Abstract
A key feature of visual crowding (impaired recognition in clutter) is its radial-tangential anisotropy in the periphery. Impending saccades are known to cause spatiotemporal mislocalizations to briefly flashed stimuli. We investigated whether and how crowding zones are affected before saccades that are targeted to a different location than the crowded stimuli. In Experiment-1a, seven observers performed a peripheral (10deg to the right) orientation discrimination task (2-AFC, ~24500 trials). In saccade trials (80% of all), a fixation spot jumped 10deg vertically, cueing the observer to make a saccade. Target stimulus (a tilted Gabor, 4cpd, tilt adjusted for each observer for 80% correct) was presented, with or without horizontally or vertically flanking vertical Gabors. The target-flanker spacing was adjusted using the method of single stimuli. We found large individual differences in both the strength and extent of crowding. The effect of saccades was also idiosyncratic. Interestingly, we found nearly isotropic crowding zones both in fixation and saccade conditions. In Experiment-1b, only the fixation trials (~2100 trials) were run to determine whether the isotropy was due to additional attentional load of the saccade task. However, the results were similar. To determine whether the lack of anisotropy was due to the stimuli/task, we used a letter recognition task (10-AFC) in Experiment-1c (~2100 trials, only fixation). Six observers showed strong anisotropy. In Experiment 2, we repeated Experiment-1a using this letter recognition task with five observers (15000 trials). Consistently, we found weaker and near perfect isotropy in the fixation and saccade conditions, respectively. For four observers, this was due to enlarged tangential zones. Our results suggest that impending saccades as well as stimuli-task combinations, which necessitate larger spatial integration reduce the anisotropy of crowding zones. The locus of spatial attention might be a common denominator for all these factors.
Acknowledgement: NIH R01-EY012810 and a UC Berkeley URAP Summer Fellowship