Abstract
Although value associations are known to modulate attention even when they are irrelevant (Raymond & O'Brien, 2009; Rutherford et al., 2010), it remains unclear whether value affects spatial attention by changing cognitive control settings or by altering data-driven priority maps. To investigate we measured conflict adaptation (CA) effects in a classical color flanker task using value-associated (Exp 1) or perceptually salient (Exp 2) distractors. In Exp 1, participants learned to associate color symbols with monetary wins or no outcome in a simple choice task. Later, all participants made speeded color categorizations of a central symbol presented with same (congruent) or different (incongruent) flanking symbols, without the expectation of monetary reward on any trial. In such tasks, responses are slowed on incongruent trials (the congruency effect, CE) and CE effects are reduced when the previous trial is incongruent. This latter effect, known as conflict adaptation (CA), indexes task-specific, transient, conflict-driven modulations of attentional selection control settings. We reasoned that if irrelevant value-associations are able to modulate control settings, then distractor value-associations should determine CA. In contrast, CA should be unaffected by the perceptual salience of distractors, a feature that drives bottom-up attention. If value associations modulate attention in a bottom-up fashion, value and perceptual salience should have similar, null effects on CA. We report that positive versus no-outcome value-associated distractors significantly reduced CA effects, whereas perceptual salience (bright/dim) had no effect. The presence of value-associated stimuli in the scene appears to terminate conflict adaptation responses instantiated by a prior incongruent trial, without influencing CEs measured after congruent trials, a finding consistent with the notion that the presence of value-associated stimuli in a scene affects attentional control settings, not bottom-up priority maps.
Meeting abstract presented at VSS 2016