Guest Speaker: Henry Jones from University of Chicago is a graduate student working with Dr. Ed Awh.

Title:Content-independent pointers mediate individual differences in working memory ability

Abstract: What is the key “currency” for storage in working memory (WM)? In the past, we have proposed that the key process of WM may be the assignment of “pointers”, which serve to bind item representations to the current context for sustained access. In this talk, I will briefly present the story of how the lab’s pointer theory developed concurrently with the development of our use of multivariate analytic methods: classification and representational similarity analysis (RSA). Next, I’ll test a key prediction of the pointer theory, that it underlies capacity limits and contributes to individual differences. To test this prediction, we designed a change detection task with a wide range of set sizes (2,4,6,8). In addition, we combined RSA with Bayesian modeling to estimate how the pointer signal changed across set sizes, isolated from other signals driven by typical confounds (spatial attention, sensory energy, and distractor filtering). Our findings revealed a sharp contrast between high and low capacity subjects: the WM load signal plateaued at 4 items in high-capacity subjects, but decreased at supraspan set sizes in low-capacity subjects. These results dovetail with past neural and behavioral studies of how high- and low-capacity subjects process supraspan displays, suggesting that pointers are a core component of individual differences in WM capacity. I’ll close with initial results from an ongoing follow-up study and some reflections on how methodological developments led to theoretical developments.