Self priming from distinctive and caricatured faces

Burton, Bruce & Johnston (1990) have recently presented an interactive activation and competition (IAC) model of face recognition. Within this architecture they present accounts of repetition priming, semantic priming and distinctiveness effects with faces. This model predicts that a short-lived priming effect should extend from a person's face to their name, this is called 'self priming'. The model also predicts that the distinctiveness of the person's face should interact with the amount of self priming found. We tested this prediction with distinctive and typical face sets (Expt 1); results confirmed the prediction. From a hybrid of Valentine's (1991 a) multidimensional space framework and Burton et al.'s IAC model we derived a second prediction, that a caricature of a face should produce more self-priming than the veridical or an anti-caricatured (shifted towards an average face) representation of the face. Using continuous-tone caricatures of photographic quality we tested this prediction in Expts 2 and 3; results again confirmed the prediction. This finding is consistent with the idea that caricaturing works by enhancing the face's distinctiveness.