Abstract
Is statistical processing in vision immune to attentional constraints? To answer this question, we manipulated attentional resources engaged in statistical processing in visual arrays. An adapted attentional blink (AB) paradigm, in which a mean size judgment task was given as the second target (T2) and a digit was the first target (T1), was used to investigate observers' ability to compute mean sizes of visual arrays inserted in fast visual stimulus streams. This paradigm enabled us to test whether observers' performance on mean size judgments would depend on reporting T1 (dual task requiring the report of both T1 and T2 vs. single task requiring the report of T2 only) and on time lag between T1 and T2. Surprisingly, mean size computation was not susceptible to limited attentional resources: neither dual-task interference nor performance drop at short lags was evident. To make sure we engaged attentional resources sufficient for the AB in our paradigm, we used single size judgments instead of mean size judgments in the same paradigm and tested if single size judgments were susceptible to attentional constraints in our procedure. The results show single size judgments were influenced by both dual-task interference and time lag. Furthermore, even when we presented a random-dot mask following visual arrays for mean size judgments as substitution masking, no evidence of the AB was found. These findings indicate our paradigm is sufficient to manipulate attentional resources and in turn, statistical processing in visual arrays is indeed immune to attentional constraints.
This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2007- 101039003-2007-8-0746).