Stimulus-driven and goal-directed orienting and encoding into memory across space and time

Poster Presentation: Friday, May 16, 2025, 3:00 – 5:00 pm, Banyan Breezeway
Session: Attention: Inattention, load

Carly Chak1, Elise Kormos2, Barry Giesbrecht3; 1University of California, Santa Barbara, 2Institute for Collaborative Biotechnologies

The selection and encoding of stimuli into working memory is dependent on the interplay between stimulus-driven and goal-directed attention. Typically, the contributions of stimulus-driven and goal-directed attention to the selection and encoding of information into working memory are studied separately. Here, we examined how these types of attention contributed to the selection and encoding of information as a function of working memory demands and individual differences in working memory capacity. Subjects (n=49) completed a spatially distributed n-back task where numbers were presented across six locations and prompted to report the 0-back, 1-back, or 2-back item. In the stimulus-driven condition, one number was displayed per frame. In the goal-directed condition, numbers were presented in each of the six locations all at once, but one cued the location of the next number to be attended. Accuracy declined when the to-be-reported number was temporally distant (η2 = .11, p < .001) and when attention was directed by the preceding number in the sequence (η2 = .41, p < .001). The magnitude of the n-back effect did not vary based on working memory capacity in either of the attention conditions. However, participants with low working memory capacity exhibited a speed-accuracy tradeoff in the goal-directed orienting condition such that they were slower in the 2-back condition, but maintained accuracy (r = .78, p = .002). Altogether, the results suggest that goal-directed orienting is particularly costly while maintaining information in working memory. Additionally, for participants with low working memory capacity, the cost manifests as a delayed response time to the n-back with preserved accuracy. This suggests that low capacity participants may volitionally encode information in space, but the retrieval of target information may involve filtering out other information (e.g., other encoded numbers or location cues).

Acknowledgements: This research was sponsored by the U.S. Army Research Office and accomplished under cooperative agreement W911NF-19-2-0026 for the Institute of Collaborative Biotechnologies.