Memory processes in repeated visual search

Visual search often involves searching the same environment, consecutively, for a number of different targets. In the experiments described in this proposal we investigate the extent to which search benefits from such previous exposure. In our pilot experiments participants searched the same display for three consecutive targets. Manual responses were faster in the second search than the first search. However, there was no additional benefit for the third search. Eye movement recordings demonstrated that the time necessary to find a target letter depended on when that letter was last fixated in the previous search even when distance was controlled for. This fixation recency effect lasted for about four fixations. These results provide evidence for a short-term memory store in this kind of visual search. The main objective of the project is to investigate what kinds of short-term memory are responsible for this effect. This objective will be achieved by conducting three sets of closely related experiments.

When we leave our homes in the morning, many of us are looking for their car keys and then for the umbrella. In the supermarket, we are searching the shelf with the dairy products for milk and then for butter. These activities are examples of repeated visual search. Despite its ubiquity this kind of search has not drawn much attention in psychology. In this research project we investigated repeated search in a series of highly controlled experiments. Participants searched a display with letters twice for different target letters while we recorded their eye movements. We found that the second target is found somewhat faster than the first one. More specifically, the second target was found faster if it had been seen within the last four fixations of the search for the first target. This is called a short-term memory recency effect. Often, in every-day life, we are being interrupted whilst searching or we are doing something else simultaneously like making a phone call. In our experiments, we showed that the recency advantage is robust to a delay between the two searches and that it is not disrupted by either a secondary task between the two searches or a concurrent secondary task. We have further demonstrated that the recency effect depends on a coding of both the identity and location of the searched objects. Finally, we showed that this memory is dependent on the continuous presence of the display across the two searches.