2025/02/18 by Daniel Tolédano, Nitzan Micher, Dominique Lamy · 1 voice
Computer Science · #Explainable Artificial Intelligence (XAI)
paper · doi:10.1037/xhp0001279
openalex publication_date 2025/02/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/31
We tend to prioritize features and locations that have recently received our attention. Surprisingly, even irrelevant features of recently attended targets enjoy increased priority. However, such irrelevant-feature priming was found for some features and not for others. Here, we inquired whether the fact that irrelevant-feature priming is sometimes absent results from a failure of encoding or from a failure of attentional guidance. To answer this question, we relied on a finding common to the visual search and attentional-control literature: when a stimulus is responded to, the features and motor response associated with the action event are bound in a common representation and can be later retrieved. In two experiments, some participants searched for a color target and others for a shape target-with shape and color serving as the target's irrelevant feature for the former and for the latter, respectively. Responding to the target required an easy discrimination (Experiment 1) or a difficult one (Experiment 2). Repeating the target's irrelevant color speeded search, but repeating its irrelevant shape did not. However, the irrelevant feature-response binding effect was similar for the two search dimensions. These findings invalidate the no-encoding account. Additional findings indicate that irrelevant-feature priming shares the main characteristics of other intertrial priming phenomena. (PsycInfo Database Record (c) 2025 APA, all rights reserved).