Evidence against early selection: stimulus quality effects in previewed displays

Abstract

H. Pashler (1984) reported that when subjects identified a probed letter in a display of 8 characters, the effect of stimulus quality on reaction time persisted full blown even when subjects had 300 ms to preprocess the display. Pashler argued that these results are incompatible with theories of late selection but that they are naturally accommodated by theories of early selection. The authors report 14 experiments using Pashler's methodology in which the effects of stimulus quality were reliably attenuated with a preview of the letter array. Pashler's results were also replicated, but only under a narrow set of conditions. Several "early selection" accounts of the interaction of probe delay with stimulus quality were examined and rejected in favor of a late selection account.