Abstract
Arbitrarily associating stimuli with participants themselves improves performance relative to stimuli associated with other people(Sui & Humphreys, 2012). An account of this self-relevance effect is that self-related information is intrinsically rewarding(Northoff & Hayes, 2011). But while reward association modulates visual attention(Anderson et al, 2011), it remains unclear whether self-association modulates it in the same manner. In the current study, we examined if self-associated stimuli modulate attentional capture. We used the shape-label matching task devised by Sui & Humphreys(2012) in the training phase, and the additional singleton task(e.g., Anderson et al, 2011) in the test phase. In the training phase, two shapes(square, diamond) were given varying social context labels; self, and other. After memorizing the pairs, participants engaged in the shape-label matching task. In this task, a shape and a label were presented together, and participants had to respond whether the shape-label pair was correctly assigned or not. Then, participants engaged in the additional singleton task. Participants search for a color singleton target among non-target items(eg. one orange circle among green circles), and respond whether the line segment included in the color singleton target is vertical or horizontal. Three distractor conditions were used:(1) a neutral condition in which no distractor with the unique shape was presented, (2) a self-associated condition in which a self-associated shape singleton distractor was presented, (3) a stranger-associated condition in which a stranger-associated shape singleton distractor was presented. Modulation of attentional capture is quantified as the response time cost between (2) and (3). Facilitation from self-association in the training phase, and attentional capture in the test phase were found, but modulation of attentional capture by self-association was not found. The current study might indicate that self-relevance effect does not transfer to visual attention, or shape-label matching task is not suitable for associative learning.