Creativity, Psychopathology, and Emotion Processing: A Liberal Response Bias for Remembering Negative Information is Associated with Higher Creativity

Abstract

To what extent do more creative people process emotional information differently than less creative people? This study examined the role of emotion processing in creativity and its implications for the creativity–psychopathology association. A total of 117 participants performed a memory recognition task for negative, positive, and neutral words; they also completed measures of emotion processing, creative achievement, and divergent thinking. Self-reported high creative achievement levels and better performance on divergent thinking tasks were associated with greater perceptual sensitivity for positive words and a more liberal response bias for negative words. Furthermore, mediational analyses revealed a relationship between creativity, emotional repair, and response bias for negative stimuli, whereby a low ability to repair negative emotions served as a mediator between creativity and response bias, with response bias predicting creativity. The significance of these findings is discussed in the context of psychopathology and its possible connection to creativity.

Acknowledgments

This research was presented at the 12th Annual Meeting of the Society for Personality and Social Psychology in San Antonio, Texas in January 2011. We thank Kofi Anim for assistance in data collection.

Notes

¹Note that we do not interpret this evidence to advocate a strong or exclusive association between mental illness and creative achievement; both constructs are too complex to be reduced to a simple relationship; moreover, the study of both topics involves many methodological and conceptual complications that contribute to the complexity (Silvia & Kaufman, 2010; see also, e.g., Prentky, 2000–2001; Richards, 2000–2001; Sass, 2000–2001; Schlesinger, 2003; Schuldberg, 2000–2001). Empirical findings that some subclinical traits, such as schizotypy, appear to be associated with some measures of creativity do not imply that creative people are full-blown schizophrenics (or vice-versa). Along these lines, Sawyer (2006) downplayed the relation between creativity and mental illness even in such luminaries as mathematician John Nash and writer Sylvia Plath, arguing that “most people afflicted with mental illness believe that their disease interferes with their creativity” (p. 87).

²The correlation between divergent thinking and self-reported achievements obtained in this study was somewhat lower than that reported by Carson, Peterson, and Higgins (2005), though still positive and statistically reliable. (Carson and colleagues found a correlation of .47 among Harvard undergraduates.)

³A separate table reflecting negative word valence as reference point is not reported since all other results were identical to those in Table 2.

⁴Because false alarm rates did not meet the assumption of homoscedasticity, and ln-transformation of the scores was not able to adjust for heteroscedasticity, no regression analyses were performed for the false alarm rates. Statistical analyses for miss rates were also omitted because the miss rates are mirror image of hit rates (one minus hit rate) and thus yield the same statistical and conceptual results.

⁵The meditation analyses for divergent thinking scores were based only on the sample of 106 participants for whom there were complete data. That is, pairwise deletion was used to test for mediation because combining the results of multiply imputed files for meditation results involves enormous analytical complications. The direction and significance of correlation between negative response bias and divergent thinking scores was not altered after performing multiple imputation, as indicated by reliable between-subject main effect of liberal response bias for negative words (β = −.059, p < .05). Consequentially, testing for a mediation relationship based on pairwise deletion is unlikely to cause biased results in this case.