Abstract
Crowding is the deleterious influence of neighboring stimuli on peripheral target perception. Typically, crowding deteriorates discrimination but not detection. However, recent studies revealed that elements are often omitted in crowded displays, resembling failures to detect. It has been suggested that omissions in repeating patterns consisting of as few as three items (triplets) are due to a mechanism called “redundancy masking”. Here, we show how redundancy masking goes hand in hand with systematic compressions of perceived space. Observers were presented with 3–7 identical, equally spaced lines at 10° eccentricity. We varied the spacing between the lines (5 levels). Using a dual task, observers first indicated the number of perceived lines. Second, observers judged the separation between adjacent lines in Experiment 1, and the horizontal extent of the entire line array from the innermost to the outermost line in Experiment 2. In both experiments, the number of perceived lines was reduced in the majority of trials, showing strong redundancy masking. Next, we compared the separation estimates of the line triplets for correct trials with those in ‘masked’ trials where only 2 lines were reported. In Experiment 1, separation estimates were larger in masked (M=1.52+/−0.16) than in correct (M=1.01+/−0.16) trials. By contrast, in Experiment 2, separation estimates were smaller in masked (M=0.83+/−0.18) than in correct (M=1.25+/−0.19) trials. Importantly, the reported separation of the two lines was approximately the same in the masked trials of both experiments, indicating that the entire array underwent spatial compression. Probe experiments to identify the perceived centroid of the arrays ruled out alternative hypotheses. We suggest that the systematic compression of visual space in redundancy masking is a major contributor to crowding in displays with repeating elements.
Acknowledgement: The Swiss National Science Foundation (PP00P1_163723 to Bilge Sayim).