How Implicit–Explicit Consistency of the Intelligence Self–Concept Moderates Reactions to Performance Feedback

Abstract

At present, it is not well understood which individual characteristics determine whether individuals will benefit or suffer from feedback about suboptimal performance. Three separate studies tested the idea that individuals' reactions to (neutral versus negative; positive versus negative) feedback as either improving or deteriorating performance in intelligence tests depended on their explicit and implicit self‐concept of intelligence (eSCI and iSCI). In all studies (Study 1, N = 177; Study 2, N = 165; Study 3, N = 132), persons with low eSCI/high iSCI showed a boost in performance after negative feedback, whereas persons with any other combination showed a decline in performance. The studies also provide an explanation for these effects in terms of mediating mechanisms. Whereas the performance boost of individuals with low eSCI/high iSCI can be best explained by achievement‐related reactance (Studies 2 and 3) and decrease in ruminative thinking (Study 3), the performance declines in the other configurations were mediated by decreased achievement motivation, increased frustration and increased outrage (Study 3). All these mediating mechanisms proved independent and incremental in a bootstrapping‐based moderated meditation analysis. The results are discussed in light of the construct validity of our Implicit Association Test (IAT) and IATs in general. Copyright © 2013 John Wiley & Sons, Ltd.

Discrepancies between implicit and explicit self-concepts of intelligence: relations to modesty, narcissism, and achievement motivation

“…Descriptive statistics, internal consistencies (Cronbach’s alpha), and intercorrelations for all measures are presented in Table 1 . Similar to the results of Dislich et al (2012) and Gerstenberg et al (2013) , the SCI-IAT and the eSCI were not found to be significantly correlated.…”

“…High eSCI was generally related to higher performance, but this relation was moderated by the iSCI. Whenever the implicit association contradicted the eSCI, subsequent performance was impaired [a pattern, fully replicated more recently by Gerstenberg et al (2013)]. More specifically, a fragile SCI (i.e., an explicit claim of being intelligent that was accompanied by a weak automatic association of the self with intelligence; high eSCI, low iSCI) was associated with reduced performance on intelligence tests compared to a consistently high SCI (high eSCI, high iSCI).…”

“…Its concurrent validity with other standard measures of crystallized intelligence has been provided by Lehrl et al (1991) . We decided to use the MWTA as it was also used in the studies by Dislich et al (2012) and Gerstenberg et al (2013) . The MWTA consists of 37 items of increasing difficulty.…”

Using Indirect Measurement Tasks to Assess the Self–Concept of Personality: A Systematic Review and Meta–Analyses

“…Future studies are needed to further investigate on which boundary conditions—and their interactions—the relation between implicit personality attributes and criterion measures depend. These boundary conditions include person‐related variables, situational variables and characteristics of the indirect measurement tasks (Friese, Hofmann, & Schmitt, ; Gerstenberg et al, ; Gschwendner, Hofmann, & Schmitt, ; Gschwendner et al, ; Schmidt et al, ). We agree with Gerstenberg and her colleagues (Gerstenberg et al, ) that ‘unless the relations between implicit constructs (and their measures) with these boundary conditions (and their interactions) are included in a data model, the predictive potential of implicit constructs and the validity of their measures will be underestimated and misjudged’ (p. 253–254).…”

Untitled

“…They used intelligence tests and demonstrated that individual reactions to feedback depend on participants' explicit and implicit self-concept of intelligence. After negative feedback, participants with low explicit and high implicit self-concepts demonstrated an increase in performance, whereas the performance of other participants was hindered (Gerstenberg et al, 2013).…”