Abstract
Chun (1997) examined repetition blindness (RB) within a letter-number attentional blink (AB) paradigm. Supporting the conclusion that the AB and RB reflect different mechanisms, he found that some manipulations selectively reduced the AB while others selectively reduced the RB. This conclusion appears to contradict the recently proposed 'perceptual wink' model of Rusconi and Huber (2017). On this account, both the AB and RB reflect perceptual habituation: RB is habituation for a character's identity (i.e., which letter or number), whereas the AB is habituation for the category of the character (i.e., a failure to perceive that the second target belonged to the target category). Because different representations underlie each effect, the same mechanism of perceptual habituation can explain the dissociation reported by Chun. To test this account, we ran a series of three experiments, manipulating the degree to which targets were readily identified as belonging to the target category. Each experiment used a between-subjects manipulation in which one group received consistent mapping (CM), with a set of characters consistently assigned to the target category across trials, whereas the other group received varied mapping (VM) by varying the assignment of characters to the target category versus the distractor category. All experiments examined the AB and RB simultaneously, but unlike Chun, these experiments used multiple choice testing to assess any strategy or bias present in the decision process. Across all three experiments, VM produced weaker AB effects as compared to CM. Numerically, RB was also weakened by the CM/VM manipulation, as expected if the loading of a character's identity into working memory requires both identifying a character (the RB) and recognizing it as being a target (the AB). The results of all three experiments were well accounted for by the perceptual wink model using the same habituation parameters.
Meeting abstract presented at VSS 2018