Possible Selves

Possible selves are specific imagined future identities — the people one hopes to become, expects to become, or fears becoming. The framework was introduced by Hazel Markus and Paula Nurius in their 1986 American Psychologist paper "Possible Selves" as a cognitive-motivational construct linking self-concept to behavior across time.

The framework operates at the level of specific imagined identities rather than general future orientation. A person does not have one "future self" but a repertoire of possible selves: the professional one might become, the parent, the patient, the recovered, the unemployed. Each possible self carries cognitive content (specific imagery, traits, situations) and motivational valence (positive, negative, mixed). Markus and Nurius proposed that this repertoire functions as a cognitive-motivational substrate: possible selves connect the abstract self-concept to concrete behavior by providing specific representations of where current action leads.

The framework distinguishes three principal categories: hoped-for selves (positive aspirational identities), expected selves (what one realistically anticipates), and feared selves (negative identities one wants to avoid). The categories can overlap or diverge in interesting ways; a hoped-for self that is far from an expected self produces a discrepancy that the literature has connected to anxiety and motivation in different proportions depending on context. The contemporary measurement and research literature (Oyserman et al., from the early 2000s onward) has elaborated on these categories and added balanced versus imbalanced repertoires as a major predictor of behavioral outcomes.

In clinical and applied psychology, possible-selves work has been used in motivational interviewing, addiction treatment, eating-disorder interventions, and identity-focused therapy. The construct provides a framework for working with future-oriented motivation that respects the multiplicity of identity rather than asking patients to commit to a single aspirational ideal. Feared possible selves often do more motivational work than hoped-for ones in clinical contexts — the literature is consistent that fear of becoming a specific negative future identity often produces more behavior change than aspiration toward a positive one.

In research on motivation and self-regulation, possible selves connect self-concept research (Markus's broader research program on self-schemas (synthesized in Markus & Wurf 1987, Annual Review of Psychology)) to behavioral outcomes. Daphna Oyserman and colleagues have shown through multiple studies that balanced possible selves — configurations with both hoped-for and feared selves in the same domain (e.g., "successful student" hoped-for + "academic failure" feared) — predict better academic outcomes than imbalanced repertoires. The mechanism: balanced repertoires support both approach (toward hoped-for) and avoidance (away from feared) motivation simultaneously, which is more durable than approach motivation alone.

In research on identity development, the framework has been applied across the lifespan from adolescence through older adulthood. Possible selves shift with age (younger adults have more hoped-for selves, older adults have more health-related feared selves) and with major life transitions. The framework has informed identity-based intervention design in education (school-to-college transition programs) and career development (career-pivot decisions, retirement planning). The conceptual link to future self continuity is that FSC measures the binding-strength of any possible self to present identity, while the possible-selves framework measures the content and distribution of specific imagined identities.

The construct was introduced by Hazel Markus (then at the University of Michigan, later at Stanford) and Paula Nurius in their 1986 American Psychologist paper. Markus's prior work on self-schemas (1977, Journal of Personality and Social Psychology) had established that self-concept is organized as multiple structured representations rather than a single global identity. The 1986 paper extended this multiplicity to the temporal dimension: in addition to the multiple selves a person currently is, there are multiple selves a person could become, and these future-oriented representations are themselves a structured part of the self-concept.

Markus and Nurius drew on older traditions for the basic intuition. William James's distinction between "I" and "me" in The Principles of Psychology (1890) had identified the self as a structured object of cognition. Carl Rogers's work on the "ideal self" provided a single-aspirational-self framework that Markus and Nurius generalized to a repertoire. Hazel Markus and Ann Ruvolo's 1989 chapter "Possible Selves: Personalized Representations of Goals" (in Pervin, ed., Goal Concepts in Personality and Social Psychology, Erlbaum) extended the framework specifically to goal-directed behavior, providing the bridge to motivation research.

The empirical research program expanded through the late 1980s and 1990s. Daphna Oyserman's work on possible selves in adolescent academic motivation (multiple papers from 1990 onward) established the balanced-repertoire finding that became central to the contemporary literature. Susan Cross and Hazel Markus (1991, Human Development) extended the framework across the lifespan. Robins and Trzesniewski later situated possible selves within broader self-concept measurement work. By the 2000s, possible selves had become one of the standard frameworks for thinking about future-oriented motivation in personality and social psychology, with a substantial intervention literature in education and clinical contexts.

The possible-selves framework operates at the level of specific imagined identities organized by motivational valence. The canonical three-part structure distinguishes hoped-for, expected, and feared possible selves.

Hoped-for possible selves are positive imagined identities the person aspires to become: the successful professional, the loving parent, the accomplished artist. These selves provide direction for approach motivation. A hoped-for self contains specific content (traits, situations, achievements) rather than abstract values. The vividness and detail of the representation matters: a clear, specific hoped-for self generates more motivation than a vague aspiration. Episodic future thinking is the cognitive capacity that produces this specific imagery.

Expected possible selves are realistic anticipations — the person one expects to be regardless of preference. These can match or diverge from hoped-for selves. A large discrepancy between hoped-for and expected can produce either healthy motivation (if the gap feels closeable) or chronic dissatisfaction (if it does not). The discrepancy literature within possible-selves research connects to broader self-discrepancy theory (Higgins 1987).

Feared possible selves are negative imagined identities to be avoided: the addict, the failed, the isolated, the unemployed. The literature has consistently found that feared selves do at least as much motivational work as hoped-for selves, sometimes more. The mechanism is avoidance motivation: vivid representation of a feared future identity generates behavior aimed at not becoming that person. Clinical interventions for substance use, for example, often work better when they engage feared possible selves than when they only build hoped-for selves.

The balanced-repertoires finding (Oyserman and colleagues) is one of the most important empirical results in the contemporary literature. Balanced repertoires — configurations with both hoped-for and feared selves in the same domain (e.g., academic) — predict better behavioral outcomes than imbalanced ones (all hoped-for or all feared). The mechanism is that approach and avoidance motivation work together in balanced repertoires, providing both direction and urgency. Educational interventions designed to produce balanced academic possible selves have documented effects on study behavior and achievement, though effect sizes vary substantially across studies and populations.

Possible-selves measurement varies more than measurement of future self continuity or delay discounting. There is no single dominant validated questionnaire; the construct's specificity (it asks about which imagined identities, not how strongly an identity is felt) requires open-ended or content-coded approaches.

The original Markus and Nurius (1986) approach uses an open-ended generation task. Participants list specific possible selves under each category prompt (hoped-for, expected, feared). The lists are content-coded for domain (academic, relational, occupational, health, material) and for specific content within domain. This approach produces rich data but is labor-intensive to score.

Domain-specific measurement uses closed-ended Likert items targeting specific domains. For example, the academic possible selves scale used in much of Oyserman's work asks participants to rate the strength of hoped-for and feared academic selves on numeric scales. This approach is faster but loses the specificity of content. The trade-off between rich open-ended content and rapid quantitative measurement is unresolved in the field.

The LBL Future Self Continuity Index does not measure possible-selves content directly. The FSCI measures the felt continuity between present and future self regardless of which specific future self is being considered. Possible-selves measurement and FSCI measurement are complementary: a person can have rich and balanced possible-selves repertoires (clear content) but low FSC (those imagined selves feel like different people), or sparse possible-selves repertoires but high FSC (vague future identity but strongly felt as oneself). Researchers needing possible-selves content should use open-ended or domain-specific instruments; the FSCI is appropriate for the identity-binding question.

The framework is one of several future-oriented self-and-identity constructs. The distinctions matter for both theory and intervention design.

vs. future self continuity — possible selves are specific imagined future identities (content); future self continuity is the felt binding-strength of any future self to present identity (relation). The two are partially independent: a person can have rich possible selves with low FSC (vivid content that feels like different people) or sparse possible selves with high FSC (vague content felt strongly as oneself).

vs. episodic future thinking — EFT is the cognitive capacity to simulate specific future events at all. Possible selves are specific imagined future identities. The capacities are related — vivid EFT supports rich possible-selves representations — but distinct. A person with intact EFT may still have impoverished possible-selves repertoires if they have not engaged the framework, and vice versa.

vs. self-discrepancy theory — Higgins's (1987) self-discrepancy theory distinguishes actual, ideal, and ought selves, with discrepancies producing specific emotional consequences (sadness/dejection for actual-vs-ideal; agitation/anxiety for actual-vs-ought). Possible-selves theory is broader: the hoped-for self is roughly the Higgins ideal self, but the framework also includes feared selves (no direct Higgins analog) and expected selves (between ideal and ought). Self-discrepancy theory is more parsimonious; possible-selves is more comprehensive.

vs. identity-based motivation theory — Oyserman's (2007) identity-based motivation theory builds on possible selves to argue that identity-congruent behaviors feel easier and identity-incongruent behaviors feel harder. The theory uses possible-selves content as input and predicts behavior as output through the felt difficulty of action. Possible-selves is the underlying content framework; identity-based motivation is one specific theory of how that content affects behavior.

vs. growth mindset / mindset research — growth mindset (Dweck) is about beliefs about the malleability of traits and abilities. Possible selves are about specific future identity content. The two interact: a growth-mindset person may have more elaborated hoped-for selves because they believe change is possible.

A student's study decision

A college sophomore deciding whether to study for tomorrow's chemistry midterm or watch a streaming series. They imagine themselves in three years applying to medical school with a strong GPA; they also imagine themselves in three years applying with a weak GPA and being rejected. They study.

The possible-selves reading: this is a balanced-repertoire engagement. Both a hoped-for self (medical-school applicant with strong GPA) and a feared self (rejected applicant) are present and specific. Oyserman and colleagues have documented that this balanced configuration predicts study behavior more reliably than either configuration alone. A student with only the hoped-for self might procrastinate (because the future feels distant); a student with only the feared self might avoid study entirely (because the feared outcome feels too aversive to engage). The balanced repertoire produces both direction and urgency.

A career-pivot consideration

A 38-year-old considers whether to leave a stable job for a more demanding but potentially more meaningful role. They list specific possible selves in each direction: the version of themselves who stays and grows comfortable; the version who leaves and thrives; the version who leaves and fails; the version who stays and grows resentful. Two are hoped-for, two are feared. They decide based partly on which feared self felt more vividly avoidable.

The possible-selves reading: career decisions often engage richer possible-selves repertoires than smaller daily decisions because the stakes invite elaborated future-self construction. The Markus and Nurius (1986) framework predicts that the relative vividness and motivational pull of the feared selves often matters more than the hoped-for ones in high-stakes pivot decisions. The literature on regret anticipation in career decisions (Gilovich & Medvec 1995) connects to this: vivid feared possible selves are essentially anticipated regret made concrete in identity terms.

The possible-selves framework has been influential for nearly four decades and has accumulated both replicated findings and methodological critiques.

Measurement heterogeneity. Unlike delay discounting or future self continuity, possible-selves research uses many different measurement approaches: open-ended content generation, closed-ended Likert items, domain-specific scales, qualitative interviews. Cross-study comparison is complicated because the constructs being measured are not standardized. The trade-off between rich open-ended data and quantitative tractability is unresolved.

Popular-framing inflation. Popular self-help adaptations of the framework often emphasize hoped-for self visualization while ignoring the framework's evidence about feared selves and balanced repertoires. "Visualize your best self" is popular advice, but the empirical literature consistently shows that hoped-for-only configurations produce less behavior change than balanced ones. The popular framing systematically understates this finding.

Cultural variance. Like much of contemporary social and personality psychology, the possible-selves literature has been built predominantly on WEIRD samples. The framework's emphasis on individual future self-construction may not generalize directly to cultural contexts with stronger interdependent self-construals, where possible selves may be constructed more collectively (family futures, community futures) rather than individually. Some adaptation work has been done but cross-cultural validation remains incomplete.

Intervention effect-size variability. Possible-selves-based interventions in education have documented effects on academic outcomes. Oyserman, Bybee and Terry (2006, Journal of Personality and Social Psychology) tested the School-to-Jobs intervention in a randomized study of n=141 experimental and n=123 control low-income 8th graders, with effects on academic initiative, GPA, and behavior sustained through two years of follow-up. Effect sizes vary substantially across samples and contexts. Meta-analytic synthesis is complicated by measurement heterogeneity. The honest reading is that balanced possible-selves repertoires predict outcomes correlationally and that interventions can shift repertoires, but the practical effect sizes are modest and require sustained implementation rather than brief manipulation.

Content versus process question. The framework specifies that possible-selves content matters but is less clear about how the content produces behavior. Multiple mediating processes have been proposed (identity-congruent feels-easy mechanism in identity-based motivation theory; approach-avoidance balance in balanced-repertoires work; felt-difficulty mechanism in Oyserman's later work). The proliferation of mechanism candidates reflects the framework's flexibility but also a relative under-specification of its predictions.

The LBL Future Self Continuity Index measures the felt continuity between present and future self — complementary to possible-selves measurement, which captures the content and distribution of specific imagined future identities. Possible selves and FSC are partially independent: a person can have rich, balanced possible-selves repertoires (clear content) but low FSC (those imagined selves feel like different people), or sparse repertoires but high FSC. Researchers needing possible-selves content data should use open-ended generation tasks or domain-specific scales; the FSCI is appropriate for the identity-binding question and connects directly to the inter-temporal-choice literature where FSC is the established mediator.

Possible selves are specific imagined future identities — the people one hopes to become, expects to become, or fears becoming. The framework was introduced by Hazel Markus and Paula Nurius in their 1986 American Psychologist paper as a cognitive-motivational construct linking self-concept to behavior. The framework distinguishes hoped-for, expected, and feared possible selves. Hazel Markus and Ann Ruvolo's 1989 chapter extended the framework specifically to goal-directed behavior. Daphna Oyserman's subsequent work established the balanced-repertoires finding: configurations with both hoped-for and feared selves in the same domain predict better behavioral outcomes than imbalanced ones, with both approach and avoidance motivation working together. The mechanism is grounded in identity-based motivation theory: identity-congruent behaviors feel easier and identity-incongruent behaviors feel harder. The framework is distinct from but complementary to future self continuity (which measures the felt binding-strength of future selves), episodic future thinking (which is the cognitive capacity to simulate future events), and self-discrepancy theory (which is more parsimonious but less comprehensive). Honest limitations include measurement heterogeneity (no dominant validated questionnaire), popular-framing inflation (hoped-for-only configurations are over-emphasized in self-help adaptations), WEIRD-sample dominance, and variable intervention effect sizes. Intertemporal choice contexts — retirement saving, career pivots, health-behavior decisions — engage possible-selves repertoires when the decisions matter enough to invite elaborated future-self construction.

This entry is intended as a citable scholarly reference. Choose the format that matches your context. The retrieval date should reflect when you accessed the page, which may differ from the entry's last-reviewed date shown above.

LifeByLogic. (2026). Bounded Rationality: Simon, Satisficing, Heuristics. https://lifebylogic.com/glossary/bounded-rationality/

LifeByLogic. "Possible Selves: Markus, Nurius, Oyserman." LifeByLogic, May 18 2026, https://lifebylogic.com/glossary/possible-selves/.

LifeByLogic. 2026. "Possible Selves: Markus, Nurius, Oyserman." May 18. https://lifebylogic.com/glossary/possible-selves/.

@misc{lblboundedrationality2026,
  author = {{LifeByLogic}},
  title = {Bounded Rationality: Simon, Satisficing, Heuristics},
  year = {2026},
  month = {may},
  publisher = {LifeByLogic},
  url = {https://lifebylogic.com/glossary/bounded-rationality/},
  note = {Accessed: 2026-05-14}
}