A sleep and memory connection with important qualifications
People who reported poorer sleep also tended to retain less information in a visual memory task in one subgroup of a new US study.
The association was observed among 56 Black adults, while the researchers did not detect a comparable association among 158 White adults.
The study involved 214 adults aged 18 to 74 and was published in Frontiers in Sleep on 14 October 2026.
It also examined depression and experiences of discrimination, factors that may overlap with sleep and memory difficulties.
Crucially, the research was observational: it cannot show that poor sleep caused memory loss or explain differences between populations.
What the scientists wanted to know
Sleep has long been studied as a factor in remembering information after a delay.
However, not every memory task measures the same abilities, and not every population has been equally represented in sleep research.
The team wanted to investigate the retention of associations between objects and their visual contexts.
They also asked whether age, depressive symptoms and discrimination were related to the observed sleep-memory patterns.
Their analysis separated participants into self-identified Black and White groups to investigate within-group associations.
Why the research focused on context
Remembering an object is not the same as remembering where or with whom it appeared.
The researchers were interested in episodic memory, which includes details about an event or experience.
Their task required people to recognise whether an object was paired with the same face or scene as before.
A participant might recognise a familiar object yet confuse the scene in which it originally appeared.
That distinction creates a demanding test of the information linking different elements of a memory.
The people who took part
The participants were US residents recruited through the online research platform Prolific.
The final analysis included 56 non-Hispanic Black adults and 158 non-Hispanic White adults.
The mean ages were 36.5 years in the Black group and 43.9 years in the White group.
All participants completed the relevant questionnaires and both sessions of the memory task.
The sample was not a representative survey of the US population, and the subgroup sizes were unequal.
Why the final sample was smaller
Initially, 278 people completed the tasks needed to measure memory retention.
The researchers excluded one extreme memory score and 63 people with incomplete questionnaire data.
That left 214 participants for the main analysis.
Excluding incomplete cases can simplify analysis but may also influence which experiences are represented.
The authors acknowledged that their sample size limited the ability to detect small effects or examine additional subgroups.
How the first memory session worked
Participants viewed pairs made from an object and either a face or a scene.
The researchers used 64 object-context pairs in the main encoding task.
People rated how easy it was to imagine the two items together.
They then completed a recognition test to identify which pairings matched the originals.
This immediate assessment established a starting point for measuring subsequent retention.
A second test after a delay
Participants returned for another online session roughly 24 to 36 hours later.
Before being tested on the original images, they learned a new set of pairings.
Some familiar objects appeared alongside different contextual images.
This created interference, making it harder to rely on simple familiarity when recalling the original combinations.
The delayed test then asked participants to identify the pairings from the first session rather than the newer material.
What memory retention actually meant
The researchers used a signal-detection measure called d-prime to assess recognition accuracy.
This approach considers both correct identifications and false alarms.
Retention was calculated by comparing delayed performance with immediate performance.
More negative retention scores represented greater loss of accuracy across the interval.
The study did not measure whether participants developed dementia or experienced clinically meaningful memory decline.
How sleep quality was measured
Participants completed the Pittsburgh Sleep Quality Index questionnaire.
It asks about aspects of sleep over the preceding month, including problems falling or staying asleep.
Higher scores indicate poorer reported sleep quality.
This is a widely used questionnaire, but it is not a direct recording of sleep stages or night-time movement.
The researchers did not use sleep-laboratory measurements or wearable devices to verify participants’ sleep patterns.
The central finding in the Black subgroup
Among the 56 Black participants, poorer reported sleep quality was associated with poorer memory retention.
The statistical model adjusted for age and years of education.
The reported sleep coefficient was minus 0.06, with a p-value of 0.014.
That result indicates evidence of an association within the analysed sample.
It does not tell us whether improving sleep would produce a particular improvement in a person’s memory score.
What happened in the White subgroup
The researchers did not find a statistically reliable sleep-memory association among the 158 White participants.
The reported sleep coefficient was approximately 0.003, with a p-value of 0.853.
A non-significant result is not proof that sleep and memory are unrelated in White adults.
It means this study did not detect an association under its specific conditions.
Different sample characteristics, sleep measures or memory tests might produce different findings.
A crucial point about comparing groups
Finding statistical significance in one subgroup but not another does not automatically prove the two subgroup effects differ.
A formal comparison of effects is needed to establish that claim confidently.
The authors focused on separate within-group analyses to explore patterns that pooled results might obscure.
Their results should therefore not be framed as a biological difference between racial groups.
The categories reflect social experiences and self-identification, not fixed biological explanations for memory performance.
Depressive symptoms and sleep
Participants also completed a questionnaire measuring depressive symptoms.
In both groups, higher depressive symptom scores were associated with poorer reported sleep.
The researchers investigated whether depressive symptoms statistically explained the sleep-memory relationship.
They did not find significant mediation effects in either group.
The finding does not rule out a role for mental health; it limits what this particular analysis can establish.
The role of reported discrimination
The Black participants reported experiences of discrimination using a separate questionnaire.
More frequent reported discrimination was associated with poorer sleep quality and higher depressive symptoms.
It was also associated with poorer memory retention in one adjusted model.
The number of different discrimination events was linked to sleep and depressive symptoms, but not significantly to retention.
These findings raise questions about the combined influence of stress and sleep, without proving a causal pathway.
Why the mediation tests matter
The team tested whether discrimination could statistically account for the observed association between sleep and memory.
Neither the number of discrimination events nor their reported frequency produced a significant indirect effect.
That means the study did not establish discrimination as the mechanism linking sleep quality and memory retention.
A factor can be associated with two outcomes without explaining the relationship between them.
The authors called for larger and longer studies to investigate these overlapping influences.
Did age change the association?
The study included adults from early adulthood into their seventies.
Researchers tested whether age altered the relationship between sleep quality and memory retention.
They did not detect a significant age interaction in either subgroup.
This does not mean age is irrelevant to sleep or memory generally.
The study may have been too small or too dependent on self-reported sleep to identify subtle age-related patterns.
Why the interference task was useful
A memory test becomes more challenging when similar information is introduced before recall.
The researchers deliberately added new object-context pairings before the delayed assessment.
This required participants to distinguish the original associations from recently learned alternatives.
The authors expected this design to place additional demands on cognitive control.
However, the observed difference between interference and non-interference trial types was not statistically significant, so those results were combined.
The limits of an online experiment
Participants completed their tasks remotely rather than in a controlled sleep or memory laboratory.
The researchers could not fully standardise background noise, interruptions or individual testing environments.
Internet access and willingness to join online research may also affect who participates.
These features limit how closely the sample reflects people with fewer digital resources.
The findings should be tested again in more varied community settings with objective sleep measurements.
Why correlation cannot answer the causal question
Sleep quality, mood, stress and cognitive performance can influence one another.
The study measured habitual sleep quality and other experiences rather than randomly changing them.
It therefore cannot determine which factor came first or whether unmeasured variables contributed to the results.
The researchers also relied on data collected during 2021, although the paper was published in 2026.
A longitudinal design or carefully controlled intervention would be needed to test whether changing sleep alters later memory retention.
What this means for sleep research
The findings add to evidence that sleep and cognition should be studied in diverse populations.
They also show why researchers need to consider social experiences rather than assuming one universal explanation.
Future studies could combine sleep trackers, laboratory memory tasks and repeated assessments over time.
Larger samples would allow more precise comparisons across demographic groups.
Research Today has also covered how a nap affected children’s contextual memory and how sleep loss may affect the control of unwanted thoughts.
The bottom line
In an online study of 214 US adults, poorer self-reported sleep was linked to weaker retention of visual associations among 56 Black participants.
The same association was not statistically detected among 158 White participants.
Higher depressive symptoms were linked to poorer sleep in both groups, and reported discrimination showed additional associations in the Black subgroup.
Neither depression nor discrimination was confirmed as a statistical mediator of the sleep-memory relationship.
The research identifies patterns worth investigating, not evidence that racial identity causes different memory abilities or that improving sleep will necessarily reverse a measured deficit.
Source Information
Original study: Black adults show cross-sectional associations between sleep quality and episodic memory retention: an investigation of underlying factors.
Authors: Emily Hokett, Jess Campbell, Aditi Arunmozhi and Audrey Duarte.
Journal: Frontiers in Sleep, 14 October 2026.
Method: Cross-sectional analysis of sleep questionnaires, depression and discrimination measures and two-session online visual associative-memory tasks in 214 US adults.
