Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Jul 29;2:73.
doi: 10.1038/s41746-019-0147-4. eCollection 2019.

Capturing Sleep-Wake Cycles by Using Day-To-Day Smartphone Touchscreen Interactions

Affiliations
Free PMC article

Capturing Sleep-Wake Cycles by Using Day-To-Day Smartphone Touchscreen Interactions

Jay N Borger et al. NPJ Digit Med. .
Free PMC article

Abstract

Body movements drop with sleep, and this behavioural signature is widely exploited to infer sleep duration. However, a reduction in body movements may also occur in periods of intense cognitive activity, and the ubiquitous use of smartphones may capture these wakeful periods otherwise hidden in the standard measures of sleep. Here, we continuously captured the gross body movements using standard wrist-worn accelerometers to quantify sleep (actigraphy) and logged the timing of the day-to-day touchscreen events ('tappigraphy'). Using these measures, we addressed how the gross body movements overlap with the cognitively engaging digital behaviour (from n = 79 individuals, accumulating ~1400 nights). We find that smartphone use was distributed across a broad spectrum of physical activity levels, but consistently peaked at rest. We estimated the putative sleep onset and wake-up times from the actigraphy data to find that these times were well correlated to the estimates from tappigraphy (R2 = 0.9 for sleep-onset time and wake-up time). However, actigraphy overestimated sleep as virtually all of the users used their phones during the putative sleep period. Interestingly, the probability of touches remained greater than zero for ~2 h after the putative sleep onset, and ~2 h before the putative wake-up time. Our findings suggest that touchscreen interactions are widely integrated into modern sleeping habits-surrounding both sleep onset and waking-up periods-yielding a new approach to measuring sleep. Smartphone interactions can be leveraged to update the behavioural signatures of sleep with these peculiarities of modern digital behaviour.

Keywords: Biomarkers; Human behaviour; Risk factors.

Conflict of interest statement

Competing interestsArko Ghosh is a co-founder of QuantActions Ltd, Lausanne, Switzerland. This company focuses on converting smartphone taps to mental health indicators. Software and data collection services from QuantActions were used to monitor smartphone activity.

Figures

Fig. 1
Fig. 1
The relationship between smartphone touchscreen interactions and gross movements. a The data from a single participant showing the extent of the overlap between smartphone interactions quantified using an App running in the background (‘tappigraphy’) and overall physical activity measured at the wrist and quantified using an actigraphy algorithm, where D ≤ 1 is indicative of physical rest. b The probability of smartphone interactions in 1-minute bins at different levels of physical activity quantified in steps of 0.25 D. The subjects are sorted according to the amount of smartphone activity at D = [0–0.25]
Fig. 2
Fig. 2
Comparison of tappigraphy-based sleep estimates with that of actigraphy. a Actigraphy watches were used to quantify the amount of ambient light, near-body temperature (not shown) and the body movements. The smartphone touches were simultaneously recorded by using an App running in the background. We used the Cole–Krpike algorithm to extract the putative sleep times from actigraphy, and a new algorithm was designed to extract the sleep times from the smartphone touches (‘tappigraphy’). Reflective example of putative sleep and wake estimates is presented in ‘a’ bottom panel. The relationship between the putative sleep-onset times (b) and wake-up times (c) determined by using actigraphy versus tappigraphy. The data were pooled by concatenation across all the subjects. The time of the day is in local time, and the inserts show the relationship between tappigraphy versus sleep diaries. PDE probability density estimate based on kernel smoothing function
Fig. 3
Fig. 3
Inter-individual differences in tappigraphy and actigraphy sleep estimates. The individual central tendencies (median) of actigraphy-based putative wake-up (a) and sleep (b) times were well correlated to the putative times determined using tappigraphy. c The distribution of median measurement error (estimated from each individual) in tappigraphy considering actigraphy as ground truth observed in the sampled population. d The relationship between the individualised measurement error and smartphone usage in the sampled population
Fig. 4
Fig. 4
Prevalence of smartphone touches during actigraphy-inferred sleep. a Kernel density plot of the probability of observing smartphone touches during ‘sleep’ in the sampled population—based on the percentage of nights with smartphone touches extracted from each participant. b The break down of the number of touches generated after sleep onset and before wake-up times with each participant represented using a different line. The probability of detecting smartphone touches and gross body movements in 3 -min bins after sleep onset (c) and before wake-up time (d). The dark line represents the central tendency of the population, and the shaded area represents the standard deviation. Note, according to the central tendencies, smartphone touches are observed with a probability of ~0.4 right after sleep onset and right before wake-up time. The statistical testing against a probability of 0 was corrected for multiple comparison correction by using Bonferroni correction
Fig. 5
Fig. 5
Integrating touchscreen interactions with actigraphy-derived sleep times to quantify sleep disturbances. a The metric of sleep fracture fraction (SFF) was derived from each night, which is essentially the longest period of smartphone un-interrupted sleep normalised to the total sleep duration according to actigraphy. b The population distribution of the 25th percentile (50th percentile as insert) values of the SFF. Note, SFF of 1 would indicate un-interrupted sleep and SFF of 0.5 indicates continuous sleep accounting for only half of the sleep duration

Similar articles

See all similar articles

Cited by 1 article

References

    1. Sadeh A, Hauri PJ, Kripke DF, Lavie P. The role of actigraphy in the evaluation of sleep disorders. Sleep. 1995;18:288–302. doi: 10.1093/sleep/18.4.288. - DOI - PubMed
    1. Martin JL, Hakim AD. Wrist actigraphy. Chest. 2011;139:1514–1527. doi: 10.1378/chest.10-1872. - DOI - PMC - PubMed
    1. Cole RJ, Kripke DF, Gruen W, Mullaney DJ, Gillin JC. Automatic sleep/wake identification from wrist activity. Sleep. 1992;15:461–469. doi: 10.1093/sleep/15.5.461. - DOI - PubMed
    1. Roenneberg T. Twitter as a means to study temporal behaviour. Curr. Biol. 2017;27:R830–R832. doi: 10.1016/j.cub.2017.08.005. - DOI - PubMed
    1. Leypunskiy E, et al. Geographically resolved rhythms in twitter use reveal social pressures on daily activity patterns. Curr. Biol. 2018;28:3763–3775.e5. doi: 10.1016/j.cub.2018.10.016. - DOI - PMC - PubMed

LinkOut - more resources

Feedback