No evidence that overnight sleep improves associative inference

Sleep can help preserve learned associations, but evidence that it also helps people draw new inferences from them is mixed. We tested both outcomes in the same task, comparing performance after a 12-hour overnight interval with performance after a daytime interval of the same length. In an online experiment, participants learned overlapping object pairs (A–B and B–C) and were tested on them immediately and again after the delay. Only at the delayed test were they asked to infer the relationship between objects linked through the shared item (A–C). We analysed accuracy and response times using Bayesian multilevel models. Memory for the learned pairs declined in both groups, and inference accuracy remained close to chance. Inferences were more likely to be correct when participants also recognised both of the pairs needed to make them. Overall, the results gave no clear evidence of an overnight benefit for memory or inference and were consistent with participants retrieving and combining the learned pairs at test.

Authors

Publication Details

Journal
PsyArXiv (OSF Preprints)
Published
2026-11-15
DOI
https://doi.org/10.31234/osf.io/836jh_v1
Primary Topic
Sleep and Wakefulness Research
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

No evidence that overnight sleep improves associative inference

Johannes Wibroe, Shane Lindsay
PsyArXiv (OSF Preprints)
Sleep and Wakefulness Research
preprint

No evidence that overnight sleep improves associative inference

Johannes Wibroe, Shane Lindsay
preprint en

Abstract

Sleep can help preserve learned associations, but evidence that it also helps people draw new inferences from them is mixed. We tested both outcomes in the same task, comparing performance after a 12-hour overnight interval with performance after a daytime interval of the same length. In an online experiment, participants learned overlapping object pairs (A–B and B–C) and were tested on them immediately and again after the delay. Only at the delayed test were they asked to infer the relationship between objects linked through the shared item (A–C). We analysed accuracy and response times using Bayesian multilevel models. Memory for the learned pairs declined in both groups, and inference accuracy remained close to chance. Inferences were more likely to be correct when participants also recognised both of the pairs needed to make them. Overall, the results gave no clear evidence of an overnight benefit for memory or inference and were consistent with participants retrieving and combining the learned pairs at test.

PsyArXiv (OSF Preprints)
Sleep and Wakefulness Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon