Common multi-day rhythms in smartphone behavior

Abstract

The idea that abnormal human activities follow multi-day rhythms is found in ancient beliefs on the moon to modern clinical observations in epilepsy and mood disorders. To explore multi-day rhythms in healthy human behavior our analysis includes over 300 million smartphone touchscreen interactions logging up to 2 years of day-to-day activities (N401 subjects). At the level of each individual, we find a complex expression of multi-day rhythms where the rhythms occur scattered across diverse smartphone behaviors. With non-negative matrix factorization, we extract the scattered rhythms to reveal periods ranging from 7 to 52 days – cutting across age and gender. The rhythms are likely free-running – instead of being ubiquitously driven by the moon – as they did not show broad population-level synchronization even though the sampled population lived in northern Europe. We propose that multi-day rhythms are a common trait, but their consequences are uniquely experienced in day-to-day behavior.

Information Dynamics of the Heart and Respiration Rates: a Novel Venue for Digital Phenotyping in Humans

“…In this study, we used freely-behaving human subjects’ longitudinal HR and RR recordings to investigate their utility for humans’ cardiorespiratory digital phenotyping [8, 83–86]. While doing so, we placed emphasis on interpretability of our analyses.…”

“…It goes with-out saying that a number of studies have utilized the digital technologies to successfully identify several behavioral signatures of physiological phenotypes. These included the pain severity [20], sleep patterns [21], behavioral traits in smartphone use [22], and the potential correlate of HRV and physical activity [23]. However, the extension of these results to continuous monitoring of humans’ physiology is nontrivial.…”

“…Using arm acceleration data, Katori et al [21] found sleep phenotypes that associated with social jet lag, individuals’ chronotype, and insomnia. The use of mobile technologies for humans’ phenotyping becomes more relevant, given the common rhythms in their multi-day traits of use by individuals [22].…”

“…A rich body of research highlights the utility of wearable devices for digital health phenotyping [17][18][19][20][21][22][23]. For instance, Smets et al [24] derived digital phenotypes from individuals' five-day physiological and contextual measurements.…”

Real-world smartphone data can trace the behavioural impact of epilepsy: A Case study

“…We employed a permutation-based clustering approach to identify clusters of 2D bins with significant differences in touchscreen interactions and to account for multiple testing associated with many individual t-tests (2500 t-statistics) when comparing JIDs bin-wise. 9 Specifically, we performed 1000 random label permutations for each comparison (pre-tapering vs. tapering and tapering vs. post-tapering). At each iteration, we did a t-test for all 2D bins within the JIDs and identified clusters in the 2D space of t-statistics.…”