THEORY-074: Implicit accessibility of the true self-concept is not what self-report scales of authenticity or of need satisfaction measure, though those two self-reports are not shown to be separate constructs

What was asked, and why the claim is narrower

The question was whether self-reported authenticity, implicit true-self accessibility and felt autonomy are one construct. The paper that bears on it measures all three in one sample, but not in a way that separates all three. It separates the implicit measure from the two self-reports. It does not separate the self-reports from each other, and its autonomy items include an authenticity item. The title says what the evidence reaches, and leaves the rest as the open half.

All three constructs belong to self-governance in its empirical form: how far what one does is felt as one’s own. The implicit measure is a measure of a self-concept, which is why the account also carries self. No unity of experience, narrative or behaviour is at issue.

Source

• Schlegel, Hicks, Arndt & King (2009), LIT-606, read in full in NOTE-469. Table 4 was checked against the PDF text for this account.

What was actually shown

The measures. True-self accessibility (TSCA) is the reversed residual of a person’s log reaction time to “me” judgments for her own true-self traits, regressed on her speed for control traits. Authenticity is the 45-item AI-3 of Kernis and Goldman. Need satisfaction is Gagné’s 21-item Basic Psychological Needs Scale, averaged across competence, relatedness and autonomy. Its autonomy items include “I feel like I am free to decide for myself how to live my life” and “I feel like I can pretty much be myself in my daily situations” (LIT-606, p. 476).

Study 4 (N = 124), Table 4. TSCA correlated .05 with authenticity and .20 with need satisfaction. Authenticity and need satisfaction correlated .52. All three correlated with presence of meaning in life (TSCA .30, need satisfaction .33, authenticity .50). In a regression with nine covariates, authenticity (β = .34) and TSCA (β = .19) each predicted meaning with the other controlled. Studies 1 and 2 found the same for TSCA against need satisfaction (TSCA β = .29 and .24; need satisfaction β = .37 and .32).

The authors’ reading. True-self accessibility “resonates with previous notions such as the awareness aspect of authenticity … or basic need satisfaction”, and the data bear on “whether these various conceptualizations in fact tap the same underlying construct. Results suggest that true self-concept accessibility differs from these constructs in that each maintained independent relationships to meaning in life. Of course, these variables also differ methodologically” (pp. 487–488).

So the separation of the implicit measure from each self-report rests on near-zero or small correlations and on independent prediction of a third variable, in one sample for authenticity and three for need satisfaction. It is confounded with method, as the authors say.

What this does not say

• It does not say authenticity and autonomy are distinct. Their r = .52 is compatible with one construct measured with error and with two related ones. The autonomy subscale was not analysed alone, and it contains an authenticity item. This is the half of the original question the record cannot answer.
• It does not say the implicit measure is the better one. TSCA predicts meaning with β ≈ .2–.3 in modest samples, and its priming evidence comes from subliminal-priming experiments of a period whose replication record has been widely questioned (NOTE-469, Limitations; that judgement is the note’s, not the paper’s).
• It does not say what a “true self” is. The paper measures the person’s avowed true self-concept, and its authors say that this “may not coincide with a person’s ‘real’ true self” (p. 475).
• It does not reach the philosophical constructs. Authenticity in Taylor’s sense (LIT-594, unread) is a moral ideal, not a trait score.

Connections

• THEORY-054 (autonomy is graded integration). SDT’s own current definition makes autonomy “self-endorsed and authentic” volition (LIT-561, as THEORY-054 quotes it), which writes authenticity into autonomy. This account does not refute that: it finds the two self-reports correlated at .52 and leaves their identity open. It does show that an implicit measure of the self they both refer to is not what either self-report tracks. The paper’s overlap index, how far one’s everyday and true self-concepts share traits, correlated with authenticity, need satisfaction and TSCA and predicted meaning on its own (β = .21; NOTE-469). It is a candidate for the measure of integration apart from endorsement that THEORY-054’s promote_when asks for, though the paper does not frame it so.
• THEORY-077 (self-concordant goals get sustained effort). Self-concordance is a fourth self-report in this family, of reasons for goals rather than of the self. No paper the record holds correlates it with the AI-3 or with TSCA.
• THEORY-029 (no attitude owns a motive by its order alone). Every measure here is of avowed ownership. None asks whether the avowal could itself be alienated, so none bears on that account.
• Kernis and Goldman (LIT-574) and Wood et al. (LIT-592) are the two authenticity scales the field uses, both filed unread. Wood et al.’s abstract reports discriminant validity from the Big Five but not from need satisfaction; neither reading would settle the open half without the design in promote_when.