A dissociation between motor and declarative memory for duration

Poster Session E - Monday, March 9, 2026, 2:30 – 4:30 pm PDT, Fairview/Kitsilano Ballrooms

Khayla Santiago¹ (khaylasantiago2030@u.northwestern.edu); ¹Northwestern University

Our recent work has demonstrated that humans can learn up to 12 arbitrary associations between words and their presentation durations (0.3-5s). These associations can be reliably retrieved via duration reproduction— robust motor-based memory. In contrast, participants performed at chance when retrieving the same learned word-durations, indicating which of a pair of words was associated with a longer duration—poor declarative memory. While the prior study examined duration encoding with a “what” feature, the current study used a “where” feature. Participants (n = 25) learned 12 durations (0.3-5s) associated with 12 locations by reproducing the duration of a square presented at each location four times in a fixed order. Following learning, participants twice reproduced the remembered duration at each location separated by a pairwise comparison task where they indicated the location associated with the longer duration for each pair of locations. Participants completed these tasks twice, first with visually presented squares and then with audiovisually presented squares (accompanied by a concurrent tone) during learning. To avoid memory confusion, a circular layout of locations was used in one condition and a grid layout in the other (counterbalanced). Motor duration memory was reliable and was improved by audiovisual presentation as in our prior study using words. Interestingly, declarative duration memory, which was at chance level with words, became reliable and substantially improved by audiovisual presentation with spatial encoding. These results suggest that while motor duration memory is equivalently supported by what and where features, declarative duration memory substantially benefits from spatial encoding.

Topic Area: PERCEPTION & ACTION: Multisensory