Preview of partial stimulus information in search prioritizes features and conjunctions, not locations

Abstract

Visual search often requires combining information on distinct visual features such as color and orientation, but how the visual system does this is not fully understood. To better understand this, we showed observers a brief preview of part of a search stimulus-either its color or orientation-before they performed a conjunction search task. Our experimental questions were (1) whether observers would use such previews to prioritize either potential target locations or features, and (2) which neural mechanisms might underlie the observed effects. In two experiments, participants searched for a prespecified target in a display consisting of bar elements, each combining one of two possible colors and one of two possible orientations. Participants responded by making an eye movement to the selected bar. In our first experiment, we found that a preview consisting of colored bars with identical orientation improved saccadic target selection performance, while a preview of oriented gray bars substantially decreased performance. In a follow-up experiment, we found that previews consisting of discs of the same color as the bars (and thus without orientation information) hardly affected performance. Thus, performance improved only when the preview combined color and (noninformative) orientation information. Previews apparently result in a prioritization of features and conjunctions rather than of spatial locations (in the latter case, all previews should have had similar effects). Our results thus also indicate that search for, and prioritization of, combinations involve conjunctively tuned neural mechanisms. These probably reside at the level of the primary visual cortex.

Keywords: Color discrimination; Conjunction processing; Feature equality; Feature preview; Orientation discrimination; Precuing; Saccades; Visual search.

Fig. 1

Precuing conditions in Experiment 1. For clarity, the bars in the three conditions have been drawn in exaggerated contrast. In the actual experiment, feature contrast were based on individually determined feature discrimination thresholds. (Color figure online)

Fig. 2

Effect of preview in Experiment 1. Panels show overall search performance (panel “hit rate”), saccadic response latency in case of hits (panel “hit latency”), color selection efficacy (panel “CSE”), and orientation selection efficacy (panel “OSE”). Error bars show standard error of the mean between observers. The dotted lines correspond to the no-preview level

Fig. 3

Additional disc precuing conditions in Experiment 2. For clarity, the discs and bars have been drawn in exaggerated contrast. For the actual experiment, feature contrasts were based on individually determined feature selection performance. (Color figure online)

Fig. 4

Effect of preview in Experiment 2. Panels show overall search performance (panel “hit rate”), saccadic response latency in case of hits (panel “hit latency”), color selection efficacy (panel “CSE”), and orientation selection efficacy (panel “OSE”). Error bars show standard error of the mean between observers. The dotted lines correspond to the no-preview level

Fig. 5

A conceptual model of prioritization in visual search. Depicted is a sensitivity space, in which putative neural mechanisms (i.e., neurons) are situated based on their sensitivity. Previews, depending on their content, cause prioritization by shifting the neural population’s sensitivity. (Color figure online)