The paradigm of cued visual selection (Nothdurft, 2017a, www.vpl-reports.de/6/) was tested on different features and the dynamics in the identification of orientation targets, color targets, and facial expressions were compared. In large arrays of up to 80 items, one item (the target) was cued at various delays after stimulus onset; observers had to identify this item and report its properties. Performance accuracy was measured for different presentation times, Δt, after the cue. With all features, accuracy improved with longer presentations. The data were fitted with cumulative Gaussian functions and the presentation time for 75% correct responses, Δt75, was estimated for each pattern and each cue delay; the values reflect the dynamics and relative strengths of underlying neural signals. For orientation, discrimination was strongly accelerated at cue delays 50-100 ms after the pattern onset. With longer delays, it slowed down again. For color, the acceleration was stronger and more prolonged. It started earlier and reached its maximum about 400 ms after stimulus onset. With longer delays, performance deteriorated but never fell back to the level at zero delay. With facial expressions, the required presentation times were generally increased; faster discrimination at short delays was less pronounced and not seen in all observers. Shortest Δt75 values were reached 50-200 ms after stimulus onset, at similar delays as for orientation. All data were used to reconstruct the signals that provide the cued discrimination of orientation, color, and facial expressions. The link to neurophysiological studies and the evidence for different impacts of these features on the control of spatial attention are discussed.

Published online: 23-Jan-2020

Citation: Nothdurft, H.C. (2020). Dynamic differences in the identification of orientation, color, and facial expressions. VPL-reports, 10, 1-27, www.vpl-reports.de/10/