Abstract
This study examined whether relevance per-se can cause blindness (i.e., failure to report clearly visible stimuli) even when there is no resources limitation. In a novel paradigm, employed in 2 similar experiments, a colored circle surrounded by a differently colored ring was presented for 500 ms, without masking. One of these stimuli was labelled as relevant at the beginning of the trial. Following the stimuli offset, the observers were asked to identify the colors of both relevant and irrelevant stimuli (report order was counterbalanced). While color identification of the relevant stimulus was near perfect, up to a quarter of the participants could not name the color of the irrelevant stimulus. Critically, a control experiment indicated that there were sufficient resources to process both stimuli. The fact that the effect of relevance was similar regardless of whether the observers had to report the irrelevant stimulus first or second suggests that this blindness does not reflect forgetting. Instead, we propose that our observers were aware of the irrelevant color when it was displayed, but due to irrelevance, its representation was insufficiently activated, resulting in recognition failure. To explore this possibility we run another experiment in which an illusory rectangle was created via modal-completion brought about by 4 colored inducers. Previous studies suggest that such modal-completion can only take place with conscious inducers. Task instructions only required identification of the rectangle orientation, but following the inducers offset the observers were also asked to report the inducers color. Like before, reports of rectangle orientation (relevant stimulus) were highly accurate (though here the order did matter), suggesting that the irrelevant inducers reached awareness, yet reporting their color often failed. Overall, these results demonstrate blindness when mental resources are clearly available, challenging attentional theories that predict strong selection only when resources are taxed.
Meeting abstract presented at VSS 2014