Abstract
Purpose. Visual crowding refers to the impaired identification of a peripheral stimulus when it is surrounded by flankers with similar patterns. According to an attention explanation, crowding happens because attention fails to individualize the target among the flankers due to limited spatial resolution of attention. As voluntary attentional selection requires visual awareness, here we ask whether visual awareness is necessary for crowding to happen.
Method. Two normally sighted observers adapted to a dicoptic stimulus located at 25° in the lower visual field for 5 s. The non-dominant eye was shown a low-contrast (×16 of the detection contrast threshold) Gabor target (oriented at 45°) with or without 4 Gabor flankers (oriented at 45° or −45°). The observers were unaware of the target and flankers because the percept was always dominated by a series of chromatic dynamic noise presented to the dominant eye. Contrast threshold at the retinal location of the target was measured by a 2IFC detection task. Threshold elevation was calculated by comparing thresholds with and without adaptation, and was measured in both crowded and isolated conditions.
Results. Threshold elevation was found only when the target and test had the same orientation. Threshold elevation in the crowded condition was 31 +/− 6% lower than the control condition.
Conclusions. The magnitude of orientation-specific adaptation was reduced in the presence of flankers, even though the observers were unaware of neither the target nor the flankers. Thus, crowding happens without visual awareness. Our findings challenge the role of voluntary attention in crowding.
Supported by HK RGC GRF grant 741808 to S.-H. C.