B.1.3. Experiment 1.1: Salience

The first experiment was conducted to evaluate objectively the salience induced by size differences and the time-course of its effects. The first goals were to ensure that size differences induced genuine exogenous salience effects and that these effects did influence the processes involved in orienting attention. The paradigm ensured that reaction time (RT) variations were due to salience and not to some perceptual effect. Indeed, RTs for targets of three sizes were measured in two conditions. In the homogeneous condition, the target was surrounded by distractors of the same size: thus, target's salience should not vary with size. In the heterogeneous condition, the target was surrounded by distractors of the two remaining sizes: therefore, the target's salience should vary with size. For instance, whereas in the homogeneous condition a large target square would be presented with two other large distractor squares, it would be accompanied by a medium and a small distractor squares in the heterogeneous condition. Thus, although the targets were identical in both conditions, the context varied, and so should do the target's saliences. Target's saliences should vary with size only in the heterogeneous condition: then the larger the target, the more salient. Consequently, the hypothesis was that RTs would vary as a function of target size in the heterogeneous condition, but not (or much less) in the homogeneous condition.

Participants had to make an orientation discrimination task (top vs. bottom), while salience was manipulated through size differences. This allowed to ensure that salience influenced attentional orienting, and not a response stage of processing, because the salient and the to-be-reported features concerned different dimensions. In addition, the salient feature was uninformative vis-à-vis the target, which countered any endogenous incentive toward this signal. To that end, the target presented this salient feature in only 1/d of the trials (where d is the number of items in the display; see Yantis and Egeth, 1999).