RESEARCH DOSSIER
Goal Setting, Expectancies, and Self-Efficacy
Abstract. Goal setting, expectancy judgments, and self-efficacy organize prospective behavior through desired performance standards, anticipated success, perceived capability, and task valuation. This fundamental research monograph critically synthesizes foundational theories, laboratory and field experiments, quantitative reviews, longitudinal tests, psychometric studies, and computational models. It separates goal specificity and difficulty from commitment, outcome expectancy from efficacy expectancy, and subjective task value from opportunity and effort costs. Primary findings on goal-setting interventions, learning versus performance goals, progress monitoring, utility-value interventions, and reciprocal efficacy–performance dynamics establish domain-dependent mechanisms and important causal qualifications. The analysis evaluates measurement validity, longitudinal directionality, generalizability, ethical consequences, and applications across education, organizations, health, and human–artificial collaboration. The resulting synthesis identifies conditions under which goals help regulate action, explains why rising confidence may be an effect of prior achievement, and specifies open questions about competing objectives and the distribution of goal regulation in the Artificial Era.
Keywords: goal setting; goal difficulty; goal specificity; expectancy for success; outcome expectancy; self-efficacy; subjective task value; perceived cost; goal commitment; goal performance; situated expectancy-value theory; learning goals; progress monitoring; reciprocal causality; Artificial Era
Scientific Overview and Research Questions
Goal setting, expectancy judgments, and self-efficacy jointly concern the organization of prospective action. A goal specifies a desired standard or end state; expectancy concerns whether a consequential outcome is anticipated; self-efficacy concerns the actor's perceived capability to perform the actions required by particular circumstances. These concepts address different explanatory questions. Goals orient attention toward criteria of success, expectancies influence which alternatives appear viable, and efficacy beliefs may affect whether a person initiates or sustains the selected activity. Their relationship unfolds through feedback, prior experience, opportunity, and the changing costs of action. A scientific account must therefore explain selection, task engagement, performance, and revision as separate outcomes, rather than equating a declared intention with actual achievement. Locke (1968) Bandura (1977) Wigfield & Eccles (2000)
This research dossier examines the experimentally testable structure of those relationships: How do the difficulty and specificity of a goal influence performance? Under what conditions does commitment translate an assigned target into effective action? How do success expectations and subjective task values shape choices over time? What distinguishes task-specific self-efficacy from optimism, self-esteem, general confidence, and outcome expectancy? Does efficacy precede performance causally, or largely register feedback from previous performance? How should competing goals and the objective availability of resources enter a model of pursuit? Answers require triangulation across foundational theory, laboratory experiments, field interventions, longitudinal designs, quantitative reviews, and competing dynamic models. Locke & Latham (2002) Sitzmann & Yeo (2013) Vancouver et al. (2010)
The methodology of this monograph is critical narrative synthesis, not a protocol-driven systematic review. Original experimental and theoretical studies provide mechanistic claims; published systematic reviews and meta-analyses inform estimates of typical intervention effects; scholarly critiques identify alternatives and possible unintended effects. A source's available abstract can establish its design, reported findings, and bibliographic record, but does not establish every unreported procedure or secondary analysis. Throughout, performance means an actual task outcome, effort means allocated resources or observable engagement, persistence means continued participation under specified conditions, and goal attainment means meeting an explicit criterion. These operational choices prevent the same motivational language from obscuring different dependent variables. Epton et al. (2017) Harkin et al. (2016)
Definitions and Construct Boundaries
Goal content refers to the desired outcome; goal level to its quantitative or qualitative difficulty; specificity to precision of the criterion; proximity to the distance in time before evaluation; commitment to the willingness to pursue an accepted goal; and feedback to information about current performance relative to that criterion. A learning goal emphasizes the acquisition of knowledge or strategies, whereas a performance goal prioritizes meeting an output standard. These distinctions matter because a task that requires discovery of an unfamiliar strategy differs from a task whose successful procedure is already mastered. Difficult goals may concentrate effort under stable conditions while creating unproductive strain where the principal need is learning. The theoretical task is to specify which mechanism, outcome and task phase a given manipulation actually alters. Locke & Latham (2002) Seijts et al. (2004)
Expectancy for success is a prospective judgment about likely success in a specified activity. Outcome expectancy concerns whether a given action will lead to a particular consequence; efficacy expectancy concerns whether the actor can organize and execute the behavior. A student may expect that solving a problem earns a good grade and simultaneously doubt the capacity to solve it. Subjective task value includes intrinsic value, attainment value, utility value, and perceived cost; perceived cost can include effort requirements, emotional strain, and foregone alternatives. Expectancy and value can be correlated because past achievements, social messages, available resources, and personal identification shape both. A product of expectancy and value is a useful formal intuition, but no universally validated single equation summarizes every motivational choice. Bandura (1977) Eccles & Wigfield (2002) Flake et al. (2015)
Self-efficacy is a capability judgment bound to a course of action, a domain and a level of challenge. It is conceptually distinct from general self-esteem, which concerns broader evaluation of oneself, and from an observer's independent assessment of objective competence. People may overestimate capability in an unfamiliar domain or underestimate capability after discouraging feedback; either discrepancy makes the belief important to study but prevents inferring actual skill from a questionnaire alone. Generalized efficacy inventories have applications for between-person comparison, but task-matched measurements are often necessary for process-level causal tests. The construct also differs from agency as a philosophical idea: measured belief in one's capacity is an empirical report, whereas claims about the metaphysics of agency require a distinct argument. Bandura (1982) Sitzmann & Yeo (2013)
Historical Formation of Goal-Setting Research
Edwin Locke's 1968 article situated conscious goals and task incentives within experimental industrial and organizational psychology. It analyzed evidence that difficult targets and specific high standards could outperform easier or indefinite instructions, and it treated goal-related intentions as mediators through which conditions such as deadlines or financial incentives influence output. This was a methodological development because goal difficulty and standard specificity could be manipulated while task performance was observed directly. Subsequent applied studies, including work in industrial settings, helped move the theory beyond laboratory demonstrations. The historical finding concerns performance under identified conditions; the original models were not intended to establish that every demanding life aspiration always improves welfare or learning. Locke (1968) Locke & Latham (2002)
Mento, Steel, and Karren's 1987 meta-analysis examined research published between 1966 and 1984, distinguishing comparisons of hard against easy goals from those contrasting specific hard goals with vague or absent ones. Their conclusion supported both central predictions while highlighting setting-dependent moderators and the interpretation limits of participative assignment. Locke and Latham's 2002 synthesis reviewed roughly thirty-five years of evidence, mechanisms and moderators, including feedback, commitment, ability, and task complexity. The underlying scientific program gained influence because it proposed measurable interventions and conditional predictions, not merely a vocabulary of aspiration. Later work on learning goals further exposed the importance of the skill-acquisition phase. Mento et al. (1987) Locke & Latham (2002) Seijts et al. (2004)
Historical continuity also requires understanding what goal-setting research left to adjacent traditions. Achievement-motivation theorists studied preferences among tasks of varying success probability, social-cognitive researchers examined beliefs about personal capability, and educational psychologists tracked value and expected success across developmental contexts. Feedback-control approaches modeled discrepancy reduction, while situated motivational accounts emphasized the changing social meanings of achievement. The intellectual history is therefore a convergence of research programs pursuing overlapping phenomena from different experimental entry points. Their constructs can be linked theoretically while remaining distinguishable by observation: assigned goal difficulty, subjective probability, perceived personal capability, and actually achieved task performance are separate measurements. Atkinson (1957) Carver & Scheier (1982) Eccles & Wigfield (2002)
Origins and Development of Self-Efficacy Theory
Albert Bandura's 1977 formulation of self-efficacy introduced a systematic account of perceived capability in behavioral change. The analysis separated expectations of successful performance from expectations about what outcomes a particular behavior would produce. It placed attention on the ways efficacy expectations could influence choice of activities, effort, coping and persistence when people encountered threatening or demanding situations. Bandura's 1982 discussion of human agency expanded the scope to coordination of skills in changing circumstances and reviewed experimental evidence involving induced efficacy beliefs. The historical contribution was a testable psychological mechanism connecting learned experience, anticipated action, and subsequent behavior. Bandura (1977) Bandura (1982)
The theory identifies four frequently discussed sources of efficacy information: mastery experiences, observation of comparable models, verbal or social persuasion, and interpreted physiological or affective states. These sources are not equal in credibility under all conditions. Directly successful performance can supply stronger information about task capability than generic reassurance, while repeated success on an irrelevant task may transfer poorly. Modeling effects depend on perceived comparability and the availability of necessary skills. Anxiety or arousal may be interpreted differently depending on context and prior learning. Each pathway suggests a study in which information source, actual capability, expected outcomes and subsequent behavior can be measured separately. Bandura (1977) Bandura (1982)
An important methodological consequence follows. A high efficacy report can be a consequence of earlier attainment as well as a predictor of later choice. A third variable, such as access to coaching, practice opportunities or prior ability, can influence both belief and performance. Longitudinal and experimental designs help evaluate these links, but even manipulated confidence may alter the amount of effort invested through competing feedback dynamics. A strong theory of efficacy consequently requires time-ordering, measurement of previous performance, intervention-specific tests, and attention to potential negative effects of unwarranted certainty. Vancouver et al. (2002) Sitzmann & Yeo (2013)
The Expectancy–Value Tradition
Achievement-related expectancy–value models ask why a person selects one activity over another, persists in one course, or withdraws from a demanding pursuit. Wigfield and Eccles's 2000 account formally distinguished ability beliefs, expected success, subjective value and developmental changes in achievement choices. Eccles and Wigfield's 2002 review integrated expectancy, task value and motivation–cognition models, illuminating how interests, perceptions of competence, social experience and anticipated outcomes contribute to decisions about education. These models move beyond target difficulty to explain why particular targets become meaningful to particular actors. The dependent variables include choice and continued engagement as well as achievement itself. Wigfield & Eccles (2000) Eccles & Wigfield (2002)
A student can value mathematics because its problems are intrinsically engaging, because competence is integral to an identity, because a desired career requires it, or because the activity provides several benefits simultaneously. These are distinguishable value components rather than interchangeable labels for enthusiasm. Cost introduces additional structure: difficult study may compete with paid work or caregiving, involve anxiety, or require sacrificing another valued activity. An expectancy measure should refer to a defined task and horizon; a value measure should specify the kind of worth being assessed; and cost should be measured rather than silently absorbed into a residual error term. Flake and colleagues' measurement study illustrates why careful operationalization can change the interpretation of the expectancy–value relation. Flake et al. (2015) Wigfield & Eccles (2000)
The more recent Situated Expectancy-Value Theory explicitly emphasizes developmental, social-cognitive and sociocultural organization. Its argument rejects a static treatment of motivational beliefs, considering how a student's experiences, personal identities and immediate context condition the meanings of success and value. This makes the model relevant to inequality of opportunity: two persons can report the same confidence yet encounter very different requirements, institutional support, discrimination or costs. Such contextual differences are not unmeasured personal motivation; they are factors that a credible explanation of choice must incorporate. Cross-cultural or intergroup claims require measurement invariance and careful testing of how tasks and options are presented. Eccles & Wigfield (2020) Flake et al. (2015)
Goal Difficulty, Specificity, and Performance
Goal-setting theory proposes that sufficiently specific and challenging goals can improve performance by directing attention toward relevant activities, mobilizing effort, increasing persistence, and stimulating the search for useful strategies. The mechanism depends on task structure. On routine tasks, higher standards can prompt additional work without requiring substantial new knowledge. On novel tasks, demanding output criteria may compete with exploratory learning, and the relevant intervention may be to set a strategy-learning target. The theory therefore yields conditional predictions for performance rather than an unconditional ranking of goals by difficulty. The actor must understand and accept the target, possess or acquire the required abilities, and receive feedback informative enough to regulate effort. Locke & Latham (2002) Seijts et al. (2004)
The goal difficulty relation also raises a problem of selection and measurement. An individually challenging target can have different absolute levels for experts and beginners, so one study's 'hard' condition may not match another's. A specific criterion can improve observed output partly because it aligns attention to the dependent measure, a benefit that becomes less meaningful if other valued aspects of work are neglected. Researchers must identify whether the outcome is speed, accuracy, durable learning, creativity or ethical quality, and whether performance gains on one indicator coincide with losses elsewhere. The 1987 meta-analysis examined task and setting moderators, while later critical discussion exposed the potential for narrowed attention, distorted incentives and gaming of targets. Mento et al. (1987) Ordóñez et al. (2009)
The research implication is to compare goal levels and goal types factorially when possible, with adequate ability assessment and objective outcome measures. Pre-registering what counts as success prevents post hoc changes in standards. For complex tasks, include delayed performance and transfer tests; for organizations, assess omitted quality dimensions and spillovers. The most interpretable estimand is the effect of assigning one specified target rather than another within a stated population, task, timeframe, and feedback regime. It is rarely informative to report that goals 'work' without specifying which comparison, which result and whose resources are at stake. Epton et al. (2017) Seijts et al. (2004)
Goal Commitment, Acceptance, and Ownership
A selected performance standard only influences conduct when it has functional priority for the actor. Goal commitment describes attachment to that standard and willingness to continue pursuing it as conditions change. The concept differs from declaring a high aim, expressing optimism, or obeying a momentary instruction. Klein, Wesson, Hollenbeck, and Alge's 1999 meta-analysis synthesized 83 independent samples and examined antecedents and consequences of goal commitment. Its relevance to goal-setting research is methodological: commitment may moderate the association between assigned difficulty and performance, and it requires independent measurement. A person can accept a demanding target because its rewards matter, because social obligations require it, or because the task expresses personally endorsed interests. Those motives may produce different persistence patterns even when the immediate performance target is identical. Klein et al. (1999) Locke & Latham (2002)
Assignment and participation are therefore variables, rather than moral labels. In structured workplaces a goal assigned by a supervisor can become effective if its rationale is understood, the individual believes the necessary actions are feasible, and appropriate resources are available. Participative goal-setting can improve information exchange, identify practical constraints, or increase ownership, but the observed association can change once the target difficulty is held constant. A careful experiment distinguishes participating in the choice of target from receiving a more feasible target, better feedback, or a different interpersonal climate. Research should also measure whether participants continue to endorse the goal after disappointing feedback. Commitment is a dynamic state shaped by anticipated reward, procedural fairness, changing opportunity and prior results. Mento et al. (1987) Klein et al. (1999)
Goal commitment can carry costs when success criteria become obsolete or incompatible with other important aims. Escalating effort toward an impossible target may be experienced as loyalty but produce poor adaptive functioning. The proper test of commitment is consequently not maximal persistence in every case; it is the association between maintaining an accepted standard, its objective attainability, and the full set of relevant outcomes. A longitudinal design should compare goal revision and eventual well-being rather than treating departure from the initial standard as a universal failure. This consideration is especially important in settings with asymmetric authority or incentives, where continued pursuit may reflect obligation rather than confident endorsement. Wrosch et al. (2003) Ordóñez et al. (2009)
Task Strategy, Learning Goals, and Feedback
Goal difficulty and effort explain only part of performance variation. A person can invest more time in an ineffective strategy and make little progress, while a modest change in technique can substantially improve results. Seijts, Latham, Tasa, and Latham's 2004 experiment used a complex business simulation to compare specific learning goals with performance goals and vague goals. Their reported findings favored the learning-goal condition in that context, with information search and self-efficacy investigated as mediating processes. The lesson is a conditional account of task structure: when the required skill is unfamiliar, the target should make strategy discovery observable. Learning goals and performance goals differ in the regulated process, not simply in how ambitious they sound. Seijts et al. (2004) Seijts & Latham (2005)
Feedback connects standards to regulation. Informative feedback communicates the discrepancy between the current state and the desired criterion, but its motivational effect depends on interpretation, reference points and perceived controllability. A failed attempt can signal a need for additional practice, a defective strategy, an overly difficult standard or a change in circumstances. Without data discriminating these possibilities, the same negative feedback may encourage skill acquisition in one participant and disengagement in another. Goal-setting research therefore treats feedback as a central moderator. To isolate mechanisms, studies need to manipulate feedback frequency and informational content separately and assess whether recipients revise skill estimates or merely adjust the quantity of effort. Locke & Latham (2002) Carver & Scheier (1982)
The practical difference between an outcome and a learning measure should be made visible in research design. A treatment group may improve short-term task output by emphasizing one metric while showing weaker retention, conceptual understanding or transfer to a new task. A learning-oriented evaluation samples multiple responses and later conditions, separating acquisition from demonstration of already learned procedures. Conversely, an intervention that increases exploratory behavior without enhancing eventual competence should not be celebrated solely for greater 'engagement.' Joint measurement of strategy quality, output, error patterns and delayed transfer supports an interpretation of how goals influence learning. Seijts et al. (2004) Harkin et al. (2016)
Expectancy Calibration and Choice Under Uncertainty
Expected success is a judgment about the relation between a chosen action and anticipated achievement, made under uncertainty. The judgment may be implicit in choice or explicitly reported as a probability, confidence rating, or perceived likelihood of meeting a standard. Different elicitation formats impose different cognitive demands: binary confidence, numerical probability and agreement with general self-belief statements need not represent the same variable. Calibration is the correspondence between predicted and observed outcome frequencies across repeated trials. Good calibration requires feedback, enough observations and a stable definition of success. A judgment can be accurate even when success is unlikely, and overconfidence can coexist with high objective ability. Wigfield & Eccles (2000) Vancouver et al. (2002)
Expectancy is both predictive and dynamically updated. Earlier achievements provide evidence of feasibility; failure can alter subsequent confidence even if intrinsic value remains constant. In a classroom, a student may choose an advanced course after repeated mastery, while a poor examination grade can reduce expectations without altering the value of the subject. Expectancy is also sensitive to perceived institutional barriers, access to learning support and social information about who succeeds. In observational datasets, expected success may therefore partly index earlier opportunities. Causal interpretation requires adjustment for previous proficiency, repeated within-person assessment and, where appropriate, randomized changes to information or support. Eccles & Wigfield (2020) Sitzmann & Yeo (2013)
The decision to attempt a goal involves possible benefits, anticipated success, effort requirements and the value of alternative uses of time. A simple expectancy-times-value expression can clarify why an attractive but implausible outcome may not be selected; it is not an estimate of the actor's complete utility function. People also consider anticipated regret, identity-relevant meanings, social obligations and heterogeneous costs. An informative study separates belief about success from estimated value and externally measured feasibility. It must also distinguish expecting a result because one can perform the action from expecting a result because external conditions favor success. This distinction protects against interpreting structural opportunity as an individual personality trait. Bandura (1977) Eccles & Wigfield (2020)
Subjective Value, Cost, and Meaning
The value of a goal can arise from interest in the activity, importance to self-identity, anticipated usefulness for future choices, or external consequences. These values often coexist. A school subject may be genuinely interesting while also offering professional utility; an assigned organizational target may have little intrinsic attraction yet contribute to a wider personally endorsed project. The scientist's task is to distinguish components in a measurement model and determine which components independently predict choice and performance. Eccles and Wigfield's framework provides such distinctions, while situated extensions attend to the social history through which values are learned and maintained. Eccles & Wigfield (2002) Eccles & Wigfield (2020)
Cost should be modeled as more than the reverse of enjoyment. Flake and colleagues in 2015 combined conceptual review with qualitative and factor-analytic work to investigate the multidimensional structure of perceived cost. Effort cost concerns expected work, emotional cost concerns anticipated distress, and opportunity cost concerns benefits surrendered by selecting one activity. Cost and task difficulty can interact without being identical: a demanding activity may be manageable for an expert but exceptionally costly for a student with limited time or equipment. Studies that omit cost can falsely attribute avoidance to low interest or low competence. They also risk explaining a structurally constrained choice as a motivational deficit. Flake et al. (2015) Wigfield & Eccles (2000)
Utility-value interventions experimentally test whether meaning can be changed. Hulleman and Harackiewicz's 2009 randomized field study encouraged high-school students to connect science material to their lives; the reported gains in interest and grades were concentrated among students with low success expectations. A related 2010 program manipulated relevance in laboratory and college classroom experiments, with stronger interest benefits in lower-expectancy or lower-performing groups. These findings support context-sensitive interventions directed at value, while the available evidence does not establish that every relevance exercise raises achievement for every learner. Research should distinguish changes in perceived utility from subsequent interest, persistence and graded outcomes, and test whether effects endure after the instructional unit. Hulleman & Harackiewicz (2009) Hulleman et al. (2010)
Self-Efficacy Formation and Behavioral Consequences
Capability beliefs are shaped by experiences that permit the actor to infer whether a particular course of action is executable. Successful mastery can alter self-efficacy because it provides task-specific evidence about competence. Observing a comparable model may change beliefs about what can be learned, while social persuasion can supply expectations that later receive confirmation or contradiction through performance. Physiological states can be interpreted as challenge, readiness or threat. The causal sequence is not fixed: beliefs also influence whether an individual attempts a demanding task and consequently whether mastery opportunities occur. This bidirectionality is central to Bandura's account, and it is measurable through repeated assessment rather than assumed from one survey. Bandura (1977) Bandura (1982)
Self-efficacy may support action by making difficult tasks appear manageable, encouraging effort and extending persistence after setbacks. At the same time, an efficacy score partly represents experience and self-knowledge: highly capable participants often report higher efficacy because they have repeatedly demonstrated the relevant skill. Researchers must therefore compare predictions after accounting for objective competence and previous attainment, and they should investigate whether confidence calibration changes exploration or attention to errors. Increased perceived capability can be beneficial when it invites engagement with an appropriate challenge, but it can also reduce information search or increase unwarranted risk when confidence exceeds skill. The direct effect of belief remains an empirical question for a given task. Vancouver et al. (2002) Sitzmann & Yeo (2013)
Generalization is an additional limit. A person who succeeds at solving one kind of problem may reasonably develop stronger expectations for similar problems but not for tasks requiring a different skill. Broad self-efficacy measures can capture dispositional self-beliefs or life experience, yet their predictive association may depend on overlap with the outcome domain. Intervention studies are strongest when they specify the capability judged, deliberately alter a credible source of that judgment, and measure transfer to independent, nontrained outcomes. Researchers should report the extent to which a confidence manipulation changes actual ability, since changing teaching quality and changing belief at the same time cannot isolate a pure efficacy effect. Bandura (1982) Sitzmann & Yeo (2013)
Interactions Among Goals, Expectancies, and Efficacy
A goal can alter performance only through a functioning system of appraisal, commitment, available means and regulation. Expected success shapes the choice of difficulty and whether the actor accepts an assigned target. Commitment determines whether the target retains priority. Feedback supplies information about the discrepancy between current and desired results, while efficacy beliefs influence the interpretation of whether new action is feasible. The variables can interact: a challenging goal may stimulate effort in an actor with adequate skill and realistic confidence but undermine persistence when the outcome is perceived as inaccessible. This interaction is a hypothesis about causal organization; its exact functional form must be tested in particular tasks. Locke & Latham (2002) Bandura (1977) Klein et al. (1999)
Statistical models can estimate these interactions by randomly varying task standards and feedback while measuring baseline competence, success expectations and task value. However, interpreting measured efficacy as a moderator requires recognizing its likely correlation with unobserved ability, experience, organizational resources and past feedback. Experiments should consider manipulating information about capability separately from actual training and from the assigned goal. Time-series data can distinguish immediate confidence changes following success from subsequent changes in effort. Structural models that treat expectancy, value and efficacy as perfectly separable latent traits risk unstable inference when questionnaire items overlap conceptually. A credible synthesis thus relies on discriminant validity tests rather than assuming that different names prove different mechanisms. Vancouver et al. (2002) Sitzmann & Yeo (2013)
The integrative purpose is explanatory rather than terminological. A person may be highly efficacious but uninterested in a low-value activity; another may value an outcome but require further instruction before the target becomes feasible. A third may possess both confidence and interest yet lack the time or institutional access needed to act. Each case predicts a different intervention: altering the target, increasing value, teaching a strategy, improving feedback, or expanding opportunity. Their consequences should be compared with meaningful endpoints, including quality and duration of performance rather than intention alone. Such conditional reasoning makes the three research traditions complementary while preserving their distinct operational commitments. Eccles & Wigfield (2020) Epton et al. (2017)
Laboratory Evidence for Specific and Challenging Goals
The strongest historical goal-setting claims concern experimentally manipulated targets with observable output. Locke's 1968 synthesis contrasted difficult with easier goals and specific high standards with vague instructions. The broader quantitative evidence was subsequently assessed by Tubbs in 1986 and Mento, Steel, and Karren in 1987. These syntheses generally favored challenging and specific goals on the tasks included, while finding heterogeneity across setting, operational definition and feedback conditions. Their historical importance comes from the aggregation of comparative effects across different work tasks, but such aggregation cannot remove variation in ability, assignment procedures or the meaning of difficulty. The empirical question is always which goal condition improves which endpoint under which constraints. Locke (1968) Tubbs (1986) Mento et al. (1987)
Interpretation depends heavily on the comparison group. A precise, demanding performance target compared with no target may improve output because it clarifies what must be done, whereas a demanding target compared with an equally specific moderate target chiefly tests difficulty. A third comparison can hold difficulty constant and vary whether the individual endorses the goal. Feedback can also alter the efficacy of either intervention. Consequently, apparent inconsistency between studies often reflects different causal estimands. A strong trial predefines its targets, ensures that the task is attainable for the sampled participants, records adherence and measures performance independently of self-report. A synthesis should retain distinctions among output quality, speed, accuracy and learning. Mento et al. (1987) Locke & Latham (2002)
Difficult goals can also generate what researchers call a ceiling or floor problem: an easy task can leave little room to improve, and an impossibly demanding task can produce little informative variation in successful attainment. Personal adaptation to the task may shift the relative difficulty over repeated trials. If experimenters adapt the target to prior ability, they should document the adaptation rule and consider whether confidence and competence predict the assigned level. If participants choose their own goals, prior performance may determine selection and confound subsequent goal–performance correlations. Random assignment, equivalent baseline distributions, reliable manipulation checks and appropriate longitudinal analyses strengthen the causal interpretation. Locke & Latham (2002) Mento et al. (1987)
Meta-Analytic Evidence for Goal Interventions
Epton, Currie, and Armitage's 2017 review investigated goal setting as a distinct behavior-change technique, extracting 384 effect sizes from 141 randomized-trial papers involving 16,523 participants. The authors reported a random-effects mean of d = 0.34 with a 95% confidence interval from 0.28 to 0.41. This is an estimate for a defined collection of intervention contrasts rather than a constant effect for every behavioral outcome. The literature spans health and other behavior-change settings with variation in how goals are formulated, delivered and assessed. The review is especially informative because it asks about goal-setting interventions beyond the controlled performance tasks common in earlier organizational research. Epton et al. (2017)
A mean effect of this magnitude should be interpreted alongside outcome heterogeneity, practical relevance, risk of bias and the target behavior's baseline frequency. A small standardized difference can matter for population interventions when costs are low and outcomes consequential; the same statistic may correspond to a less important improvement on a proxy questionnaire. Multiple effect sizes extracted from a study require appropriate dependence handling. A meta-analysis can estimate average effects under an explicit search and inclusion protocol but may be influenced by selective reporting or the quality of the underlying trials. The intervention's content must also be checked: some studies compare goal setting with usual support, while others test goal specificity or goal distance. These are different interventions. Epton et al. (2017) Mento et al. (1987)
The synthesis supports the causal usefulness of specifying goals under certain conditions, with effect sizes smaller than some universal motivational promises suggest. It does not support replacing skill acquisition, material access or sustained monitoring with a one-time verbal commitment. Long-term outcomes require follow-up beyond immediate task response and must distinguish behavioral maintenance from initial improvement. Researchers should also record potential displacement of other behaviors; meeting one narrowly measured target may come at the expense of an unmeasured objective. Such concerns become crucial in organizational contexts governed by performance metrics and in educational settings where broad understanding matters more than a single score. Epton et al. (2017) Ordóñez et al. (2009)
Experimental Tests of Efficacy and Performance
The self-efficacy literature has a distinctive causal challenge: the actor's previous successes predict both current confidence and future skill-based performance. Between-person correlations therefore cannot by themselves demonstrate that increasing confidence will improve outcomes. Vancouver, Thompson, Tischner, and Putka's 2002 article directly investigated this complication through analytic-game studies, including an experimental efficacy manipulation. Their reported results found circumstances in which increased efficacy was associated with poorer subsequent task performance, consistent with proposed overconfidence and reduced error sensitivity. These findings challenge a simple, monotonic interpretation of the relation between confidence and achievement while leaving room for positive effects in other situations. Vancouver et al. (2002)
Sitzmann and Yeo's 2013 meta-analysis examined within-person associations across time. The authors reported a corrected self-efficacy–performance correlation of 0.23, which became 0.06 and statistically nonsignificant after controlling for a linear trajectory. The analogous past-performance–later-efficacy relationship remained 0.30 after controlling for trajectory. These are reported meta-analytic estimates, not a license to assert that efficacy never affects performance. The result emphasizes that improved task competence can produce rising confidence and rising performance together, yielding an apparent causal direction from belief to result. Different moderator conditions yielded different self-efficacy relationships, further arguing for time-sensitive and task-specific models. Sitzmann & Yeo (2013)
A discriminating empirical program measures competence before belief, manipulates credible efficacy-related information without accidentally changing actual teaching, and examines later output at repeated intervals. Experiments should check whether altered confidence changes information search, persistence after failure, willingness to choose difficulty or attention to feedback. A manipulation that raises confidence and simultaneously provides better strategies does not isolate the psychological effect of confidence. Likewise, study designs using only two waves may conflate long-term growth with short-term deviations. The methodological implication is to model both directions of influence and to report competing dynamic explanations, including negative feedback loops. Vancouver et al. (2002) Sitzmann & Yeo (2013) Vancouver et al. (2010)
Goal Progress Monitoring and Feedback Evidence
Progress monitoring is the repeated comparison of actual movement toward a criterion with a desired standard. It is conceptually downstream from choosing a goal but tightly connected to efficacy and expectancy because new information changes predictions of future success. Harkin and colleagues' 2016 meta-analysis synthesized 138 randomized studies involving 19,951 participants. Interventions increased monitoring frequency with a reported standardized effect of d+ = 1.98, 95% CI [1.71, 2.24], and increased goal attainment with d+ = 0.40, 95% CI [0.32, 0.48]. The review also reported larger goal-attainment effects when monitoring outcomes were recorded physically or made public, and a mediation analysis relating increased monitoring to attainment. Harkin et al. (2016)
The design establishes the usefulness of monitoring interventions within the included settings. Yet a statistical mediation result is not automatically proof of a unique causal mechanism, because intervention-driven processes may change together and mediator measurement may be imperfect. Making progress public may affect social accountability, perceived approval, anxiety or social comparison, as well as simple information access. Recording a measure can alter attention and subsequent decisions. A careful follow-up experiment would randomize these elements separately and examine whether the effect persists after recording ends. Researchers should also distinguish self-monitoring of behavior from measurement of the eventual target outcome, and test whether reported gains reflect reactive measurement. Harkin et al. (2016) Carver & Scheier (1982)
Feedback has a special relevance to self-efficacy because participants infer competence from progress. Poor progress may lower confidence while simultaneously stimulating effort when a goal remains achievable; extremely poor progress may motivate disengagement. The effects depend on whether discrepancy is interpreted as a solvable strategy problem, a lack of skill, insufficient time or an external barrier. Repeated feedback designs can model these transitions instead of treating motivational strength as constant. In applied settings, useful monitoring must address accuracy, privacy, organizational incentives and the difference between tracking activity and improving meaningful outcomes. Harkin et al. (2016) Vancouver et al. (2010)
Field Experiments on Utility, Interest, and Expectancy
Expectancy and value have been manipulated in educational settings to test predictions beyond questionnaire correlations. Hulleman and Harackiewicz's 2009 randomized classroom field experiment asked students to connect the science they were studying to personally relevant examples. The published abstract reports improved interest and grades among students with low prior success expectations. The logic is theoretically important: a relevance intervention targeting subjective utility may be particularly beneficial when students are otherwise unlikely to identify worthwhile reasons to persist. This is not equivalent to instructing every student to adopt a higher performance target. The intervention attempts to modify the subjective meaning of the educational activity. Hulleman & Harackiewicz (2009)
Hulleman, Godes, Hendricks, and Harackiewicz's 2010 study reported two randomized experiments, one in the laboratory and one in a college classroom, manipulating the relevance of material through writing exercises. The abstract reports increases in perceived utility and interest, especially for lower-expectancy or lower-performing participants. Its reported mediation analysis implicated perceived utility. These findings connect value to both emotional engagement and performance without presuming that greater efficacy is the sole mechanism. The design also suggests differential treatment effects: an intervention may have relatively little impact for learners already highly interested or convinced of their success. Replication across classrooms, ages and subjects is necessary before asserting universal transportability. Hulleman et al. (2010)
The field context brings additional measurement challenges. Grades reflect teacher assessment and course tasks; interest questionnaires can be sensitive to wording and social desirability; sustained future enrollment is a different endpoint from performance in a single semester. Treatment adherence may vary in how thoughtfully students connect course material to personal priorities. For subsequent research, preregistered moderation by baseline expectancy and verified long-term outcomes would improve interpretation. The strongest conclusion is that subjective utility is a manipulable, context-dependent component of achievement motivation that can interact with initial expectations. The evidence also demonstrates why a scientific account of goal pursuit requires a more differentiated motivational model than 'set a harder target.' Hulleman & Harackiewicz (2009) Hulleman et al. (2010) Eccles & Wigfield (2020)
Multiple Goals, Competition, and Dynamic Regulation
Individuals rarely pursue one objective in isolation. A worker may balance speed, accuracy, safety, learning and family obligations; a student balances several courses and social commitments. A change in expected success can influence which aim receives resources even when each aim's subjective value remains unchanged. Vancouver, Weinhardt, and Schmidt's 2010 computational work modeled multiple-goal pursuit, integrating selection and striving. Their model reproduced task-specific findings in which relative discrepancies influenced resource allocation differently early and late in a session. This illustrates that goal discrepancy does not always predict the same behavioral response; its effect can vary with time remaining and competing alternatives. Vancouver et al. (2010)
An integrated dynamic model requires state variables for progress, changing expectancy, accumulated skill, time remaining, available action alternatives, and perhaps goal-specific costs. Such models can produce hypotheses about switching and persistence that static cross-sectional regressions cannot specify. Their explanatory power, however, must be evaluated through prospective predictions and parameter identifiability. Fitting observed data with enough flexible parameters is not equivalent to demonstrating that the proposed psychological mechanism is uniquely correct. Model comparison should test different causal accounts against held-out behavior, including conditions in which goals become newly unattainable or unexpectedly easy. Vancouver et al. (2010) Sitzmann & Yeo (2013)
The distinction between persistence and flexibility also matters for outcomes. An actor who abandons an obsolete target may be exercising effective regulation, whereas another who continues may simply be unable to update a commitment. The relevant dependent variable is not persistence alone but the suitability of action relative to current evidence, costs and valued alternatives. Studies need to quantify what alternatives were realistically available and whether withdrawal leads to meaningful goal reengagement. This line of inquiry connects goal-setting research to the broader psychology of adaptation without duplicating neighboring dossiers devoted specifically to implementation planning, habit formation or clinical coping. Wrosch et al. (2003) Ordóñez et al. (2009)
Measurement and Psychometric Interpretation
Scientific comparisons across goals, expectations and efficacy beliefs require discriminant measurement. Goal difficulty can be manipulated experimentally as a standard, or estimated relative to the participant's ability and current opportunity. Specificity can be coded from the precision of the performance criterion. Commitment is normally assessed through expressed attachment, priority and intended persistence, but its validity must be compared with actual sustained behavior. Expectancy should be elicited for a specified success event and time horizon; self-efficacy should refer to the actions the participant believes can be executed. Using a generic confidence question as a proxy for all these constructs makes the theory difficult to test. The instruments must have clear referents, adequate reliability and evidence that they measure the intended constructs. Bandura (1977) Klein et al. (1999)
Value and cost scales require particular care because people may assign multiple meanings to the same activity. Flake and colleagues' 2015 analysis of cost emphasized multiple forms of perceived burden; collapsing them into a single item would obscure whether disengagement reflects strenuous work, anxiety, or competing opportunities. Longitudinal measurement also requires attention to whether an item's meaning changes after training. For example, rating a task 'difficult' may index novelty before instruction but effort after learning. Measurement invariance tests and cognitive interviews can assess whether the scale functions similarly across groups, time points and educational contexts. In observational research, separating reported beliefs from independently assessed opportunity and ability is essential to construct validity. Flake et al. (2015) Eccles & Wigfield (2020)
Measurement of achievement is equally consequential. A task score may combine speed and quality or reward accurate final answers without observing exploratory learning. Behavioral tracking can quantify initiation, duration, error correction and actual goal completion, while interviews reveal priorities and perceived obstacles. Physiological effort indices and digital trace measures provide complementary evidence but require domain-specific validation: more arousal or more clicking does not necessarily mean stronger motivational commitment. A robust study triangulates self-report, objective task performance and contextual variables, then explains which measure tests each proposed pathway. No one indicator can serve as a universal 'motivation meter.' Locke & Latham (2002) Harkin et al. (2016)
Causal Identification, Time, and Model Comparison
The causal graph of goal pursuit contains at least four reciprocal relations. Previous performance influences efficacy; efficacy can influence the difficulty a person accepts; chosen difficulty can influence effort; and obtained feedback updates both future efficacy and future goal selection. Ability, resources and social expectations may influence all of these nodes. Cross-sectional structural equation models can describe associations but are vulnerable to omitted variables and ambiguous temporal direction. Randomized assignment of goals addresses one causal question, whereas repeated measurement of belief and performance addresses another. A scientifically meaningful causal analysis specifies its target intervention and does not interpret every longitudinal regression coefficient as a causal pathway. Vancouver et al. (2002) Sitzmann & Yeo (2013)
For efficacy, a high-quality experimental program might compare a credible performance-information manipulation, an actual skills-training manipulation, and their combination, while holding task requirements constant. Baseline competency should be measured before treatment; self-efficacy and expected success should be measured at multiple later points; subsequent performance should be assessed with independent scoring. This factorial design can help determine whether learning and belief have separable effects. Researchers also need to evaluate adverse outcomes, such as confidence-induced reduction in vigilance, to avoid presupposing a favorable direction. Field studies should document selection into programs, attrition, instructor behavior and concurrent institutional changes that could account for improvement. Vancouver et al. (2002) Vancouver et al. (2010)
Formal models create further tests by spelling out the mechanism that maps observed feedback to goal priority and effort. The multiple-goal model of Vancouver and colleagues represents allocation over competing standards, while expectancy–value models represent beliefs and valued alternatives. Models should be evaluated against held-out predictions of goal switching, performance after unexpected failure, and behavior under resource constraints. Fit to a single time series is weak evidence if free parameters can mimic several causal stories. Preregistered competing predictions, informative manipulations and analyses of parameter recovery would advance evidence more than an ever-expanding inventory of constructs. Vancouver et al. (2010) Sitzmann & Yeo (2013)
Development, Socialization, and Opportunity
Motivational beliefs develop through accumulated task experience and the feedback supplied by teachers, peers, families, organizations and institutions. Differences in expected success may reflect genuine differences in instruction and performance history alongside self-interpretation. The situated expectancy–value framework incorporates the social and cultural meaning of achievement, personal identities and perceived costs. In studying developmental changes, one must compare equivalent tasks and account for differences in the option sets accessible to each age group. A child's expressed utility for an activity cannot be interpreted in the same way as an adult's choice among employment, caregiving and education without modeling the surrounding environment. Eccles & Wigfield (2020) Wigfield & Eccles (2000)
Cross-cultural generalization depends on the equivalence of both measurement and the circumstances in which behavior is observed. An item about personal confidence may be expressed differently when achievement is framed collectively or individually; organizational hierarchies may change whether participants feel free to decline a goal. These are concrete reasons to test response processes, translated instruments and the functional meaning of commitment before comparing mean levels. Researchers should describe local opportunity structure and task incentives rather than treating a questionnaire's numerical comparability as proof of a universal psychological mechanism. At the level of experimental explanation, one can assess whether within-context changes in feedback or goal specificity produce similar effects under matched conditions while leaving mean value differences to separate analysis. Eccles & Wigfield (2020) Flake et al. (2015)
Developmental trajectories and the accumulation of skill also alter feedback interpretation. Repeated success can increase efficacy and permit the choice of greater challenge; repeated failure under inadequate support may depress future expectations even when the immediate task is within reach after training. Effective research therefore tracks the co-development of performance, belief, skill and opportunity. Participants who remain in a course or job may differ systematically from those who leave, producing biased results if attrition is ignored. When distinguishing persistence from adaptation, investigators should examine alternative routes to achievement, satisfaction and well-being. Sitzmann & Yeo (2013) Wrosch et al. (2003)
Education, Organizations, Health, and Performance
Educational applications of the evidence center on specifying feasible and worthwhile targets, aligning goal type with skill level and offering informative feedback. Challenging performance goals are reasonable for mastered procedures, while learning goals may be more suitable for acquiring unfamiliar strategies. Utility-value exercises illustrate an alternative intervention that addresses why the subject matters, especially for students with initially lower expectations. The studies do not establish that manipulating a student's beliefs can replace adequate teaching, educational access, or assessment quality. Applied programs should report proficiency, persistence, course selection, delayed transfer and possible disparities in participation. Seijts et al. (2004) Hulleman & Harackiewicz (2009) Hulleman et al. (2010)
Organizational performance systems make goal-setting effects directly observable, but also introduce incentive distortions. Quantified targets can direct attention to measurable output while neglecting unmeasured safety, collaboration and learning. Ordóñez, Schweitzer, Galinsky and Bazerman's 2009 critical analysis warned that inappropriate goal prescriptions can narrow focus, elevate risk taking and foster unethical behavior. Such consequences depend on task incentives and measurement structure rather than representing inevitable properties of having goals. Evaluation should include quality-adjusted output, unintended consequences, shared goals and the freedom to revise targets when conditions change. A production target that raises throughput while worsening error rates is an incomplete measure of intervention success. Ordóñez et al. (2009) Locke & Latham (2002)
Health, rehabilitation and sport contexts introduce additional constraints: functional capacity, symptom burden, changing physical condition, team support and safety. Goal-setting interventions studied across behavior-change domains show average benefits, while the estimated magnitude cannot be assumed for a specific diagnosis, treatment or athlete population without domain-specific trials. Setting an attainable exercise target differs from determining medical suitability or prescribing an intervention. Clinical or rehabilitation translation requires qualified practitioners, individualized safety assessments and direct evidence for the relevant population. Self-efficacy may affect engagement, but changes in symptoms, recovery or function cannot be attributed to confidence without controlling treatment, baseline function and opportunity. Epton et al. (2017) Bandura (1977)
Unintended Effects, Ethics, and the Limits of Performance Metrics
Specific targets can create incentives to optimize what is counted. A student rewarded exclusively for a test score may narrow study to tested content; an employee pressured by a quota may sacrifice accuracy; a digital platform optimizing engagement may favor readily measured actions over the person's longer-term priorities. These outcomes follow from the measurement and incentive arrangement and should be treated as testable intervention effects. The presence of a clear goal gives researchers an independent variable, while the completeness of the outcome measure determines what claims about improvement are warranted. Evaluating an intervention means measuring benefits and harms across the relevant consequence space. Ordóñez et al. (2009) Locke & Latham (2002)
Ethical assessment also requires examining who sets the standard and who bears the cost. People can appear committed when they lack realistic power to refuse a task; apparently low motivation can reflect inaccessible resources or competing obligations. Such patterns are empirical questions about institutions as well as interpretive questions about fairness. A program's effectiveness should therefore be assessed with respect to autonomy, opportunity and sustained functioning, not solely compliance. Group averages can conceal differential costs for participants with limited time, caregiving responsibilities or reduced access to materials. Transparent reporting of subgroup evidence, feasibility and unintended consequences makes the intervention's boundaries visible. Eccles & Wigfield (2020) Flake et al. (2015)
Public goal monitoring deserves careful handling. Harkin and colleagues' meta-analysis found that public reporting and physically recording progress were associated with larger intervention effects, but those formats also expose personal behavioral information and may alter social pressure. Research designs should evaluate privacy, informed consent and the possibility that participants change behavior to satisfy observers. In institutional applications, confidentiality and proportionality are part of intervention design rather than afterthoughts. A useful goal is one that supports the person's valued functioning under transparent conditions and appropriate oversight. Harkin et al. (2016) Ordóñez et al. (2009)
Artificial Era and Distributed Goal Regulation
Artificial systems increasingly participate in goal setting by offering suggested targets, forecasting completion, scheduling activities, supplying feedback and summarizing progress. Their inclusion changes the information environment of human motivation. A tool that decomposes a difficult project into manageable stages may alter perceived task difficulty, expected success and the amount of individual cognitive effort required for planning. It may also change the standard by which performance is evaluated, for instance by improving the speed of output while leaving independent skill untouched. These are mechanistic hypotheses suitable for controlled study, drawing on established goal-setting and efficacy constructs. The relevant outcome includes both assisted performance and later unassisted transfer. Locke & Latham (2002) Sitzmann & Yeo (2013)
Human–artificial collaboration creates a distinctive potential for confidence miscalibration. A person may attribute an AI-generated result to newly acquired personal competence, or may undervalue personal capability after comparing performance with an automated system. Evaluating these possibilities requires separating the actor's own performance, the system's contribution, independent accuracy, self-efficacy reports and goal adoption. In randomized comparisons, supported and unsupported groups should receive equivalent task opportunities, with additional measures for trust, correction of erroneous suggestions and retained skill. A short-term productivity improvement does not establish a durable increase in self-efficacy. Claims about artificial subjective experience or artificial autonomy remain separately testable theoretical questions. Bandura (1977) Vancouver et al. (2002)
The broader historical horizon Artificial Era · From Homo to Artificial invites a comparison of goal organization in biological and artificial systems. For software agents, researchers can operationalize goals as optimization criteria, policies, constraints and stored task representations, observing persistence, conflict resolution and adaptation under interventions. Human self-efficacy is a report of perceived capability grounded in an individual's psychological and developmental history. Transferring the same construct to artificial agents requires carefully defining what observable process corresponds to capability estimation, what counts as self-reference and how predictions are validated. A shared mathematical language for expectation and action selection may support comparative research without prejudging the phenomenology of either system. Vancouver et al. (2010) Eccles & Wigfield (2020)
Critical Scientific Discussion
Across diverse research programs, a convergent conclusion is that clear, challenging and accepted goals can improve defined task performance when feedback, ability and strategies permit progress. Expectancy–value research explains why the same target is not equally attractive or feasible to all actors, and self-efficacy research explains how perceived capacity enters decisions and task persistence. These findings have different evidential bases. Randomized interventions provide stronger evidence for short-term effects of specified manipulations; prospective observations clarify developmental and ecological patterns; meta-analyses estimate conditional averages; computational models express mechanism candidates. Convergence across them does not entail that one variable or one universal intervention causes every successful pursuit. Locke & Latham (2002) Epton et al. (2017) Eccles & Wigfield (2020)
The self-efficacy dispute is theoretically consequential because it challenges an intuitive direction of causality. Between-person associations, experimentally induced confidence and within-person trajectories can produce divergent results. Vancouver and colleagues demonstrated conditions in which stronger efficacy related to worse performance; Sitzmann and Yeo showed why controlling for performance trends changes the within-person predictive relation. These findings complement Bandura's model by demonstrating the need to specify feedback history, baseline competence, task demands and competing mechanisms. Neither unconditional efficacy enhancement nor the dismissal of all efficacy effects follows from the evidence. Research is strongest when its predictions differentiate a belief-driven change from an ability-driven or feedback-driven change. Bandura (1977) Vancouver et al. (2002) Sitzmann & Yeo (2013)
Important unresolved questions include the transportability of goal-setting effects to complex, long-duration projects; how goals affect creativity, collaboration and welfare when multiple metrics compete; which forms of perceived cost predict disengagement independently of expectancy; how to measure efficacy without conflating it with prior skill; when overconfidence impairs performance; how goal commitment interacts with autonomy; and how artificial assistance alters the calibration of perceived personal capability. Substantial progress will require factorial trials, intensive longitudinal observation, cross-cultural measurement tests, preregistered mechanism comparisons and transparent reporting of unintended outcomes. Those designs can explain when a goal strengthens achievement and when a wiser response is revision or withdrawal. Wrosch et al. (2003) Vancouver et al. (2010) Ordóñez et al. (2009)
Conclusion
Goal setting, expectancy judgments and self-efficacy form an interacting set of psychological mechanisms for organizing intended futures into present action. Goals establish criteria; expectancies estimate success; subjective value and perceived cost determine whether a pursuit warrants priority; self-efficacy concerns the perceived ability to perform requisite actions; and feedback updates the actor's commitments and beliefs. Their effects are conditional on task complexity, skills, resources, social context and alternatives. Scientific explanation therefore depends on separating the operational constructs, preserving their reciprocal relations and evaluating concrete outcomes. The evidence supports structured goal interventions while specifying why confidence, commitment and difficulty cannot stand alone as universal explanations of success. Locke & Latham (2002) Eccles & Wigfield (2020) Sitzmann & Yeo (2013)
Historical research from Locke and Bandura through contemporary field experiments and quantitative syntheses shows how motivational psychology has advanced by making subjective belief and goal properties experimentally tractable. The causal role of efficacy remains an active question requiring careful temporal methods; stronger goal specificity is useful when the criterion aligns with meaningful performance; utility and cost determine whether a seemingly attractive pursuit becomes practical. An ethically adequate application examines the full consequences of goal adoption, including the costs of persistence, adaptability and opportunity to learn. The conclusion is an integrated research agenda grounded in measured behavior, scientifically characterized beliefs and realistic settings. Mento et al. (1987) Epton et al. (2017) Harkin et al. (2016)
The Artificial Era expands this agenda because planning and feedback now occur across human actors, institutions and artificial systems. Comparative research can investigate the distribution of information and control while distinguishing demonstrated changes in human skills from assistance-generated performance. Reliable scientific progress will require designs that identify who selects a target, how expected success is produced, what capabilities are actually present, and whether the resulting action remains beneficial as circumstances evolve. The durable contribution of the existing literature is a set of conditional, empirically testable explanations for how people establish standards, estimate feasibility and regulate effort in a changing world. Vancouver et al. (2010) Locke & Latham (2002)
References
- Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior. Psychological Review, 64(6, Pt. 1), 359–372. https://doi.org/10.1037/h0043445
- Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
- Bandura, A. (1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122–147. https://doi.org/10.1037/0003-066X.37.2.122
- Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality–social, clinical, and health psychology. Psychological Bulletin, 92(1), 111–135. https://doi.org/10.1037/0033-2909.92.1.111
- Eccles, J. S., & Wigfield, A. (2002). Motivational beliefs, values, and goals. Annual Review of Psychology, 53, 109–132. https://doi.org/10.1146/annurev.psych.53.100901.135153
- Eccles, J. S., & Wigfield, A. (2020). From expectancy-value theory to situated expectancy-value theory: A developmental, social cognitive, and sociocultural perspective on motivation. Contemporary Educational Psychology, 61, Article 101859. https://doi.org/10.1016/j.cedpsych.2020.101859
- Epton, T., Currie, S., & Armitage, C. J. (2017). Unique effects of setting goals on behavior change: Systematic review and meta-analysis. Journal of Consulting and Clinical Psychology, 85(12), 1182–1198. https://doi.org/10.1037/ccp0000260
- Flake, J. K., Barron, K. E., Hulleman, C. S., McCoach, D. B., & Welsh, M. E. (2015). Measuring cost: The forgotten component of expectancy-value theory. Contemporary Educational Psychology, 41, 232–244. https://doi.org/10.1016/j.cedpsych.2015.03.002
- Harkin, B., Webb, T. L., Chang, B. P. I., Prestwich, A., Conner, M., Kellar, I., Benn, Y., & Sheeran, P. (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin, 142(2), 198–229. https://doi.org/10.1037/bul0000025
- Hulleman, C. S., & Harackiewicz, J. M. (2009). Promoting interest and performance in high school science classes. Science, 326(5958), 1410–1412. https://doi.org/10.1126/science.1177067
- Hulleman, C. S., Godes, O., Hendricks, B. L., & Harackiewicz, J. M. (2010). Enhancing interest and performance with a utility value intervention. Journal of Educational Psychology, 102(4), 880–895. https://doi.org/10.1037/a0019506
- Klein, H. J., Wesson, M. J., Hollenbeck, J. R., & Alge, B. J. (1999). Goal commitment and the goal-setting process: Conceptual clarification and empirical synthesis. Journal of Applied Psychology, 84(6), 885–896. https://doi.org/10.1037/0021-9010.84.6.885
- Locke, E. A. (1968). Toward a theory of task motivation and incentives. Organizational Behavior and Human Performance, 3(2), 157–189. https://doi.org/10.1016/0030-5073(68)90004-4
- Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57(9), 705–717. https://doi.org/10.1037/0003-066X.57.9.705
- Mento, A. J., Steel, R. P., & Karren, R. J. (1987). A meta-analytic study of the effects of goal setting on task performance: 1966–1984. Organizational Behavior and Human Decision Processes, 39(1), 52–83. https://doi.org/10.1016/0749-5978(87)90045-8
- Ordóñez, L. D., Schweitzer, M. E., Galinsky, A. D., & Bazerman, M. H. (2009). Goals gone wild: The systematic side effects of overprescribing goal setting. Academy of Management Perspectives, 23(1), 6–16. https://doi.org/10.5465/amp.2009.37007999
- Seijts, G. H., Latham, G. P., Tasa, K., & Latham, B. W. (2004). Goal setting and goal orientation: An integration of two different yet related literatures. Academy of Management Journal, 47(2), 227–239. https://doi.org/10.5465/20159574
- Seijts, G. H., & Latham, G. P. (2005). Learning versus performance goals: When should each be used? Academy of Management Executive, 19(1), 124–131. https://doi.org/10.5465/ame.2005.15841964
- Sitzmann, T., & Yeo, G. (2013). A meta-analytic investigation of the within-person self-efficacy domain: Is self-efficacy a product of past performance or a driver of future performance? Personnel Psychology, 66(3), 531–568. https://doi.org/10.1111/peps.12035
- Tubbs, M. E. (1986). Goal setting: A meta-analytic examination of the empirical evidence. Journal of Applied Psychology, 71(3), 474–483. https://doi.org/10.1037/0021-9010.71.3.474
- Vancouver, J. B., Thompson, C. M., Tischner, E. C., & Putka, D. J. (2002). Two studies examining the negative effect of self-efficacy on performance. Journal of Applied Psychology, 87(3), 506–516. https://doi.org/10.1037/0021-9010.87.3.506
- Vancouver, J. B., Weinhardt, J. M., & Schmidt, A. M. (2010). A formal, computational theory of multiple-goal pursuit: Integrating goal-choice and goal-striving processes. Journal of Applied Psychology, 95(6), 985–1008. https://doi.org/10.1037/a0020628
- Wigfield, A., & Eccles, J. S. (2000). Expectancy–value theory of achievement motivation. Contemporary Educational Psychology, 25(1), 68–81. https://doi.org/10.1006/ceps.1999.1015
- Wrosch, C., Scheier, M. F., Miller, G. E., Schulz, R., & Carver, C. S. (2003). Adaptive self-regulation of unattainable goals: Goal disengagement, goal reengagement, and subjective well-being. Personality and Social Psychology Bulletin, 29(12), 1494–1508. https://doi.org/10.1177/0146167203256921