Self-Talk and the Two Egos Bridge Bank

Verified research bank, gathered by parallel research agents; every quoted line was verification-checked at collection time. Raw reference material, not a wiki-register page.

================================================================================ LANE: motivation-split SUMMARY: The motivation-science strand is where the two egos were first named, operationalized, and measured — and it is the only literature that recorded the actual sentences each mode speaks. Nicholls (1984) supplied the root mechanism: achievement behavior aims at either DEVELOPING or DEMONSTRATING ability, and which one is live depends on whether ability is judged self-referentially (mastery gains = competence) or as normative capacity (mastery gains alone prove nothing). That single fork generates a cruel corollary for the self-regard mode — since demonstrating capacity requires achieving more with equal effort, or equal results with less effort, high effort becomes evidence of low ability, so the ego-involved inner voice must suppress or hide trying. Diener & Dweck (1978) is the strand’s empirical crown jewel for this researcher’s exact question: they continuously recorded children’s verbalizations during failure and found the two modes talk in categorically different grammars — helpless children issued ability attributions and dwelt on the CAUSE of failure, while mastery-oriented children made surprisingly few attributions at all and instead produced self-monitoring and self-instructions aimed at REMEDIES. SDT then named the voice: Ryan & Deci (2000) define introjection as “regulation by contingent self-esteem” and identify ego involvement as its classic form, which crucially reframes self-regard talk as intrapersonal control — the scoreboard is external in origin but the guard is now inside. The selfhood side is specified by Csikszentmihalyi’s flow (loss of reflective self-consciousness under clear goals, immediate feedback, and challenge-skill balance) and sharpened by Leary’s hypo-egoic model, which makes the most portable move in the whole strand: the fix is not less self-thought but a level shift from abstract/evaluative to concrete/present self-thought. Hatzigeorgiadis et al. (2011) supply the causal warrant that changing self-talk content actually moves performance (ES = .48), and Hulleman et al. (2010) supply the necessary corrective — across 91,087 participants, “outperform others” and “look good / avoid looking bad” split cleanly (r = .14 vs r = -.14), so the catching-practice must target appearance and evaluation, not comparison as such. NOTE ON METHOD: this session’s WebSearch budget was exhausted before my first query, so every work here was verified instead through OpenAlex, Europe PMC, Semantic Scholar, and OpenLibrary APIs plus direct retrieval and local text-extraction of the Ryan & Deci (2000) primary PDF; all titles, authors, years, journals, volumes, pages, DOIs, and the quoted findings/effect sizes below come from those retrievals, and no key_line is included unless I read it verbatim in a verified source.

Achievement motivation: Conceptions of ability, subjective experience, task choice, and performance — John G. Nicholls (1984) [mechanism] MECHANISM: Nicholls defines achievement behavior as directed at developing or demonstrating high rather than low ability, and shows ability can be conceived two ways. Judged against one’s own past performance, gains in mastery themselves indicate competence; judged as capacity relative to others, a gain in mastery alone indicates nothing — to demonstrate high capacity one must achieve more than others with equal effort, or equal results with less effort. He then specifies the conditions under which each conception becomes the operative goal and what subjective experience follows, deriving and testing different predictions for task choice and performance. The self-referenced conception underwrites task involvement; the normative-capacity conception underwrites ego involvement. SPECIFIC: Develop-vs-demonstrate ability as the root fork, plus the effort-ability inverse rule: under the normative conception, needing high effort is itself evidence of low capacity. That gives a precise catch-cue — any inner sentence of the form ‘if I have to work this hard at it, I must not be good at it’ is the ego-involved conception talking. GOAL: This is the definitional floor for the whole split, stated in measurable terms rather than moral ones, so selfhood-mode self-talk gets an operational test: does this sentence reference my own prior performance, or someone else’s? The effort-ability corollary is the single most useful catch, because it explains why self-regard mode quietly generates withheld effort, hidden practice, and pre-emptive excuses.

An analysis of learned helplessness: Continuous changes in performance, strategy, and achievement cognitions following failure — Carol I. Diener and Carol S. Dweck (1978) [mechanism] MECHANISM: Rather than inferring cognition after the fact, Diener and Dweck continuously monitored children’s verbalizations while failure was actually unfolding, capturing the nature, timing, and relative frequency of achievement-related thoughts. Helpless children produced the expected attributions to lack of ability and focused on the cause of failure. Mastery-oriented children made surprisingly few attributions of any kind and instead engaged in self-monitoring and self-instructions, focusing on remedies for failure. These verbal differences were accompanied by striking differences in strategy change under failure, and the authors conclude that the timing or even the occurrence of attributions — not just their content — is a critical individual difference. SPECIFIC: Attribution talk versus self-instruction talk, and the diagnostic that the mastery marker is the ABSENCE of attributions rather than the presence of better ones. The portable pair: ‘cause of failure’ talk versus ‘remedy for failure’ talk. GOAL: This is the closest thing in the literature to a direct transcript of the two egos speaking, which makes it the best template for a catching-practice: the tell is not negativity but the grammar shift into verdict-and-cause about the self. It also corrects a common error — the target is not to replace bad attributions with good ones, but to stop issuing attributions and resume issuing instructions.

Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being — Richard M. Ryan and Edward L. Deci (2000) [mechanism] MECHANISM: Within organismic integration theory, introjected regulation is defined as taking in a regulation but not fully accepting it as one’s own — a relatively controlled form in which behaviors are performed to avoid guilt or anxiety or to attain ego enhancements such as pride. Ryan and Deci restate this as regulation by contingent self-esteem, and name ego involvement as its classic form. Critically, they note that although internally driven, introjected behaviors still have an external perceived locus of causality and are not really experienced as part of the self, so introjection is grouped with external regulation as controlled motivation — external regulation being interpersonally controlled and introjection intrapersonally controlled. Identification, by contrast, reflects a conscious valuing of a goal such that the action is accepted or owned as personally important. SPECIFIC: The interpersonally-controlled versus intrapersonally-controlled distinction, and the introjected-to-identified test: does this sentence enforce a should in order to protect my worth, or express something I consciously value as mine? ‘Regulation by contingent self-esteem’ is the compact name for the self-regard voice. KEY LINE: “A classic form of introjection is ego involvement (deCharms, 1968; Nicholls, 1984; Ryan, 1982), in which people are motivated to demonstrate ability (or avoid failure) in order to maintain feelings of worth.” GOAL: This is the strand’s hinge, because it explicitly welds SDT’s introjection to Nicholls’s ego involvement and shows the self-regard scoreboard does not need an audience to operate — the internalized guard runs it. The introjected/identified pair gives a directly usable rewrite protocol: convert ‘I should, or I’m worthless’ into ‘I value this, and here is why it is mine.‘

Mindset: The New Psychology of Success — Carol S. Dweck (2006) [self-regard-promoting] MECHANISM: Dweck’s account is that a belief about the nature of ability installs a goal, and the goal installs a monologue. In a fixed mindset, people believe their basic abilities, intelligence, and talents are just fixed traits — they have a certain amount and that’s that — so the operative goal becomes looking smart and never looking dumb, which turns every task into a verdict on the person. In a growth mindset, abilities are understood as developable through effort, good teaching, and persistence, so the same task becomes information rather than judgment. Her praise findings supply the installation route: praising children’s intelligence harms motivation and harms performance, because ability praise teaches the person to protect a rating. SPECIFIC: The prove-versus-improve framing, and the praise lever: what a voice praises is what it will later have to defend. Ability praise breeds a defended rating; process praise breeds a workable practice. KEY LINE: “They have a certain amount and that’s that, and then their goal becomes to look smart all the time and never look dumb.” (Dweck’s own formulation, verified from her 2012 interview with James Morehead rather than from the book text.) GOAL: This is the popular, most transmissible version of the self-regard monologue and therefore the best vocabulary for naming it in the moment — ‘that’s a looking-smart sentence.’ The praise mechanism also extends the catching-practice inward: self-praise for being talented is not harmless, it is the deposit that the prove-voice will later be collecting on.

Flow: The Psychology of Optimal Experience — Mihaly Csikszentmihalyi (1990) [selfhood-promoting] MECHANISM: Flow is characterized by six factors: intense and focused concentration on the present moment, merging of action and awareness, a loss of reflective self-consciousness, a sense of personal control or agency, a distortion of temporal experience, and experience of the activity as intrinsically rewarding — the autotelic experience. The self-monitoring channel is not suppressed by willpower; it is crowded out, because clear proximal goals and immediate feedback consume the attentional capacity that self-evaluation would otherwise occupy, provided challenge is matched to skill. Agency is retained while reflective self-consciousness drops, which is precisely the configuration the two-egos distinction is after: the instrument is fully present while the image-monitor is offline. SPECIFIC: ‘Loss of reflective self-consciousness’ as a named, measurable component — and the three engineering conditions that produce it: clear goals, immediate feedback, and challenge-skill balance. Autotelic experience is the term for the activity carrying its own reward rather than reporting to a scoreboard. GOAL: Flow gives the selfhood mode a positive specification rather than a mere absence, and shows it is designed into the task rather than talked into existence — if self-regard chatter keeps intruding, the first suspect is a missing feedback loop or a challenge-skill mismatch, not insufficient discipline. The retained sense of control alongside dropped self-consciousness is the proof that quieting self-regard does not cost you judgment or agency.

Hypo-egoic self-regulation: Exercising self-control by diminishing the influence of the self — Mark R. Leary, Claire E. Adams, and Eleanor B. Tate (2006) [selfhood-promoting] MECHANISM: Leary and colleagues argue that intentionally trying to control one’s behavior sometimes reduces the likelihood of achieving one’s goals, and examine hypo-egoic states — including flow, deindividuation, and transcendence — characterized by diminished self-awareness together with concrete, present-focused thinking. They identify two routes for deliberately cultivating such a state: reducing self-awareness through practice and automation, or increasing concrete self-thoughts via mindfulness and concrete mental frameworks. The second route is the substantive one for self-talk, because it does not require silencing the inner voice at all — it changes the level of abstraction at which the voice addresses the self. The model predicts people should strategically adopt hypo-egoic regulation exactly when deliberate self-control would impair performance. SPECIFIC: The abstract-versus-concrete axis of self-referential thought. You do not stop thinking about yourself; you drop the level from abstract and evaluative (‘am I any good at this’) to concrete and present (‘what is my hand doing right now’). Two named cultivation routes: automation through practice, and concreteness through mindfulness. GOAL: This is the most directly actionable protocol in the strand for cultivating selfhood-mode self-talk, because it replaces the unworkable instruction ‘stop being self-conscious’ with a level shift you can actually execute mid-task. It also explains why task-focused self-talk works: concreteness is the active ingredient, and abstraction is the tell that self-regard has taken the microphone.

Self-Talk and Sports Performance: A Meta-Analysis — Antonis Hatzigeorgiadis, Nikos Zourbanos, Evangelos Galanis, and Yannis Theodorakis (2011) [mechanism] MECHANISM: Working from the premise that what people think influences their actions, the authors synthesized 32 studies yielding 62 effect sizes and found a moderate positive overall effect of self-talk interventions on performance (ES = .48). Effects were larger for tasks requiring fine motor control than for gross movements, and larger for novel than for well-practiced tasks. Instructional self-talk outperformed motivational self-talk for fine motor tasks. Interventions that included structured self-talk training produced better results than those that simply assigned cues. SPECIFIC: The instructional-versus-motivational self-talk distinction, with ES = .48 overall and instructional winning on fine and novel tasks. Plus the training finding: self-talk works as a trained skill, not as an instruction issued once. GOAL: This is the causal warrant the whole project rests on — deliberately changing self-talk content measurably changes performance, so a catching-and-replacing practice is not merely expressive. Instructional self-talk is the operational form of the task-involved register (cues about execution, not exhortations about the self), and its advantage on novel and fine-grained work is exactly where the self-regard voice is loudest.

A meta-analytic review of achievement goal measures: Different labels for the same constructs or different constructs with similar labels? — Chris S. Hulleman, Sheree M. Schrager, Shawn M. Bodmann, and Judith M. Harackiewicz (2010) [against] MECHANISM: Across 243 correlational studies and 91,087 participants, the authors coded achievement goal items by their actual conceptual content rather than by their scale labels. Goal-outcome and goal-goal correlations differed significantly depending on which scale was chosen and which individual items were used. Decisively for the two-egos mapping, performance-approach scales worded in normative terms correlated positively with performance (r = .14), while performance-approach scales worded in appearance terms correlated negatively (r = -.14). The authors conclude that researchers apply identical labels to conceptually distinct constructs, and that this mismatch between conceptual and operational definitions undermines theory development and practical application. SPECIFIC: The normative/appearance split inside ‘performance goals,’ with opposite-signed correlations (r = .14 versus r = -.14) attached to it. Wanting to outperform is empirically not the same construct as wanting to look competent or avoid looking incompetent. GOAL: This is the necessary correction to a too-clean mapping: comparison per se is not the pathogen, and a catching-practice tuned to flag every competitive thought will over-fire and cost the researcher real drive. The precise target is appearance and evaluation content — ‘how will this look, what will they conclude about me’ — while normatively framed ambition can sit inside selfhood mode without contradiction.

LANE: self-worth-machinery SUMMARY: This strand traces the self-worth machinery: how staking worth in a domain, and comparing within it, converts inner speech from task-talk into rank-talk — and what the non-evaluative substitute looks like. The literature converges on a single structural claim that maps almost exactly onto the two egos: self-regard mode is not a trait but a STATE, induced by conditions, and the condition that induces it is having self-worth contingent on the outcome at hand. Nicholls (1984) names the two states directly — ego involvement (ability judged against others, self as object of attention) vs task involvement (ability judged against one’s own prior mastery, task as object) — which is the cleanest conceptual twin available. Crocker’s program supplies the architecture underneath: worth is staked domain by domain, those domains array on a measurable external-to-internal simplex, and the cost of the pursuit falls on learning, autonomy, and self-regulation. Tesser explains why the spike is sharpest exactly where identity is staked, and names the three tells (distance from the peer, redefine the domain away from self, or diminish them). Two works give the actual install and the actual transcript: Mueller & Dweck show one sentence of person-directed praise flipping goal orientation, attribution, and even honesty about scores; Diener & Dweck recorded children thinking aloud through failure and found the two modes verbatim — attribution talk about the self vs self-monitoring and self-instruction about the task. That 1978 study is the strongest single artifact in this strand for the researcher’s question, because it is close to a literal transcript of both modes. The meta-analytic warrant for treating comparison as the thing to catch is Gerber/Wheeler/Suls: upward target selection is the default and contrast (self-deflation) is the dominant reaction at d ≈ -0.65 to -0.75. Neff supplies the substitute — a way of relating to oneself that requires no positive self-evaluation, hence no comparison and no domain to win in, with the empirical payoff that it delivers self-esteem’s affective benefits while correlating negatively with exactly the rank-monitoring cluster (social comparison, public self-consciousness, self-rumination, anger). The dissent is real and worth keeping: Harackiewicz et al. argue the damage sits in the AVOIDANCE flavor (“don’t look stupid”), not in every comparison thought, and that approach-flavored performance goals can coexist with mastery goals. A catching-practice built without that caveat will over-fire on ordinary competitive noticing. Two verification notes: Kross was excluded per instruction, and Festinger (1954, Human Relations 7(2), 117-140 — verified) is folded into the Gerber entry rather than given its own slot, since that meta-analysis IS the six-decade test of his propositions. All eight works were verified against OpenAlex, Crossref-linked DOIs, and abstract text; two machine-reconstructed abstracts (Diener & Dweck, Neff & Vonk) initially inverted a polarity and were re-verified against a second source before use.

The Costly Pursuit of Self-Esteem — Jennifer Crocker & Lora E. Park (2004) [mechanism] MECHANISM: Self-esteem matters as a PURSUIT, not as a level. People stake their worth in particular domains (contingencies of self-worth), and when a staked contingency is engaged, attention converts to self-validation goals — proving worth rather than doing the thing. Crocker & Wolfe’s earlier architecture (Psychological Review, 2001, 108(3), 593-623) formalizes the domain structure, and the Contingencies of Self-Worth Scale (Crocker, Luhtanen, Cooper & Bouvrette, JPSP, 2003, 85(5), 894-908) validates seven domains — academics, appearance, others’ approval, competition, family support, God’s love, virtue — that form a simplex running from EXTERNAL to INTERNAL sources of esteem. The documented costs of the pursuit land on learning, autonomy, relationships, and self-regulation, with short-term emotional benefits outweighed by long-term costs. SPECIFIC: ‘Contingencies of self-worth’ plus the external-to-internal simplex. The diagnostic is two-step: which domain is your worth staked in, and does that domain’s standard sit outside you (approval, appearance, competition) or inside (virtue)? Externally contingent domains are where rank-monitoring self-talk fires. KEY LINE: the importance of self-esteem lies more in how people strive for it rather than whether it is high or low GOAL: Gives the catching-practice its map: self-regard self-talk is domain-specific and predictable — it fires when a staked contingency is engaged, not uniformly. The selfhood move it implies is migrating one’s staked standards from external contingencies to internal ones.

Achievement motivation: Conceptions of ability, subjective experience, task choice, and performance — John G. Nicholls (1984) [mechanism] MECHANISM: Achievement behavior is directed at developing or demonstrating high rather than low ability — and ‘ability’ can be construed two incompatible ways. In the LESS DIFFERENTIATED conception, ability is judged against one’s own prior mastery and learning; effort and ability are not separated. In the DIFFERENTIATED conception, ability is judged relative to others, and effort becomes an inverse signal of ability (needing to try hard means you lack it). Nicholls names the first state TASK INVOLVEMENT and the second EGO INVOLVEMENT, and treats them as states the same person moves between as conditions change — in ego involvement the self is the object of attention; in task involvement the task is. SPECIFIC: Task involvement vs ego involvement — two inducible states, not two personality types. Plus the effort-ability inverse rule: under ego involvement, having to try hard is itself read as evidence of low ability, which is what makes visible effort feel humiliating. KEY LINE: directed at developing or demonstrating high rather than low ability GOAL: The cleanest conceptual twin of the two egos in the empirical literature, and the most useful reframe for the researcher’s design goal: selfhood is a state to be induced by arranging conditions, not a trait to be acquired. Also supplies a specific tell — inner speech that treats effort as incriminating is ego-involved by definition.

An analysis of learned helplessness: Continuous changes in performance, strategy, and achievement cognitions following failure — Carol I. Diener & Carol S. Dweck (1978) [selfhood-promoting] MECHANISM: Children’s verbalizations were recorded CONTINUOUSLY as they were moved from a run of success into failure, which makes this one of the few studies that captured the two modes of inner speech as they diverged in real time. Helpless children made the expected attributions for failure to lack of ability. Mastery-oriented children made surprisingly few attributions and instead engaged in self-monitoring and self-instruction — strategy talk, effort statements, task focus. The behavioral consequence tracked the talk: helpless children’s problem-solving strategies deteriorated under failure while mastery-oriented children’s held or improved. SPECIFIC: The two verbalization profiles as directly observable categories: ATTRIBUTION TALK (a verdict about the self — ‘I’m not good at this’) vs SELF-MONITORING AND SELF-INSTRUCTION (a directive about the task — ‘slower, check that one again’). Note the finding that mastery-oriented children did not counter-attribute positively; they simply did not run the attribution routine at all. KEY LINE: made surprisingly few attributions but instead engaged in self-monitoring and self-instructions GOAL: The closest thing in this literature to a transcript of both modes, which makes it the most portable practice target in the strand: the goal is not to replace a negative verdict with a positive verdict, but to stop issuing verdicts and issue instructions instead. Catching is easy to operationalize — any failure sentence whose grammatical subject is ‘I’ as a kind of person.

Praise for intelligence can undermine children’s motivation and performance — Claudia M. Mueller & Carol S. Dweck (1998) [self-regard-promoting] MECHANISM: Six experiments varied a single line of feedback after success: praise for intelligence versus praise for effort. Children praised for intelligence subsequently preferred PERFORMANCE goals over LEARNING goals, showed less persistence and less task enjoyment after setbacks, attributed failure to low ability, and performed worse than they had before. Ability praise also induced a fixed view of intelligence, whereas effort praise induced a malleable one. In a further result, children praised for intelligence were more likely to misrepresent their scores when reporting to peers. SPECIFIC: Ability/person praise vs effort/process praise. The score-misrepresentation result is the sharpest single datum in the strand: once worth is pinned to a trait verdict, managing the appearance of the number becomes rational, and honesty degrades. GOAL: This is the INSTALL mechanism — evaluative feedback aimed at the person rather than the process trains self-regard-mode self-talk, and it takes only one sentence. For catching, it names the exact grammatical form to flag: person-level verdicts, whether flattering (‘I’m smart’) or punishing (‘I’m a fraud’), since both pin worth to a trait and both produce the same downstream fragility.

A social comparison theory meta-analysis 60+ years on — J. P. Gerber, Ladd Wheeler & Jerry Suls (2018) [self-regard-promoting] MECHANISM: A meta-analysis of six decades of research testing Festinger’s original propositions (A Theory of Social Comparison Processes, Human Relations, 1954, 7(2), 117-140 — verified). Two results matter here. On TARGET SELECTION: when comparison options exist, people show a strong preference for UPWARD targets absent threat, shifting toward downward targets when threatened. On REACTION: CONTRAST is the dominant response rather than assimilation, with contrast effect estimates in the range of roughly d = -0.75 to -0.65, while assimilation effects are notably weaker and depend on how they are measured. SPECIFIC: Contrast is the default reaction to comparison (d ≈ -0.65 to -0.75), and upward is the default target choice. Combined: the comparison you spontaneously reach for is the one most likely to leave you diminished — and the shift to downward comparison is itself a threat signal worth noticing. GOAL: Supplies the empirical warrant for treating comparison sentences as the primary catch target rather than as neutral information-gathering. It also gives the practice a live diagnostic: catching yourself reaching for a DOWNWARD comparison is evidence that a threat has already registered, making it an early-warning marker rather than merely a bad habit.

Toward a Self-Evaluation Maintenance Model of Social Behavior — Abraham Tesser (1988) [mechanism] MECHANISM: Another person’s good performance feeds two opposed processes. REFLECTION raises your self-evaluation by association with a close other’s success; COMPARISON lowers it. Which process dominates is governed by RELEVANCE — whether the performance domain is central to your self-definition — moderated by CLOSENESS and RELATIVE PERFORMANCE. When a close other outperforms you in a self-defining domain, comparison dominates, and people restore self-evaluation through three moves: reducing closeness to the person, reducing the domain’s relevance to their self-definition, or acting to impair the other’s performance. SPECIFIC: Relevance × closeness × relative performance → reflection or comparison, plus the three restoration moves. The counterintuitive prediction is that closeness AMPLIFIES rather than buffers the threat when the domain is self-defining — friends hurt more than strangers. GOAL: Explains why self-regard self-talk spikes precisely where identity is staked, and names three behavioral tells that are easier to catch than the thought itself: the urge to distance from a peer, the urge to declare the domain unimportant, and the urge to diminish their work. The selfhood alternative it implies is holding the domain as work you do rather than as a definition of who you are — which is exactly the relevance variable, and it is the one variable under voluntary control.

Self-Compassion: Stop Beating Yourself Up and Leave Insecurity Behind — Kristin Neff (2011) [selfhood-promoting] MECHANISM: Self-compassion has three components — self-kindness, common humanity, and mindfulness — and its structural feature is that it requires no positive self-EVALUATION. Because it does not depend on judging oneself favorably, it needs no comparison target and no domain to win in. The empirical anchor is Neff & Vonk (Journal of Personality, 2009, 77(1), 23-50): in Study 1 (N = 2,187), self-compassion predicted more STABLE feelings of self-worth than self-esteem and showed stronger NEGATIVE associations with social comparison, public self-consciousness, self-rumination, anger, and need for cognitive closure, while self-esteem — but not self-compassion — was positively associated with narcissism. In Study 2 (N = 165), the two were statistically equivalent predictors of happiness, optimism, and positive affect. SPECIFIC: Evaluative vs NON-EVALUATIVE self-relating. The load-bearing result is the dissociation: self-compassion matches self-esteem on the affective upside while inversely tracking the entire rank-monitoring cluster that self-esteem drags along. Practically, common humanity is the component that does the anti-comparison work — it reframes failure as evidence of membership rather than of rank. GOAL: This is the substitute, and a catching-practice needs one — suppressing a self-regard sentence leaves a vacuum that the same sentence refills. It supplies an inner-speech form that is neither self-attack nor self-promotion, which is what makes it compatible with selfhood-as-instrument rather than a softer variety of scoreboard.

Revision of achievement goal theory: Necessary and illuminating — Judith M. Harackiewicz, Kenneth E. Barron, Paul R. Pintrich, Andrew J. Elliot & Todd M. Thrash (2002) [against] MECHANISM: Directly contests the claim that performance and comparison framing are uniformly corrosive, on three grounds: approach strivings must be separated from avoidance strivings; performance-approach goals have genuine positive potential; and performance-approach goals can COMBINE with mastery goals to promote optimal motivation. Under the resulting 2×2 framework (mastery/performance × approach/avoidance), the four goals show distinct empirical profiles, and much of the damage historically attributed to ‘performance goals’ localizes to the performance-AVOIDANCE cell. In the multiple-goals perspective the authors defend, performance-approach goals and mastery goals predict different outcomes and both contribute unique variance. SPECIFIC: The 2×2 achievement goal framework and the multiple-goals perspective. The operative cut: ‘do better than them’ (performance-approach) behaves differently from ‘don’t be shown up’ (performance-avoidance), and only the latter reliably carries the costs. GOAL: Forces precision on the catching-practice and prevents it from over-firing. The pattern worth catching is AVOIDANCE-flavored rank-monitoring — image protection, fear of exposure — not every instance of noticing where one stands. It also warns against a purity model in which any competitive thought counts as a lapse, since the evidence supports mastery and performance-approach goals coexisting.