpretation regarding young children’s goal focus.

With the help of other paradigms, we can speculate whether the means or the outcome is more salient for infants and small children. One informative paradigm was introduced by Woodward (1998) . It consists of a habituation phase in which infants see a person repeatedly grasping the same of two toys in the same location. After habituation, the location of the toys is swapped, and infants see the person either grasp the “old” toy in the “new” location or the “new” toy in the “old” location. Because infants around the age of 6 months look longer when the “new” toy is grasped, Woodward (1998) inferred that infants expect the person to grasp the same toy as before instead of making the same movement (i.e., reach to the “old” location). Apart from indicating infants’ ability to perceive goal-directedness, this can be interpreted as a focus on the outcome instead of the means. Further support for a stronger focus on the outcome than the means in infancy and early childhood comes from Carpenter et al. (2005) : When infants saw a toy mouse being moved into a house in one of two action styles (either hopping or sliding), they ignored the action style and simply put the mouse in the house. However, when the house – a salient end position of the action – was absent, they were much more likely to imitate the action style. The authors conclude that in the house condition, infants perceived the ‘putting into the house’ as the intended outcome, while in the no-house condition they viewed the action as an outcome in itself (for similar results in adults, see Schachner & Carey, 2013 ). Therefore, in both cases (as well as in the looking-time paradigm by Woodward, 1998 ) infants focused on the outcome in their imitation. However, one might argue that in the two mentioned studies the means and outcomes were not always salient or distinct enough (e.g., the direction of grasping in Woodward, 1998 and the endpoint in Carpenter et al., 2005 in the no-house condition). For the imitation paradigm by Carpenter et al. (2005) , this means that infants either perceive the action style or the endpoint (the house) as the relevant outcome in this situation, depending on the salience of the endpoint and the distinctness of means and outcomes. This indicates that characteristics of the paradigm exert an influence on infants’ behavior.

Furthermore, research on means-end behavior in infants provides similar evidence for an outcome focus. By the age of 7-10 months, infants start to produce two-step actions (e.g., pulling a cloth towards oneself in order to grasp a toy) that are directed towards the higher-order outcome (i.e., receiving the toy) instead of the first action step (i.e., pulling on a certain cloth; e.g., Gerson et al., 2015; Sommerville & Woodward, 2005; Willatts, 1999 ). Before this age, infants seem to pull the cloth towards them for its own sake (Willatts, 1999) . It has been stated that this development reflects a shift from a focus on the means (i.e., the cloth-pulling) towards the outcome during skill acquisition (e.g., Gerson et al., 2015 ). Alternatively, one could argue that instead of focusing on the means, younger infants also focus on the outcome but perceive what has been labeled a means as an outcome in itself (e.g., receiving a certain cloth; similar to the no-house condition in Carpenter et al., 2005 ). Independent of the specific interpretation with regard to goal focus, this work suggests that infants around that age learn to engage in means-end behavior, and also start to differentiate between means and outcomes when observing others, dependent on their own experience (e.g., Gerson et al., 2015; Willatts, 1999 ).

In order to clearly distinguish between means and outcomes, Wagner et al. (2008) tested the imitation of different directed motion events in 33-month-old children. These events consisted of a motion path (up or down a ramp), a movement manner (hopping/sliding), and an end position (in or on a bowl). For example, a puppet could hop up the ramp and sit on the bowl. Unlike usual imitation paradigms, children could not imitate exactly what they had seen, but had to choose whether to imitate path and manner (i.e., the means) or the end position (i.e., the outcome) of the action, as the location of the two bowls was swapped on their ramp. Children showed evidence for a focus on the means; they primarily imitated the means at the expense of the outcomes. However, the low salience of the outcomes compared to the means might have impacted these results (Wagner et al., 2008) .

Elsner and Pfeifer (2012) adapted the paradigm to investigate the role of the salience of the outcomes in the imitation choice task. In this new setup, the authors added a condition with more salient outcomes (i.e., the puppet aimed at arriving at a bench or a boat instead of the more unusual bowls used in the previous experiments) as well as conditions including verbal cues to highlight parts of the action. The variation in setup had a strong impact on children’s imitation behavior: When confronted with the more salient outcomes, children preferentially imitated them at the expense of the means. There were no significant effects in the condition with the less salient outcomes (i.e., the bowls). The importance of outcomes has also been recognized for the perception of goal-directedness (for a review, see Elsner, 2007 ).

Together the reported findings suggest that infants and young children are more likely to focus on the outcome than the means of a given goal. Importantly, the relative salience of the means and outcomes (i.e., goal focus) also seems to depend on the material presented in the paradigms, cautioning against strong interpretations regarding the preferred goal focus in early childhood.

Research on overimitation (i.e., the imitation of action elements that are causally irrelevant for a certain outcome; e.g., Horner & Whiten, 2005; Keupp et al., 2015 ) and norm enforcement (e.g., Rakoczy et al., 2008 ) suggests that children around preschool age no longer primarily pay attention to the outcomes others achieve, but also to the means others use to achieve an outcome, even if some of the behaviors are causally irrelevant for the outcome. One major explanation for overimitation constitutes its social function. Not only does overimitation allow children to learn about norms and certain rituals, but it might also be a means to affiliate with others (for a review, see Hoehl et al., 2019 ). Additionally, children’s overimitation is context specific. Depending on the social cues (instrumental vs. conventional), children are more or less likely to overimitate and therefore the specific means seem to be more or less salient (e.g., Legare et al., 2015 ). Independent of the function of overimitation, we cannot know whether the means actually become more important than the outcomes or only more important than they used to be, because participants in these studies are always able to imitate both aspects, the means and outcomes. However, social context seems to modulate the salience of the means, which might indicate that goal focus can be influenced by the social context.

With respect to possible mechanisms explaining infants’ and young children’s goal focus, we propose that cognitive development plays a major role. After having formed representations of the outcomes of actions, infants and young children might be likely to explore the means in greater detail and, therefore, shift towards a stronger process focus, or at least focus more on the process than before. We could not identify any particular motivational process that could explain a stronger focus on the means or outcomes in childhood. As mentioned above, social motivation can explain the phenomenon of overimitation and social context might impact goal focus, but there is no reason to assume that social motivation influences goal focus in a general manner. This is not to say that there are no individual differences in motivation that might contribute to goal focus in early childhood (e.g., a strong achievement motive might be associated with a stronger outcome focus), but these individual differences are unrelated to early development. One might wonder about the development and impact of goal orientation in early childhood. To our knowledge, orientation towards gains, maintenance, or loss avoidance has not been investigated for this age group. One might speculate that children generally are more gain oriented, because they have not had the chance to accumulate many resources yet and additionally expect growth across different domains (Riediger et al., 2014) . However, because there is no reason to assume that the primary orientation towards gains might change across childhood and adolescence, goal orientation cannot provide an explanation for potential changes in goal focus for this age range.

In sum, we propose that infants and young children focus on the outcomes of goal pursuit, primarily based on the development of their (social-)cognitive competencies. Findings in different lines of research support this notion by indicating a predominant sensitivity for outcomes in different paradigms. However, differences in the salience of means and outcomes in the research paradigms also impact infants’ and children’s attention to the means versus outcomes of an action. On the basis of findings on overimitation, we hypothesize that the means gain in importance as children grow older, though not necessarily more than the outcomes and across social contexts.

Goal Focus in Later Childhood and Adolescence

Research on the development of action perception is scarce in later childhood and adolescence (for notable exceptions for research on action perception across the lifespan, see e.g., Nielsen & Tomaselli, 2010; Wermelinger et al., 2019 ). To our knowledge, there are two studies documenting that overimitation increases throughout childhood and into adulthood (e.g., Nielsen & Tomaselli, 2010 ; for a review, see Hoehl et al., 2019 ). As mentioned above, the findings on the development of overimitation suggest an increasing sensitivity for means with age which might render them more important than before, or maybe even more important than the outcomes, at least in particular contexts (e.g., when learning about a culture; Nielsen & Tomaselli, 2010 ). However, given that overimitation increases into adulthood, but young adults also focus more on the outcomes than the means of goal pursuit, one could argue that overimitation might not indicate a focus on the means versus the outcomes. Instead, it seems to imply that both, means and outcomes, are similarly salient and worth imitating if possible. This similar degree of salience does not seem to apply beyond the context of overimitation paradigms; else one should observe neither a process nor outcome focus in young adults.

There is no research directly investigating goal focus for school children and adolescents. Therefore, we turn to the development of goals in the school context as well as the development of adolescents’ goals in more general. This research provides insights into how goal focus might change in later childhood and adolescence. Given that there is only very little research on the development of goals as cognitive representations for this portion of the lifespan, we also draw on the literature on self-representations. This literature provides general insights into the development of cognitive representations that can also be applied to goals. Furthermore, we consider decision-making in adolescence, as decision-making involves cognitive and motivational processes that might also impact goal focus.

Goals Inside and Outside the School Context

The work by Eccles et al. (1993) provides an entry point for potential changes in goal focus during school age. These authors have argued that the school environment becomes increasingly grade-focused and competitive from elementary to secondary school. When grades are seen as the major outcome of learning, school-aged children might become (even) more outcome-focused as they move through school and adopt the school’s emphasis on grades. Similar findings have been reported with regard to performance and mastery goal orientation, in that students had higher levels of performance goals with the transition to middle school (Anderman & Anderman, 1999) . These findings support the notion that the school environment impacts students’ goal-related thoughts and beliefs (Anderman & Anderman, 1999) . Together with the finding that the school environment can influence identity formation (Kaplan & Flum, 2012) , the school context of an increasing emphasis on grades might also have an impact on goal focus in that it increases a focus on outcomes throughout later childhood and adolescence. More generally, the school context constitutes a potential societal influencing factor on goal focus which might also vary from culture to culture (e.g., in some cultures the schools’ focus on grades or other tangible outcomes might not increase across the years).

Regarding goal development in general, adolescence constitutes a time which is marked by the development and selection of goals that pave the way for the transition to adulthood (e.g., Erikson, 1968; Nurmi, 2013; Salmela-Aro, 2009 ). During this phase of selecting goals, adolescents need to consider the potential outcomes of different alternatives (Marcia, 1966) . From this perspective, one could argue that adolescents are in a predecisional phase as described in the model of action phases (H. Heckhausen & Gollwitzer, 1987) . In this phase, people weigh different possible outcomes against each other to decide which goals to pursue. Therefore, the outcomes should be more salient for adolescents when they reflect upon what they want to achieve in the future (Freund et al., 2019) .

In contrast to these findings that might speak for an increasing outcome focus from later childhood throughout adolescence, Nurmi et al. (1994) showed that the temporal extension of many of the landmark goals for adolescents (e.g., graduating from high school) decreases with age, as they move closer in time to their realization (e.g., graduating from high school is much closer at the age of 16 compared to 13 years). Consequently, one could also argue that adolescents’ goals turn from high-level to low-level construals and that, therefore, the means might become more salient throughout adolescence, at least for certain goals.

Development of (Self-)Representations in School-Aged Children and Adolescents

Research on the development of self-representations can provide insight on how cognitive representations of self-related characteristics and attributes develop in general. Overall, the literature suggests that self-representations develop from childhood to adolescence from more concrete to more abstract descriptions (e.g., Harter, 1999 ). In case this is a general trend in the development of representations, goals likely develop in a similar way.

Regarding the development of self-representations, Harter (1999) differentiates between three phases in childhood and adolescence, respectively. Whereas children at the age of 4 years describe themselves in terms of concrete behaviors or characteristics (e.g., “I can run fast. I have brown eyes”), children by the age of 7 years start to combine competencies in representational sets (e.g., “being good at schoolwork,” instead of “being good at reading, spelling and counting”). By age 11, children integrate their descriptions into higher-order generalizations (e.g., “being popular”). During adolescence, self-representations become more abstract (e.g., “extraverted,” “intelligent”) and initially isolated trait labels become, after a phase of confusion and conflict, more integrated and consistent. When applying this pattern of increasing abstractness to goals, different to younger children, older children and adolescents are more likely to represent their goals more abstractly, that is in terms of abstract outcomes rather than in terms of more concrete means (e.g., for thriving at school “receiving a good education” instead of “learning for the exam tomorrow”). Consequently, the increasing ability to think abstractly could render an outcome focus in later childhood and adolescence more likely than a focus on the means.

Furthermore, with increasing age children improve in their ability to represent future events (at least the ones that lie in the further future) and orient their thoughts and actions towards them (for an overview, see Nurmi, 2005 ). Together with the core tenet of construal level theory (Trope & Liberman, 2003) , that more distant events are represented more abstractly, this speaks for an increasing outcome focus from childhood to adolescence.

Decision-Making in Adolescence

Some research on planning and decision-making suggests that adolescence is a phase of increased risk taking when the (even potentially harmful) longer-term consequences of actions are often not taken into account (e.g., Steinberg, 2007 ). Moreover, young adolescents are less future oriented than older adolescents and adults in that they plan less ahead and are worse in anticipating consequences of certain actions (e.g., Steinberg et al., 2009 ). Together with the assumption of construal level theory that near future actions are represented in terms of concrete details (Trope & Liberman, 2003) , the finding that adolescents plan less ahead should lead them to concentrate on near future events and hence focus more on the “how” aspects of actions in the “here and now” (i.e., focus on the means) than on outcomes that often lie in the future, at least with regard to more complex goals. At first glance, this seems to contrast the aforementioned increasing ability to focus on abstract outcomes. However, young adolescents are particularly unlikely to plan ahead and anticipate consequences compared to older adolescents and young adults (Steinberg et al., 2009) . Therefore, adolescents’ focusing on the means is likely to decrease as they grow older, which then fits with the assumption of an age-related increase of the focus on outcomes through adolescence. Consequently, this line of argumentation suggests a change from focusing on the means in early adolescence towards an outcome focus in later adolescence.

Apart from this “cognitive inability” perspective on adolescents’ risk-taking and decision-making, motivational changes in adolescence might also contribute to a stronger focus on the means compared to outcomes in this phase of life: Blakemore (2018) argues that the risk-taking observed in adolescence might be less due to the cognitive inability to foresee potentially harmful consequences of one’s actions, but rather be driven by a strong orientation towards social goals related to peers. If, as Blakemore suggests, the immediate social rewards of doing something with one’s peers are more important than potential later outcomes, this should increase a focus on the means over an outcome focus. However, this depends on what is regarded as the means and the outcome. If “having fun with peers” is perceived as an outcome, Blakemore’s suggestion does not have any implications for goal focus. Further, it remains unclear when and how exactly this pattern might change in the course of adolescence.

Taken together, different patterns of the trajectory of goal focus in adolescence are theoretically plausible: Most theories on adolescents’ development presented here support the notion of an early shift in that older children and adolescents become gradually more outcome-focused (i.e., through an increasing focus on grades in school, the need to select future goals, the increasing ability to form more abstract representations, to orient towards future events, and the increasing ability and motivation to take long-term consequences into account). However, there are also reasons to assume a later shift in goal focus: Changes in the temporal extension of goals might suggest that adolescents focus more on the means as they grow older, nearing the realization of many of their goals. If this were the case, the question would be when and why goal focus shifts back to a stronger outcome focus in young adults. It is currently an open question whether the assumed shift from process focus to outcome focus already takes place in later childhood, over the course of adolescence, or only when adolescents transition into young adulthood. As to the involved mechanisms, we assume that both, cognitive and motivational processes, as well as societal influences might be at play.

Based on the literature on the development of action perception in childhood, self-representations and other processes in adolescence, and motivational changes reflected in goals as cognitive representations across adulthood, we assume that goals change across the lifespan on multiple dimensions, such as complexity, temporal extension, and motivational orientation: As children grow older, they become increasingly able to represent complex and temporally distant goals due to cognitive development. During adulthood, the availability of resources changes and remaining lifetime decreases, which influence the motivational orientation of adults’ goals. Based on these cognitive and motivational changes, we propose that lifespan changes in goal focus are mainly due to cognitive development in the early phases of the lifespan and motivational development across adulthood. Whereas Figure 1 depicts more specific cognitive and motivational processes that might impact goal focus, Figure 2 provides a more specific overview on how goal focus might change across the lifespan.

Note. The vertical arrow indicates that it is unclear whether processes weigh more than outcomes in this age range or only more relative to infancy and early childhood. The dashed versus dotted line and horizontal arrow indicate that both an earlier and a later shift towards outcome focus are possible with the specific timing of the change being unclear.

Note. The vertical arrow indicates that it is unclear whether processes weigh more than outcomes in this age range or only more relative to infancy and early childhood. The dashed versus dotted line and horizontal arrow indicate that both an earlier and a later shift towards outcome focus are possible with the specific timing of the change being unclear.

More specifically, we suggest that in early development, when children learn about actions and their effects in the world, the outcomes are likely more important than the means. This idea has its roots in the ideomotor theory (Greenwald, 1970; Shin et al., 2010) and is supported by findings that infants are more likely to perceive an action as goal-directed when it has a salient outcome (for a review, see Elsner, 2007 ). As children grow older and have developed stable action-outcome associations, they likely focus increasingly on the specific means by which an outcome can be achieved, as indicated by overimitation and normative criticism (e.g., Rakoczy et al., 2008 ). However, it is unclear if the means increase in salience with age but might still be less salient than the outcome, or if the means actually become more salient than the outcome (as indicated by the vertical arrow in Figure 2 ). Further, the salience of the means seems to be context specific (Legare et al., 2015) . Regarding broader developmental processes, we assume that cognitive development plays a major role for changes in goal focus in infancy and early childhood, whereas motivation plays a less important role (see Figure 2 ). This is not to say that motivational or social influences are irrelevant for children’s behavior in the first years of life. However, there are no theoretical reasons or empirical evidence leading us to assume that motivational or social processes lead to changes in goal focus.

As for later childhood and adolescence, we propose a shift from process focus to outcome focus, with different possible temporal patterns:

There might be an early shift towards a stronger outcome focus starting in later childhood (see Figure 2 , dashed line). Cognitive development enables children to form more abstract representations (Harter, 1999) . As outcomes are usually more abstract than means, the development of this ability might lead to an increase in outcome focus starting in later childhood throughout adolescence. Furthermore, children and adolescents become increasingly able to orient their thoughts and actions towards the future and might therefore focus more on temporally distant outcomes (Nurmi, 2005) . Over the course of primary and middle school, grades gain in importance and competition increases (Eccles et al., 1993) . This likely renders outcomes more salient than the means of goal pursuit. Additionally, research on decision-making and planning (e.g., Steinberg et al., 2009 ) suggests a greater focus on the near future and therefore on the temporally closer means rather than the more distant outcome for younger adolescents. As they grow older, adolescents become more likely to take long-term consequences into account, which might focus their attention more on the outcomes of goal pursuit. Finally, adolescents find themselves in a phase of selecting future goals (Nurmi, 2005, 2013) , where evaluating and comparing different outcomes might render them more salient than the means.

Alternatively, as adolescents grow older, they also move closer towards the realization of landmark goals. Thus, older adolescents might construe them on lower levels and focus more on the means over time. Apart from this, changes in goal content might affect goal focus in adolescence: As social goals gain in importance in adolescence (Blakemore, 2018) , the very pursuit of goals (i.e., engaging in certain means) together with peers might become more important than its consequences, again suggesting a higher focus on the means than outcomes in adolescence. This would speak for a prolonged phase of focusing on the means throughout adolescence, with a shift towards outcome focus only occurring in young adulthood (as indicated by the dotted line in Figure 2 ).

Taken together, we assume that in later childhood and adolescence, cognitive as well as motivational processes impact the development of goal focus, which makes it difficult to formulate directed predictions. On the one hand, cognitive processes such as abstract thinking slowly reach adult levels, allowing for a broad range of representations, from very concrete to very abstract. This should also give more leeway to how goals are represented. On the other hand, motivational influences, such as the setting and pursuit of long-term goals, gradually gain in importance. These cognitive and motivational processes probably do not act independently from one another. For instance, a certain degree of ability to represent the future is needed to set and pursue longer-term goals. On top of these processes, social and societal influences likely impact what kinds of goals people pursue at different points in the lifespan, and whether they focus more on the means or outcomes. From a theoretical standpoint, both, an early and a late shift from process to outcome focus seem plausible (see Figure 2 , dashed and dotted line). How fast and when exactly this shift from process to outcome focus might take place, remains unclear (indicated by the horizontal arrow in Figure 2 ) and needs empirical investigation.

In adulthood, we propose that relative to older adults, younger adults focus more on the outcomes, because of their goal orientation towards gains as well as their extended future time perspective. These assumed age-related differences in goal focus have been supported empirically (e.g., Freund et al., 2010; Mustafić & Freund, 2012b ). These motivational changes might in part be influenced by societal factors such as expectations of when certain developmental tasks should be completed. We do not assume that cognitive processes have a major impact on goal focus across adulthood because 1) the level of cognitive abilities needed to represent goals of varying abstractness and temporal scope should be reached in the course of adolescence and 2) cognitive decline in later adulthood should not impact goal focus, as the crystallized abilities relevant for goal focus typically remain fairly stable across adulthood.

In sum, we propose that changes in goal focus across the lifespan are driven by two different processes: Because action perception and the development of (self-)representations in childhood are largely characterized by cognitive development, we assume that cognitive processes play a major role for goal focus and its development in that age range (e.g., establishing action-outcome associations; being able to form abstract representations). In later adolescence and adulthood, where cognitive abilities no longer act as constraints, we predict motivational processes to drive changes in goal focus (e.g., changes in goal orientation). In addition, it is possible that societal influences, such as social expectations, impact people’s goal focus (as mentioned in the school context), even though it is hard to make specific age-related predictions in that case.

The literature on goals integrated in this paper stems from different psychological perspectives (e.g., motivation, action perception) and consequently includes a variety of goals, which do not only differ with regard to their content but also their temporal scope, abstractness, and complexity. Importantly, these differences in temporal scope, abstractness and complexity are largely confounded with age, as infants and young children are not yet able to represent (or at least communicate) more complex goals, which often span into the distant future. One of the factors contributing to these age differences might be the development of the comprehension of time. Without an understanding of what “tomorrow” or “next week” mean, the representation of complex, temporally distant goals seems difficult if not impossible. As a consequence of the confounding of age with the mentioned concepts, we cannot disentangle whether age-related differences in goal focus might actually be related to the complexity of the respective goal.

Apart from the variety of goals that have been investigated in the studies reported above, the studies also differ regarding their methodology and the situations under investigation. These differences reflect on the one hand the large differences in participants’ age and, on the other hand, also the different research traditions. The differences between studies make it difficult to compare and integrate the results into an overarching lifespan account of goal focus. As we have shown in the action perception part in infancy and childhood, even seemingly small changes in setup can have large effects on the results. Given the lack of research directly investigating goal focus in childhood and adolescence, we considered the extant literature that provides a basis for understanding the development of goal focus. Finally, as there are no longitudinal studies on goal focus, we had to rely on cross-sectional data with its obvious drawback that differences between age-groups do not necessarily reflect development over time (Baltes et al., 1977) .

As mentioned throughout the paper, studies explicitly investigating goal focus are scarce and mainly limited to adult development. Thus, we considered diverse literatures which include a great variety of goals to draw conclusions on the development of goal focus. Obviously, this theoretical approach would benefit from an empirical complement with studies that are explicitly designed to investigate goal focus and its influencing factors across the lifespan. To this end, multiple steps are needed. The first important step is to engage in conceptual work by providing a definition of goals and goal focus that can be used for a broad range of goals and across age groups (as we have attempted in this paper). Then, suitable measures with appropriate goals need to be developed. Furthermore, dimensions on which goals differ need to be identified (such as time extension and goal complexity) to systematically investigate their impact on goal focus across and within different age groups. Finally, relevant developmental processes (such as the formation of abstract representations) need to be assessed along with goal focus. Ideally, such an empirical approach would include cross-sectional as well as longitudinal comparisons to answer the following questions: How does goal focus change across the lifespan? How much do these changes vary individually? Which developmental processes contribute to these changes? Which goal dimensions impact goal focus (and how goal specific is goal focus)? Which consequences does goal focus have for development?

On a more general note, we want to stress the importance of research spanning the entire lifespan, including infancy and childhood as well as adulthood and aging. Looking at the different parts of the lifespan separately not only keeps developmental psychologists from seeing the bigger picture of development, but also from gaining valuable conceptual and methodological insights from research in other parts of the lifespan. Therefore, we encourage developmental psychologists to conduct, or at least consider on a theoretical level, lifespan research that actually spans the entire life (in addition to cultural and species comparisons).

The goal of this paper was to combine different approaches and research findings on the development of goal focus across the entire lifespan. Goal focus is particularly important to consider across the lifespan because it predicts the success of goal pursuit and well-being. Thus, it likely contributes to successful development. The literature points to the importance of considering different factors impacting developmental change across the lifespan (Baltes et al., 1977) but also to the need to engage in conceptual work before comparing different literatures. We acknowledge these theoretical and empirical challenges of considering the entire lifespan for understanding age-related changes in goal focus. However, we believe that research including the entire lifespan is worthwhile in that it contributes to a fuller understanding of human development than considering the different life phases separately: Neither does development in adulthood take place without the preceding influences of childhood and adolescence, nor does childhood development take place without the prospect of adult development.

Drafted and/or revised the article: LM, MMD, AMF

Approved the submitted version for publication: LM, MMD, AMF

During the work on her dissertation, LM was a pre-doctoral fellow of LIFE (International Max Planck Research School on the Life Course; participating institutions: Max Planck Institute for Human Development, Freie Universitaet Berlin, Humboldt-Universitaet zu Berlin, University of Michigan, University of Virginia, University of Zurich). We are grateful to the members of the Child Development Team and the Life-Management Team at the University of Zurich for fruitful discussions of the ideas presented in this article.

Throughout her dissertation, LM was funded by the Jacobs Foundation (F-63206-18-01).

The authors have no conflicts of interest to declare.

This is a theoretical contribution. Therefore, no databases have been used.

In keeping with the tradition of action theory in philosophy (e.g., Davidson, 1963 ), we reserve the term “action” to denote intentional behavior. In case we cannot assume intentionality, we will use the term “movement” for motor behavior, and the term “behavior” when we want to include also non-motor external reactions.

Note that sometimes means and outcome are hardly distinct, such as in the goal “I want to enjoy running,” which Kruglanski et al. (2018) have called means-ends fusion and conceptualized as intrinsic motivation.

With resources we refer to goal-relevant means that are limited at any given point in time and, once spent, no longer available until replenished (such as money, time, or social support; Freund & Riediger, 2001 ).

Pathological cognitive aging such as dementias, however, likely affect the representation of time and counterfactuals, and might therefore also impact goal focus.

When speaking about the different age groups, we refer to the following broad categorization: Infancy (up to 1 year), early childhood (comprising toddlerhood: 1-3 years, and preschoolers: 3-5 years), later childhood/ school-aged children (6-10 years), adolescence (early: 11-14 years, late: 15-17 years), young adulthood (18-29 years), middle adulthood (30-64 years), early older adulthood (65-84 years), and late older adulthood (85 years and older).

With imitation we refer to all variants of reproducing behavior, independent of whether the imitator understands the intention of the actor.

Supplementary data

Recipient(s) will receive an email with a link to 'For Whom Is the Path the Goal? A Lifespan Perspective on the Development of Goal Focus' and will not need an account to access the content.

Subject: For Whom Is the Path the Goal? A Lifespan Perspective on the Development of Goal Focus

(Optional message may have a maximum of 1000 characters.)

Citing articles via

Email alerts, affiliations.

Stay Informed


About UC Press

This Feature Is Available To Subscribers Only

Sign In or Create an Account


PSYC 322 Lifespan Developmental Psychology

Online reference books, useful books, online books, picking a topic.

Psychology in the News

Profile Photo

empirical research article on lifespan development

Useful Websites that talk about Psychology in the News


www. psychologytoday .com/


Explore. Discover. Create.

Copyright ©  2022  Pepperdine University


  1. (PDF) Using Empirical Article Analysis to Assess Research Methods Courses

    empirical research article on lifespan development

  2. How do I know if a research article is empirical?

    empirical research article on lifespan development

  3. (PDF) General Analysis of an Empirical Research Article (Doctoral Assignment)

    empirical research article on lifespan development

  4. Development Economics Theory Empirical Research and Policy Analysis 1st Edition Schaffner

    empirical research article on lifespan development

  5. How do I...?

    empirical research article on lifespan development

  6. What is an Empirical Article?

    empirical research article on lifespan development


  1. Lifespan development presentation

  2. Development: 2.2 Research Methods Ethics

  3. Chapter 1 Introduction to Lifespan Development


  5. The Story of Participatory Research: History & Future Webinar

  6. METASCIENCE 2021 9 25 Preregistration in the Social Sciences Empirical Evidence of its Effectivenes


  1. Child health and human development over the lifespan

    A study of health, development, and well-being over the lifespan. ... but human development is a life span issue, so research in childhood

  2. A Motivational Theory of Life-Span Development

    The past 15 years of conceptual and empirical work have shown that a central feature of adaptive capacity is the regulation of motivation. An individual's

  3. Considering Generations From a Lifespan Developmental Perspective

    For example, Baltes and Reinert (1969) note: “Schaie (1965) formulated a well-explicated general model for the study of developmental problems which is based on

  4. Connecting Life Span Development with the Sociology of the Life

    Attempts to link this sub-discipline with life span developmental psychology have been called for but with little sign of success. In this paper

  5. Spelling out some unaddressed conceptual and methodological

    Nevertheless, theoretical or empirical research that actually encompasses the entire lifespan, that is, from early childhood to old age, is rare. This is not

  6. Development and Health of Adults Formerly Placed in Infant Care

    This study uses data from a survey on the health and development of 420 children mostly under the age of three, placed in 12 infant care

  7. Learning as an Important Privilege: A Life Span Perspective with

    We conclude that addressing barriers to lifelong learning would advance theories on life span cognitive development and raise the bar for

  8. For Whom Is the Path the Goal? A Lifespan Perspective on the

    For the adult segment of the lifespan, we review the theoretical and empirical work on developmental changes in goal focus across adulthood.


    Lifespan Development: A Psychological Perspective ... development is such a vast topic of study that it requires the theories, ... empirical support.

  10. Finding Articles

    Includes all journal articles, book reviews, letters to the editor ... To find articles that report on empirical research, some keywords you