# Passive, mobile assessment of sleep, circadian timing, and keyboard dynamics to prospectively predict depression severity, cognition, emotion processing, and emotion regulation

> **NIH NIH R21** · UNIVERSITY OF ILLINOIS AT CHICAGO · 2020 · $199,875

## Abstract

Depression is a prevalent, debilitating illness characterized by emotional dysregulation and
cognitive impairment. Yet, our understanding of depression remains inadequate due in part to
features of the illness that are difficult to measure such as disturbances in sleep, circadian
disorganization, and atypical moment-to-moment variation in affective state. Actigraphy and
‘BiAffect’ are well-suited to evaluate these aspects of depression. Actigraphy utilizes
accelerometer technology to monitor rest-activity patterns and is validated to estimate naturalistic
sleep and circadian timing. Disturbances in sleep, in particular insomnia, and circadian timing
(e.g., delayed sleep timing) are highly prevalent in depression and evidence suggests they play
a role in depression severity and symptomatology (e.g., emotional dysregulation, cognitive
difficulties). BiAffect is an innovative smartphone app comprising a secure virtual keyboard that
utilizes dynamic variation in typing behavior that is sensitive to mood and cognitive function
dynamics. Separate lines of research provide support for these technologies in the study of
depression. Recent pilot data comprising 28+ un-medicated patients with primary or comorbid
depression showed more fragmented sleep over the course of 1 week was significantly
associated with greater depression level. More fragmented sleep also corresponded with more
atypical brain response during emotion processing (e.g., less mid-frontal neural activity during
error detection), independent of depression severity. For BiAffect, our published data showed
depression was significantly predicted by typing behavior and movement; specifically, greater
severity was predicted by more interkey delay, more autocorrect rate, and more accelerometer
displacement in 7 patients over a 6-week period suggesting less focus/concentration and/or more
psychomotor activity (e.g., agitation) portended depression severity. Pilot data also demonstrated
interkey delay dynamics reflected diurnal patterns indicating BiAffect may serve as a proxy of
circadian organization. Altogether, findings provide support for the feasibility of BiAffect. The
proposed 2-year study endeavors to validate the novel BiAffect app and fill important gaps in the
literature. Over the course of 6 weeks we will combine wrist actigraphy with Biaffect in 70
participants with depression, 50 participants with insomnia, and 50 healthy controls. We expect
actigraphy and BiAffect data will each prospectively predict weekly depression severity and
cognitive function and bi-weekly neurocognitive and brain-behavioral response during emotion
processing and emotion regulation. We expect these effects will be more robust in the depressed
group relative to the insomnia group, which will be more robust compared to healthy controls.

## Key facts

- **NIH application ID:** 10016797
- **Project number:** 5R21MH121852-02
- **Recipient organization:** UNIVERSITY OF ILLINOIS AT CHICAGO
- **Principal Investigator:** Heide Klumpp
- **Activity code:** R21 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2020
- **Award amount:** $199,875
- **Award type:** 5
- **Project period:** 2019-09-11 → 2022-06-30

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10016797

## Citation

> US National Institutes of Health, RePORTER application 10016797, Passive, mobile assessment of sleep, circadian timing, and keyboard dynamics to prospectively predict depression severity, cognition, emotion processing, and emotion regulation (5R21MH121852-02). Retrieved via AI Analytics 2026-05-24 from https://api.ai-analytics.org/grant/nih/10016797. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*