Abstract
We investigated how cue predictability affects reaction time (RT). We manipulated number of locations, the presence of distracters and the amount of information (in bits) provided by the spatial cues. In each experiment, a trial began with a fixation field, followed by a central cue, and finally a target letter. In Experiment 1, we varied cue information by varying the number of target locations. In separate blocks, there were either two or six possible target locations. On half of the trials, a noninformative cue (a + sign) was presented. On the remaining trials an arrow cue was 100% predictive of target location, corresponding to 1 and 2.58 bits of information in the two and six location displays respectively. We found that although the cue had a significant effect on RTs, the cueing effect did not vary as a function of information. Experiment 2 was an exact replication of Experiment 1 except that we added distractors that appeared in nontarget locations. In this experiment, the cueing effect was significantly larger with six locations than with two locations. Thus with distractors, attention was modulated by the information provided by the cue. Last, we parametrically investigated this effect. On some trials, the target appeared in the cued location (valid) and on some trials it indicated a nontarget location (invalid). The proportion of valid trials varied across session. There were four levels of cue predictability corresponding to 0.0, .51, 1.02, and 1.55 bits of information. We found that the relation between the cueing effect in RT and information provided by the cue was linear. In summary, in the absence of distractors, effects of cue predictability were similar regardless of the amount of cue information. In contrast, when distractors were present, as cue information increased, so did the RT cueing effects.
Meeting abstract presented at VSS 2012