Abstract and Concrete Sequences Rely on Separable Working Memory Resources

Poster Session B - Sunday, April 14, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC

Jorja Shires¹ (jshires@nevada.unr.edu), Jenna N. Pablo¹, Bodie Stein¹, Theresa M. Desrochers², Marian E. Berryhill¹; ¹University of Nevada, Reno, ²Brown University

Daily life involves accomplishing overarching abstract goals (e.g., be productive) with subordinate tasks (e.g., running a participant). Typical working memory (WM) tasks involve the retention and manipulation of sensory stimuli themselves. A newer paradigm, abstract cognitive task sequences (ACTS), requires participants to retain a series of instructions to apply to stimuli. These are operations to apply independently of the specific stimuli. Here, we alter the task design to test whether abstract and concrete sequences are maintained in separate stores. An ACTS-WM modification doubled WM demands: participants maintained instructions and stimuli. Participants encoded a four-item instruction sequence and applied them to sequentially presented pictures, reporting answers after stimulus presentation. If the abstract instructions and concrete stimuli are maintained uniformly, increasing demands would impair performance. WM capacity limits would limit both if maintained in the same store. Undergraduates (n = 64) completed ACTS and ACTS-WM tasks. Accuracy was significantly higher for the ACTS-WM task (p < 0.001). To probe further, we examined errors. WM maintenance of the concrete stimuli and the abstract sequence were similar between tasks (p = .42), suggesting separable storage. However, 67% of errors were in the sequence, whereas 22% were stimulus errors (p < 0.001). These findings suggest both abstract and concrete sequences were maintained independently, and the abstract representations were disrupted more easily.

Topic Area: EXECUTIVE PROCESSES: Working memory