Illusory correlation and cognitive processes: A multinomial model of source-monitoring

95 The formation of illusory correlation is a mechanism of stereotype development which has largely been studied in adults, but until recently has not received attention from a developmental standpoint. Illusory correlation refers to an erroneous inference about the relationship between the two categories of events that are not correlated or are correlated to a lesser degree than one perceives (Chapman, 1967). In the classic illusory correlation paradigm (Hamilton & Gifford, 1976, Experiment 1) people read 26 statements describing the behaviors of individuals belonging to a group (A) and 13 statements describing behaviors of individuals belonging to another group (B). Although the ratio of favourable to unfavourable behaviors was the same for both groups, people perceived an association between the minority group (numerically smaller) and behaviors occurring less frequently (undesirable behaviors). Specifically, an illusory correlation was observed on three measures: a) Group B received less favourable evaluative ratings, b) an inflated number of previously presented negative behaviors was assigned to Group B in the assignment task, and c) the frequency of negative behaviors in Group B was overestimated, whereas the estimated frequency of negative behaviors in group A was more correct. Distinctiveness-based account. Hamilton and Gifford (1976) explained the effect of illusory correlation in terms of the biased encoding of infrequently occurring information. The co-occurrence of infrequent, and therefore distinctive, events was supposed to draw particular attention and to lead to better encoding. That is, members of the minority Group B performing undesirable behaviors should be particularly salient, and the increased attention to these distinctive and paired events should result in deeper encoding of the relationship between the minority group and negative behaviors. As a result, these events are assumed to be particularly well represented in memory, leading to the enhanced availability at retrieval (Hamilton, Dugan, & Trolier, 1985) and, thereby, to biases in judgments (Tversky & Kahneman, 1973). Illusory correlations in children. Despite two decades of research on illusory correlation in adults and the important implications of this phenomenon for understanding the development of stereotypes, research on the development of illusory correlation in children has just begun to emerge. To investigate illusory correlations in children, Primi and Agnoli (2002) followed Hamilton and Gifford’s paradigm but adapted it to the capacities of children. Participants were shown drawings of 15 positive behaviors and 6 negative behaviors performed by members of a majority group, called Pines, or a minority group, called Firs. There were 10 descriptions of desirable behaviors ascribed to members of Group Pines, whereas 4 statements specified undesirable behaviors of Group Pine members; in Group Firs, there were 5 desirable and 2 undesirable behaviors; thus, the ratio Illusory correlation and cognitive processes: A multinomial model of source-monitoring