Template-based and saliency-driven attentional control converge to coactivate on a common, spatially organized priority map.

Abstract

Visual attention can be controlled both by a match to known target attributes (template-based guidance) and by physical salience (saliency-driven guidance). However, it remains unclear how these mechanisms interact to determine attentional priority. Here, we contrasted two accounts of this interaction. Under a coactive mechanism, template-based and saliency-driven guidance are simultaneously integrated in a common priority signal. Under a noncoactive mechanism, the two sources of control do not converge on a common priority signal, either because they are separated architecturally (separate-activations model) or temporally (sequential model). In a redundancy-gain paradigm, search targets were defined either as a match to a shape cue (template-based), the presence of a singleton-colored item (saliency-driven), or both (redundant). We assessed whether the response time distribution in the redundant condition contained a substantial proportion of trials that were faster than could have been generated by the faster of the two individual guidance processes operating independently in parallel, that is, violation of the race model inequality (RMI). This effect can be generated only by a coactive mechanism. The results showed robust violations of the RMI when both features appeared at the same location, consistent with a coactive model. In addition, violations of the RMI were eliminated when redundant features were displayed at different locations, indicating that guidance signals combine on a spatially organized priority map. (PsycInfo Database Record (c) 2025 APA, all rights reserved).