Sleep

An open science resource for accelerating scalable digital health research in autism and other neurodevelopmental conditions.

TL;DR

The Simons Sleep Project is an open-science resource containing data from >3,600 days and nights from 102 autistic children and 98 nonautistic siblings, demonstrating that digital devices have higher accuracy and reliability compared to parent reports, and that autistic children have longer sleep-onset latencies associated with behavioral difficulties.

Key Findings

The Simons Sleep Project (SSP) contains data from over 3,600 days and nights collected from 102 autistic children and 98 nonautistic siblings aged 10-17 years.

  • Participants included children with idiopathic autism and their nonautistic siblings
  • Age range of participants was 10-17 years
  • Data were collected using Dreem3 EEG headbands, multi-sensor EmbracePlus smartwatches, and Withings' sleep mats
  • The dataset also includes parent questionnaires, daily sleep diaries, and whole-exome sequencing for all participants
  • The resource is designed as an open-science platform to accelerate digital health research

Digital devices demonstrated higher accuracy and reliability compared to parent reports for measuring sleep and daily behaviors.

  • Multiple digital devices were used including EEG headbands, smartwatches, and sleep mats
  • Harmonized data across devices were presented to demonstrate breadth of available measures
  • Parent reports were used as a comparison benchmark for device accuracy
  • The finding highlights advantages of using digital devices over subjective reporting in pediatric research

Autistic children had longer sleep-onset latencies than their nonautistic siblings.

  • Sleep-onset latency differences were detected between the 102 autistic children and 98 nonautistic sibling controls
  • The sibling-control design allowed comparison within families, reducing confounding from shared environmental factors
  • Objective digital device measurements were used to assess sleep-onset latency
  • This finding replicates and extends prior literature on sleep difficulties in autism using objective digital measures

Longer sleep-onset latencies were associated with behavioral difficulties in all participants, regardless of autism diagnosis.

  • The association between sleep-onset latency and behavioral difficulties was found in both autistic children and nonautistic siblings
  • This transdiagnostic finding suggests sleep-onset latency may be a broadly relevant behavioral marker
  • Behavioral difficulties were assessed across all 200 participants (102 autistic and 98 nonautistic)
  • The result indicates that sleep problems and behavioral difficulties are linked beyond the specific context of autism

The SSP enables whole-exome sequencing access for all participants, linking genetic data to digital phenotyping measures.

  • Whole-exome sequencing is available for all 200 participants in the dataset
  • This genetic data can be linked to the multi-modal digital and behavioral data collected
  • The integration of genomic and digital phenotyping data is presented as a key opportunity afforded by the SSP
  • The dataset covers both idiopathic autism cases and nonautistic sibling controls, enabling genetic comparison studies

The SSP is presented as an open-science resource designed to support development of broad digital phenotyping techniques for autism and neurodevelopmental conditions.

  • The resource is intended to accelerate scalable digital health research
  • Multi-modal data collection included EEG, actigraphy-type wearables, sleep mats, questionnaires, and diaries
  • The harmonized dataset is made accessible to the broader research community
  • The authors highlight the resource's potential for studying autism and developing digital phenotyping methods more broadly

What This Means

This research introduces the Simons Sleep Project (SSP), a large open-science dataset designed to help researchers study sleep and daily behaviors in autistic children. The project collected more than 3,600 days and nights of data from 102 autistic children (ages 10-17) and 98 of their non-autistic siblings using wearable devices including EEG headbands, smartwatches, and under-mattress sleep sensors, alongside parent questionnaires and genetic testing. By making this data openly available, the researchers aim to speed up scientific progress in understanding how sleep problems relate to autism and behavior. The study found that electronic devices were more accurate and reliable than parent reports for measuring sleep. It also found that autistic children took longer to fall asleep compared to their non-autistic siblings — a difference captured objectively by the devices. Importantly, longer time to fall asleep was linked to more behavioral difficulties in all children, whether autistic or not, suggesting that sleep problems may broadly affect behavior across different groups of children. This research suggests that wearable technology can provide more objective and reliable information about sleep than caregiver recall alone, and that addressing sleep-onset difficulties could be relevant for children regardless of whether they have an autism diagnosis. The open-access nature of the dataset, which also includes genetic data, means other researchers around the world can use it to investigate questions about autism, sleep, and neurodevelopment without needing to collect their own large datasets from scratch.

Have a question about this study?

Citation

Hacohen M, Levy A, Kaiser H, Green Snyder L, Amatya A, Gundersen B, et al.. (2026). An open science resource for accelerating scalable digital health research in autism and other neurodevelopmental conditions.. Nature neuroscience. https://doi.org/10.1038/s41593-025-02146-3