S. Friedmanhill et Jm. Wolfe, 2ND-ORDER PARALLEL-PROCESSING - VISUAL-SEARCH FOR THE ODD ITEM IN A SUBSET, Journal of experimental psychology. Human perception and performance, 21(3), 1995, pp. 531-551
Visual search tasks in which participants searched for an odd element
in a subset of items were investigated. Participants searched for an i
tem of odd orientation in the red subset. The target was a led line of
X degrees, distracters were green fines of X degrees and red lines of
Y degrees. The orientations, X and Y, changed on every trial. In this
task, orientation information was useful only after color had been us
ed to select the relevant subset. Results show that response time (RT)
and error data were different from standard color X orientation conju
nction searches (Experiment 1). RT X Set Size functions had slopes nea
r 0 ms per item (Experiment 2). The selection of the subset appeared t
o take 200-300 ms (Experiments 2 and 3). Subset selection was based on
properties of the relevant subset, not the irrelevant subset (Experim
ent 4). It was more difficult (perhaps impossible) to select a subset
defined by 2 colors (Experiment 5). Random variation in an irrelevant
dimension did not disrupt subset search (Experiment 6).