Madilyn Dungan¹ (), Nadia Iranpour¹, Michael Wenger², Sarah Newbolds²; ¹Undergraduate, ²University of Oklahoma

This study investigates the binding of sound and color in working memory, examining how these features interact within the context of attention-driven memory processes. Previous research has explored this question from various theoretical and methodological angles, but we apply general recognition theory (GRT) to mathematically assess the relationship between sound and color in working memory. GRT characterizes binding in three ways: violations of informational independence, informational separability, and decisional separability. These constructs are defined mathematically, allowing for rigorous tests for how sound and color features are or are not bound together. Stimuli consisted of a single visual form (triangle) whose color (purple) varied in saturation (low, medium, and high), that was paired with a pure sine-wave tone that varied in frequency/pitch (low, medium, and high). Three working memory conditions were tested using an n-back memory task at three levels of demand: 0-, 1-, and 2-back. In the 0-back condition, participants identified the saturation and pitch of the current stimulus pairing. In the 1-back and 2-back conditions, participants compared the current stimulus with the one presented 1 or 2 items before. Data were analyzed at the level of individual participants. Although there were individual differences, the GRT analyses generally pointed to independence and separability in both immediate perception (0-back) and working memory (1- and 2-back), with there being very few failures of marginal response invariance. Results are considered in the context of current accounts of feature binding in working memory and suggest a variety of steps to be pursued in this line of work.