Large cognitive fluctuations surrounding sleep in daily living

Age-related behavioral resilience in smartphone touchscreen interaction dynamics

We experience a life that is full of ups and downs. The ability to bounce back after adverse life events such as the loss of a loved one or serious illness declines with age, and such isolated events can even trigger accelerated aging. How humans respond to common day-to-day perturbations is less clear. Here, we infer the aging status from smartphone behavior by using a decision tree regression model trained to accurately estimate the chronological age based on the dynamics of touchscreen interactions. Individuals (N = 280, 21 to 87 y of age) expressed smartphone behavior that appeared younger on certain days and older on other days through the observation period that lasted up to ~4 y. We captured the essence of these fluctuations by leveraging the mathematical concept of critical transitions and tipping points in complex systems. In most individuals, we find one or more alternative stable aging states separated by tipping points. The older the individual, the lower the resilience to forces that push the behavior across the tipping point into an older state. Traditional accounts of aging based on sparse longitudinal data spanning decades suggest a gradual behavioral decline with age. Taken together with our current results, we propose that the gradual age-related changes are interleaved with more complex dynamics at shorter timescales where the same individual may navigate distinct behavioral aging states from one day to the next. Real-world behavioral data modeled as a complex system can transform how we view and study aging.

Examining Human-Smartphone Interaction as a Proxy for Circadian Rhythm in Patients With Insomnia: Cross-Sectional Study

BACKGROUND

The sleep and circadian rhythm patterns associated with smartphone use, which are influenced by mental activities, might be closely linked to sleep quality and depressive symptoms, similar to the conventional actigraphy-based assessments of physical activity.

OBJECTIVE

The primary objective of this study was to develop app-defined circadian rhythm and sleep indicators and compare them with actigraphy-derived measures. Additionally, we aimed to explore the clinical correlations of these indicators in individuals with insomnia and healthy controls.

METHODS

The mobile app "Rhythm" was developed to record smartphone use time stamps and calculate circadian rhythms in 33 patients with insomnia and 33 age- and gender-matched healthy controls, totaling 2097 person-days. Simultaneously, we used standard actigraphy to quantify participants' sleep-wake cycles. Sleep indicators included sleep onset, wake time (WT), wake after sleep onset (WASO), and the number of awakenings (NAWK). Circadian rhythm metrics quantified the relative amplitude, interdaily stability, and intradaily variability based on either smartphone use or physical activity data.

RESULTS

Comparisons between app-defined and actigraphy-defined sleep onsets, WTs, total sleep times, and NAWK did not reveal any significant differences (all P>.05). Both app-defined and actigraphy-defined sleep indicators successfully captured clinical features of insomnia, indicating prolonged WASO, increased NAWK, and delayed sleep onset and WT in patients with insomnia compared with healthy controls. The Pittsburgh Sleep Quality Index scores were positively correlated with WASO and NAWK, regardless of whether they were measured by the app or actigraphy. Depressive symptom scores were positively correlated with app-defined intradaily variability (β=9.786, SD 3.756; P=.01) and negatively correlated with actigraphy-based relative amplitude (β=-21.693, SD 8.214; P=.01), indicating disrupted circadian rhythmicity in individuals with depression. However, depressive symptom scores were negatively correlated with actigraphy-based intradaily variability (β=-7.877, SD 3.110; P=.01) and not significantly correlated with app-defined relative amplitude (β=-3.859, SD 12.352; P=.76).

CONCLUSIONS

This study highlights the potential of smartphone-derived sleep and circadian rhythms as digital biomarkers, complementing standard actigraphy indicators. Although significant correlations with clinical manifestations of insomnia were observed, limitations in the evidence and the need for further research on predictive utility should be considered. Nonetheless, smartphone data hold promise for enhancing sleep monitoring and mental health assessments in digital health research.

Common multi-day rhythms in smartphone behavior

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.

Spontaneous motor tempo over the course of a week: the role of the time of the day, chronotype, and arousal

The spontaneous motor tempo (SMT) or internal tempo describes the natural pace of predictive and emergent movements such as walking or hand clapping. One of the main research interests in the study of the spontaneous motor tempo relates to factors affecting its pace. Previous studies suggest an influence of the circadian rhythm (i.e., 24-h cycle of the biological clock), physiological arousal changes, and potentially also musical experience. This study aimed at investigating these effects in participants' everyday life by measuring their SMT four times a day over seven consecutive days, using an experience sampling method. The pace of the SMT was assessed with a finger-tapping paradigm in a self-developed web application. Measured as the inter-tap interval, the overall mean SMT was 650 ms (SD = 253 ms). Using multi-level modelling (MLM), results show that the pace of the SMT sped up over the course of the day, and that this effect depended on the participants' chronotype, since participants tending towards morning type were faster in the morning compared to participants tending towards evening type. During the day, the pace of the SMT of morning types stayed relatively constant, whereas it became faster for evening-type participants. Furthermore, higher arousal in participants led to a faster pace of the SMT. Musical sophistication did not influence the SMT. These results indicate that the circadian rhythm influences the internal tempo, since the pace of SMT is not only dependent on the time of the day, but also on the individual entrainment to the 24-h cycle (chronotype).

A model of healthy aging based on smartphone interactions reveals advanced behavioral age in neurological disease

Smartphones offer unique opportunities to trace the convoluted behavioral patterns accompanying healthy aging. Here we captured smartphone touchscreen interactions from a healthy population (N = 684, ∼309 million interactions) spanning 16 to 86 years of age and trained a decision tree regression model to estimate chronological age based on the interactions. The interactions were clustered according to their next interval dynamics to quantify diverse smartphone behaviors. The regression model well-estimated the chronological age in health (mean absolute error = 6 years, R2 = 0.8). We next deployed this model on a population of stroke survivors (N = 41) to find larger prediction errors such that the estimated age was advanced by 6 years. A similar pattern was observed in people with epilepsy (N = 51), with prediction errors advanced by 10 years. The smartphone behavioral model trained in health can be used to study altered aging in neurological diseases.

Disentangling personalized treatment effects from “time-of-the-day” confounding in mobile health studies

Ideally, a patient’s response to medication can be monitored by measuring changes in performance of some activity. In observational studies, however, any detected association between treatment (“on-medication” vs “off-medication”) and the outcome (performance in the activity) might be due to confounders. In particular, causal inferences at the personalized level are especially vulnerable to confounding effects that arise in a cyclic fashion. For quick acting medications, effects can be confounded by circadian rhythms and daily routines. Using the time-of-the-day as a surrogate for these confounders and the performance measurements as captured on a smartphone, we propose a personalized statistical approach to disentangle putative treatment and “time-of-the-day” effects, that leverages conditional independence relations spanned by causal graphical models involving the treatment, time-of-the-day, and outcome variables. Our approach is based on conditional independence tests implemented via standard and temporal linear regression models. Using synthetic data, we investigate when and how residual autocorrelation can affect the standard tests, and how time series modeling (namely, ARIMA and robust regression via HAC covariance matrix estimators) can remedy these issues. In particular, our simulations illustrate that when patients perform their activities in a paired fashion, positive autocorrelation can lead to conservative results for the standard regression approach (i.e., lead to deflated true positive detection), whereas negative autocorrelation can lead to anticonservative behavior (i.e., lead to inflated false positive detection). The adoption of time series methods, on the other hand, leads to well controlled type I error rates. We illustrate the application of our methodology with data from a Parkinson’s disease mobile health study.

Temporal clusters of age-related behavioral alterations captured in smartphone touchscreen interactions

Smartphones touchscreen interactions may help resolve if and how real-world behavioral dynamics are shaped by aging. Here, in a sample spanning the adult life span (16 to 86 years, N = 598, accumulating 355 million interactions), we clustered the smartphone interactions according to their next inter-touch interval dynamics. There were age-related behavioral losses at the clusters occupying short intervals (∼100 ms, R² ∼ 0.8) but gains at the long intervals (∼4 s, R² ∼ 0.4). Our approach revealed a sophisticated form of behavioral aging where individuals simultaneously demonstrated accelerated aging in one behavioral cluster versus a deceleration in another. Contrary to the common notion of a simple behavioral decline with age based on conventional cognitive tests, we show that the nature of aging systematically varies according to the underlying dynamics. Of all the imaginable factors determining smartphone interactions, age-sensitive cognitive and behavioral processes may dominatingly shape smartphone dynamics.

•

The timing of smartphone touchscreen interactions varies from one person to the next

•

A diverse range of interactions was studied by quantifying the next interval dynamics

•

The smartphone dynamics reflected the performance in cognitive tests and age

•

The way aging shapes behavior may depend on its underlying temporal dynamics

Smartphone keyboard interaction monitoring as an unobtrusive method to approximate rest-activity patterns: Inter-individual and metric-specific variations (Preprint)

BACKGROUND

Sleep is an important determinant of individuals' health and behavior during the wake phase. Novel research methods for field assessments are required to enable the monitoring of sleep over a prolonged period and across a large number of people. The ubiquity of smartphones offers new avenues for detecting rest-activity patterns in everyday life in a noninvasive an inexpensive manner and on a large scale. Recent studies provided evidence for the potential of smartphone interaction monitoring as a novel tracking method to approximate rest-activity patterns based on the timing of smartphone activity and inactivity throughout the 24-hour day. These findings require further replication and more detailed insights into interindividual variations in the associations and deviations with commonly used metrics for monitoring rest-activity patterns in everyday life.

OBJECTIVE

This study aimed to replicate and expand on earlier findings regarding the associations and deviations between smartphone keyboard-derived and self-reported estimates of the timing of the onset of the rest and active periods and the duration of the rest period. Moreover, we aimed to quantify interindividual variations in the associations and time differences between the 2 assessment modalities and to investigate to what extent general sleep quality, chronotype, and trait self-control moderate these associations and deviations.

METHODS

Students were recruited to participate in a 7-day experience sampling study with parallel smartphone keyboard interaction monitoring. Multilevel modeling was used to analyze the data.

RESULTS

In total, 157 students participated in the study, with an overall response rate of 88.9% for the diaries. The results revealed moderate to strong relationships between the keyboard-derived and self-reported estimates, with stronger associations for the timing-related estimates (β ranging from .61 to .78) than for the duration-related estimates (β=.51 and β=.52). The relational strength between the time-related estimates was lower, but did not substantially differ for the duration-related estimates, among students experiencing more disturbances in their general sleep quality. Time differences between the keyboard-derived and self-reported estimates were, on average, small (<0.5 hours); however, large discrepancies were also registered for quite some nights. The time differences between the 2 assessment modalities were larger for both timing-related and rest duration-related estimates among students who reported more disturbances in their general sleep quality. Chronotype and trait self-control did not significantly moderate the associations and deviations between the 2 assessment modalities.

CONCLUSIONS

We replicated the positive potential of smartphone keyboard interaction monitoring for estimating rest-activity patterns among populations of regular smartphone users. Chronotype and trait self-control did not significantly influence the metrics' accuracy, whereas general sleep quality did: the behavioral proxies obtained from smartphone interactions appeared to be less powerful among students who experienced lower general sleep quality. The generalization and underlying process of these findings require further investigation.

Artificial neural network trained on smartphone behavior can trace epileptiform activity in epilepsy

A range of abnormal electrical activity patterns termed epileptiform discharges can occur in the brains of persons with epilepsy. These epileptiform discharges can be monitored and recorded with implanted devices that deliver therapeutic neurostimulation. These continuous recordings provide an opportunity to study the behavioral correlates of epileptiform discharges as the patients go about their daily lives. Here, we captured the smartphone touchscreen interactions in eight patients in conjunction with electrographic recordings (accumulating 35,714 h) and by using an artificial neural network model addressed if the behavior reflected the epileptiform discharges. The personalized model outputs based on smartphone behavioral inputs corresponded well with the observed electrographic data (R: 0.2-0.6, median 0.4). The realistic reconstructions of epileptiform activity based on smartphone use demonstrate how day-to-day digital behavior may be converted to personalized markers of disease activity in epilepsy.

Spontaneous Motor Tempo: Investigating Psychological, Chronobiological, and Demographic Factors in a Large-Scale Online Tapping Experiment

The spontaneous motor tempo (SMT) describes the pace of regular and repeated movements such as hand clapping or walking. It is typically measured by letting people tap with their index finger at a pace that feels most natural and comfortable to them. A number of factors have been suggested to influence the SMT, such as age, time of the day, arousal, and potentially musical experience. This study aimed at investigating the effects of these factors in a combined and out-of-the-lab context by implementing the finger-tapping paradigm in an online experiment using a self-developed web application. Due to statistical multimodality in the distribution of participants' SMT (N = 3,576), showing peaks at modes of around 250 ms, a Gaussian mixture model was applied that grouped participants into six clusters, ranging from Very Fast (M = 265 ms, SD = 74) to Very Slow (M = 1,757 ms, SD = 166). These SMT clusters differed in terms of age, suggesting that older participants had a slower SMT, and time of the day, showing that the earlier it was, the slower participants' SMT. While arousal did not differ between the SMT clusters, more aroused participants showed faster SMTs across all normalized SMT clusters. Effects of musical experience were inconclusive. With a large international sample, these results provide insights into factors influencing the SMT irrespective of cultural background, which can be seen as a window into human timing processes.