Abstract
We investigate the effect of manipulating the probability of the target being in a particular location in a search task and what processes underpin these effects. Participants searched for a target (Landolt C with gap at top) amongst 7 distractors (Landolt Cs with gap at base). For each trial, the latency and landing position of the first saccade was analyzed. In Experiment 1 the target was twice as likely to appear on one side of the display as the other. The first saccade was directed to the target more often when it occurred in the more frequent locations. This benefit could be a result of sequential dependencies or implicit learning of the spatial probabilities. Experiment 2 demonstrated that there were sequential dependencies in this task. When the target appeared with equal probability at each location, performance on the current trial improved when the target had appeared at the same location on either of the two preceding trials. In Experiment 3 the target was twice as likely to appear on one side of the display as the other (cf. Experiment 1), but with the additional constraint that the target location could not repeat within a sequence of 4 trials. The results showed that when short term target location repetitions were eliminated there was no longer an advantage for more frequent locations. These results suggest that the benefits for more frequently occurring locations result from short-term target location repetitions rather than implicit learning of the spatial probabilities.