Abstract
Abstract: :
Purpose:Crowding between adjacent letters has been investigated primarily as a spatial effect. The purpose of this study was to investigate the temporal properties of letter crowding. Specifically, we examined the hypothesis that the degradation in letter identification performance in the presence of nearby letters can be alleviated when adjacent letters are presented with a temporal asynchrony. Methods:We measured proportion–correct performance for identifying the middle target letter in strings of three lowercase letters (trigrams) at the fovea and 10 deg in the inferior visual field. Letters making up each trigram were 100% contrast letters chosen randomly from all 26 letters of the Times–Roman alphabet. Letter sizes corresponded to the sizes that yielded ca. 90% correct for single–letter identification at each eccentricity. The center–to–center separation between adjacent letters ranged from 0.8 to 4X the x–height. The target and the flanking letters were presented simultaneously, or with a stimulus onset asynchrony (SOA) between –200 and 250 ms, where positive SOAs mean that the target preceded the flanking letters. Target and flanking letter durations were always the same, and were either 50 or 150 ms. Results:When the target and its flanking letters were presented simultaneously (SOA = 0), performance for identifying the target letter showed the classical crowding effect —— highest at the largest separation, and dropped with smaller separations. The reduction in performance was larger (from 90% to 10% correct) at 10 deg eccentricity than at the fovea (from 90% to 50% correct). However, performance improved with SOAs (either positive or negative), even at small letter separations. At 10 deg eccentricity and at 0.8x letter separation, performance improved from ∼10% at SOA = 0 to ∼80% at SOA = 150 ms. The improvement was smaller at larger letter separations, and at the fovea. Conclusions:Consistent with our hypothesis, performance for identifying flanked letters improved when the target and flanking letters were presented at different times. The reliance of letter identification performance on SOAs and how it changes with letter separations implies that the crowding effect can be traded between space and time. Our findings argue that crowding should be considered as a spatio–temporal, and not simply a spatial, effect.