Does sleep change how accurately we remember a story?

Details

The episodic context account (Gaskell et al., 2019) proposes that understanding language creates an episodic representation of a discourse and that this representation is susceptible to sleep-related memory effects.

The researchers tested this prediction in three experiments. Participants read or listened to naturalistic stories, and their memory was tested after a 12-hour interval that included either daytime wakefulness or overnight sleep. Discourse memory was assessed using sentence recognition in Experiment 1 (N = 386), free story recall in Experiment 2 (N = 96), and cued recall in Experiments 2 and 3 (N = 192).

There was no evidence of sleep-related effects in sentence recognition or free recall. In cued recall, a fill-in-the-blank task, time-related distortion was lower after sleep than after wakefulness, as measured by both subjective categorisation and Latent Semantic Analysis.

The researchers suggest that sleep’s effect on discourse memory is modest but observable. They propose that it may be constrained by retrieval processes, including recollection versus familiarity and associative versus item retrieval; may occur at a qualitative level that is difficult to detect with an all-or-nothing scoring metric; and may be situated primarily at the textbase level of the tripartite model of discourse processing.