Schedule of Events | Search Abstracts | Invited Symposia | Symposia | Poster Sessions | Data Blitz

A Multiple Trace Model Unifying Statistical Learning and Intertrial Priming.

Poster Session F - Tuesday, March 10, 2026, 8:00 – 10:00 am PDT, Fairview/Kitsilano Ballroom

Dock Duncan¹ (), Sander Los¹, Jan Theeuwes¹; ¹Vrije Universiteit Amsterdam

Visual selective attention is guided to a large extent by where we have encountered targets and distractors in the past. The influence of past behavior is often called “selection history”. Two prominent forms of selection history effects are Intertrial priming (referring to effects on selection that are due to processing taking place during immediate preceding trial) and statistical learning (effects that are due to the overall average regularities during the overall experimental session). Traditionally, these effects have been thought of as separate mechanisms even though there is evidence that they interact. In the current talk, we present evidence that both effects are driven by the same core cognitive mechanism involving a short-term and long-term expression of the same cognitive process. By applying a multiple trace computational framework (in which each trial leaves a memory trace which decreases in influence over time according to a power function towards a nonzero asymptote), we present a model that can explain immediate effects of intertrial priming and higher order statistical learning effects. Together, these processes work to improve attentional behavior by biasing it towards global and local averages. We demonstrate that a multiple trace model successfully captures both short-term and long-term effects, including extinction, thereby unifying these disparate fields of study

Topic Area: ATTENTION: Spatial