Real-time reorienting of preparatory sustained attention lapsing during episodic retrieval using closed-loop pupillometry

Poster Session E - Monday, April 15, 2024, 2:30 – 4:30 pm EDT, Sheraton Hall ABC
Also presenting in Data Blitz Session 1 - Saturday, April 13, 2024, 1:00 – 2:30 pm EDT, Ballroom East.

Shawn T. Schwartz^1,2 (stschwartz@stanford.edu), Khanh K. Nguyen¹, Haopei Yang¹, Megan T. deBettencourt¹, Tammy T. Tran^1,3, Kevin P. Madore¹, Anthony D. Wagner^1,2; ¹Stanford University, ²Wu Tsai Neurosciences Institute, Stanford University, ³Stanford University School of Medicine

Successful goal-directed knowledge expression is modulated, in part, by moment-to-moment lapses in preparatory sustained attention (assayed by pupillometry/scalp EEG alpha-power) in pre-goal periods immediately preceding memory retrieval attempts. While theoretically informative, correlations between attention and retrieval success yield limited evidence regarding the causal role of attention lapsing on memory retrieval and constrain implications for whether moment-to-moment attention can be intervened upon to optimize performance. Here, we leveraged real-time readouts of trial-to-trial pupil diameter to trigger attention-reorienting probes just prior to retrieval attempts. After completing a goal-directed associative memory encoding task, 75 young adults (18-25 yrs) indicated whether they remembered test probes as having been encountered in one of two task goals during encoding. Memory was assayed at the trial-level (hits/misses) and individual-level (memory d’). At the beginning of each block, we built participant-specific distributions of baseline-pupil during tonic fixation periods for trigger-thresholding. Critically, then, moment-to-moment pupillary dilations/constrictions which exceeded the empirical threshold triggered deployment of salient, real-time attention-reorienting probes just prior to retrieval probe delivery. As predicted, we found a correlation between hit/miss item memory and pre-stimulus tonic pupil diameter. Attention-reorienting triggers on attention lapsing trials rescued performance (item-memory d’), returning close to that of control trials (those with no detected attention lapses/triggers), although there was marked variability in reorienting efficacy across participants. These initial findings set the stage for better understanding the causal mechanisms underlying arousal-based attention lapsing and its effects on episodic retrieval, as well as considerations for designing more personalized attention-reorienting interventions.

Topic Area: LONG-TERM MEMORY: Episodic