Loss aversion and evidence accumulation in short-video addiction: A behavioral and neuroimaging investigation

Excessive use of short-video platforms not only impairs decision-making processes but also predisposes individuals to addictive behaviors. This study investigated the relationship between short-video addiction (SVA) symptoms and loss aversion (LA), delving into the underlying computational and neural mechanisms using the drift diffusion model (DDM) and the inter-subject representational similarity analysis (IS-RSA). Behavioral analyses revealed a significant negative correlation between SVA symptoms and the LA coefficient (lnλ). Additionally, the DDM-based drift rate (v) was found to mediate this relationship. Neuroimaging analyses further indicated that SVA symptoms were negatively associated with gain-related activity in the right precuneus, while positively correlating with loss-related activity in the right cerebellum and left postcentral gyrus. Notably, precuneus activation during gain processing mediated the relationship between SVA symptoms and both lnλ and drift rate. IS-RSA revealed that inter-subject variations in SVA symptoms were significantly associated with distinct activation patterns related to gain processing in the frontoparietal network (e.g., frontal pole, inferior frontal gyrus, and supramarginal gyrus) and motor network (e.g., precentral), as well as loss-related activation patterns in the motor networks (e.g., postcentral and pre-supplementary motor area). Similar patterns emerged when examining simultaneous gain and loss-related activation patterns. Mediation analyses further demonstrated that functional activation patterns in the motor network mediated the relationships between inter-subject variations in SVA symptoms and both loss-aversion and psychological processing patterns (e.g., decision threshold, drift rate, and non-decision time). These findings provide novel insights into the cognitive and neural mechanisms underlying the influence of SVA symptoms on loss aversion, and suggest the critical roles of evidence accumulation speed and specific brain activation patterns—particularly within the cognitive control and motor network—in shaping decision-making biases associated with addiction.