Abstract
Our previous research using a ‘window’ and ‘scotoma’ paradigm suggested that early in processing, scene gist recognition is better for central than peripheral vision, followed thereafter by converging performance, consistent with the hypothesis of attentional expansion over time. However, an alternative explanation is that the ‘scotoma’ captures attention early in processing, resulting in worse gist performance for peripheral information. We tested these hypotheses by manipulating attention to focus either centrally or peripherally, and examining its effects on scene gist over time.
Scenes were manipulated by presenting imagery only in a central circular region and blocking information outside it (the “window”), or conversely, by blocking imagery in the central circular region (the “scotoma”) and presenting information only outside it. A single critical radius divided the circular images into mutually exclusive center and surround regions, which produced equal gist performance in both conditions when unmasked. Images were presented for 24ms with masking SOAs of 36, 70, 105, and 376 ms. Eyetracking ensured central fixation during scene presentation. Attention was manipulated between-subjects by presenting 80% of the trials as either window or scotoma images.
Early use of central versus peripheral information differed significantly as a function of attention. Specifically, at 36ms SOA, when attention was centrally focused, performance was significantly better in the window condition than the scotoma condition, whereas when attention was peripherally focused, there was no difference between either condition. Thereafter, with increasing processing time, gist performance equalized, as predicted by the use of the critical radius to create the stimuli.
Thus, at early processing times, attention moderates gist recognition between central and peripheral vision. However, with additional processing time, performance converges to produce equal gist performance between central and peripheral information, consistent with the hypothesis that attention expands out from the center of a scene over a single fixation.