The role of action in perceptual decision making
Loading...
Date
2025-01-29
Authors
Μπαλντούκας, Κωνσταντίνος
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The motor system plays a pivotal role in shaping sensory processing and decision-making during goal-directed actions, such as reaching into a bag to locate a specific object. To explore this interplay, we conducted a two-alternative forced-choice experiment where 16 participants lifted two objects of varying weights to decide which was heavier or lighter. Participants reported their decisions by deviating the trajectory of their lifting movements either to the left or right, while electromyographic (EMG) signals were recorded from nine muscles of the lifting arm. Using single-trial multivariate Linear Discriminant Analysis (LDA) on the EMG data, we identified two distinct neural signals: one reflecting the participants’ choices and another encoding the presented stimuli. Both signals demonstrated a consistent trend, with classifier performance (Az) progressively improving midway through the trial, capturing a decision-making component. Cognitive modeling with a Hierarchical Drift Diffusion Model (HDDM) revealed key decision-making dynamics. The drift rate, representing the speed of evidence accumulation, was positively correlated with choice accuracy, while the non-decision time parameter, encompassing sensory encoding and motor execution processes, predicted reaction times. Crucially, the first timepoint at which significant weight discrimination emerged in EMG signals for each participant was strongly predictive of their median reaction time and, more specifically, the HDDM-derived non-decision time. This finding links the pre-decisional EMG activity to the sensory encoding phase of the decision-making process. HDDM analyses further revealed that EMG components associated with weight discrimination increased evidence accumulation rates (higher drift rate) and response caution (higher decision boundary) during trials involving heavier stimuli. Overall, this study establishes a comprehensive framework for understanding how motor signals interact with cognitive processes to shape perceptual decisions. By integrating neurophysiological measures and computational modeling, it provides novel insights into the shared mechanisms underpinning sensory encoding, motor execution, and decision dynamics.
Description
Keywords
Αctive sensing, Decision-making, Linear Discriminant Analysis (LDA), Hierarchical Drift Diffusion Model (HDDM), Motor control