In analyzing the role of the self-concept in judgements of similarity between persons, authors have made use of two notions: “the self as a habitual reference point” and “the self as a prototype”. We argue that these two notions must be kept distinct, and present four studies aimed at examining the hypothesis that the self could be a prototype in the strong sense of the term, namely a central exemplar in a category structured by a typicality gradient. The results from these studies show that the self is a prototype only in categories of positively valued persons. This prototype-like function does not seem to depend upon contextual priming.

No abstract