Effect of self-discrepancy on specificity of autobiographical memory retrieval

Abstract

Research using a cue word paradigm has consistently shown that depression, in both adults and adolescents, is associated with difficulties in retrieving specific autobiographical memories. Inspired by previous work stating that depressed feelings are related to a perceived discrepancy between attributes of the actual and the ideal self, the present study aimed to investigate the hypothesis that cues bringing discrepancies between the actual and ideal selves to the foreground might promote or facilitate the recall of overgeneral (instead of specific) autobiographical memories. In two studies adolescents provided autobiographical memories in response to 10 high-discrepant and 10 low-discrepant words. As predicted, results in both studies showed an effect of cue word discrepancy on the specificity of autobiographical memories such that participants retrieved a smaller proportion of specific and a greater proportion of overgeneral memories in response to high-discrepant words as compared to low-discrepant words. The findings are discussed in terms of the self-memory system (SMS) as a conceptual framework of autobiographical memory (Conway & Pleydell-Pearce, 2000).

Keywords:

• Autobiographical memory
• Self-discrepancy
• Depression

ACKNOWLEDGEMENTS

This research was supported by a Research Programme of the Research Foundation-Flanders (FWO) (G.0339.08) awarded to the last author. The authors would like to thank Koen Vanbrabant for his comments on a previous version of the manuscript.

Notes

¹Number and proportion of specific and categoric memories are not independent, given the nature and scoring of the Autobiographical Memory Task (AMT). As a consequence we use the terms overgeneral memory (OGM) and reduced memory specificity (rAMS) interchangeably to indicate lower levels of specific recall as well as higher levels of categoric recall.

²A post-study was conducted (based on the valuable suggestion of an anonymous reviewer) wherein 56 non-clinical last-year high school students rated the HD and LD cues in alternating order for the extent to which each word describes their “ideal” and “current” self. These students had not previously participated in Study 1 or Study 2. We calculated the degree of discrepancy in order to compare the two sets of words. Results showed that the two sets differed significantly with respect to the extent they promoted actual vs ideal self-discrepancy, with greater current–ideal discrepancies for the HD vs LD words, t(55) = 7.73, p<.001.