Self-deception and failure to modulate responses despite accruing evidence of error

Abstract

Two studies assessed performance on a gambling-type card playing task ( Newman, Patterson, & Kosson, 1987) by males defined as high or low in self-deception. Monetary success in this task depends upon the ability to modulate reward-seeking responses, by attending to information indicative of task-failure. In Study 1, 28 13-year-old boys categorized as high in self-deception using Eysenck’s Junior Lie Scale ( Eysenck & Eysenck, 1975), played more cards and won significantly less money than 143 categorized as low in self-deception. Study 2 replicated these findings in a sample of 42 male Harvard undergraduates defined as high or low in self-deception using Eysenck’s Lie scale ( Eysenck, Eysenck, & Barrett, 1985) and the Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991). Also, a higher proportion of high self-deceivers played until the end of the task in both samples, thereby losing all their money, despite the fact that 19 of the last 20 cards were losing. These findings support a model of self-deception as ignoring evidence of error and reinforce the argument that self-deception may be maladaptive.