Repetition blindness for pseudoobject pictures

In this study the nature of type activations that underlie repetition blindness (RB) was addressed. According to the token individuation hypothesis put forward to explain RB, both instances of a repeated stimulus make contact with the mental representation, or type, for that stimulus. In the resulting confusion only the first stimulus is encoded as an episodic instance or token. Type representations have traditionally been thought of as preexisting and are often linked within a network of nodes. The authors developed and tested a picture frequency task, which does not require stimulus naming, and used it to examine repetition performance for unfamiliar nonobject pictures. RB was found for these stimuli, despite the fact that they had no prior phonological or semantic representation. These results suggest that the reactivation of a newly formed visual type is sufficient to produce RB. Implications for the role of types in the token individuation hypothesis are discussed.