Abstract
Purpose. Information near filled-in regions alters the perception of interpolated shape, but it is unknown whether this process depends on top-down influences. Here, we consider whether observer strategy can reduce filling-in effects when interpolation normally occurs, or elicit such effects when interpolation normally does not occur. Method. Subjects discriminated briefly-presented, partially-visible fat and thin shapes, the edges of which either induced or did not induce illusory contours (relatable and unrelatable condition, respectively). Half the trials of each condition incorporated task-irrelevant distractor lines, known to disrupt the filling-in process. Half of the observers were asked to treat the visible parts of the target as belonging to a single thing (group strategy); the other half were asked to treat the parts as disconnected (ungroup strategy). A strategy was encouraged by giving subjects pictures of the fat and thin response templates in the instruction phase of the experiment, and at the end of each trial. These pictures depicted either unitary shapes or fragmented shapes, depending on the strategy. Results. There were three main results. First, distractor lines impaired performance in the relatable condition (p<0.001), but not in the unrelatable condition (p>0.7). Second, for both relatable and unrelatable stimuli, strategy did not alter the effects of the distractor lines (p>0.7). Finally, the attempt to group relatable fragments improved performance (p<0.001) while the attempt to group unrelatable fragments did not (p>0.3). Conclusions. These results suggest that a) filling-in during illusory contour formation cannot be easily removed via top-down strategy; b) filling-in cannot be easily manufactured from stimuli that fail to elicit interpolation; and c) actively grouping fragments can readily improve discrimination performance, but only when those fragments form interpolated contours. These findings indicate that while discriminating filled-in shapes depends on strategy, filling-in itself is relatively encapsulated from top-down influences.
This research was supported by a grant to the first author from the American Psychological Association.