Toward a Mobile Platform for Real-world Digital Measurement of Depression: User-Centered Design, Data Quality, and Behavioral and Clinical Modeling.

BACKGROUND: Although effective mental health treatments exist, the ability to match individuals to optimal treatments is poor, and timely assessment of response is difficult. One reason for these challenges is the lack of objective measurement of psychiatric symptoms. Sensors and active tasks recorded by smartphones provide a low-burden, low-cost, and scalable way to capture real-world data from patients that could augment clinical decision-making and move the field of mental health closer to measurement-based care. OBJECTIVE: This study tests the feasibility of a fully remote study on individuals with self-reported depression using an Android-based smartphone app to collect subjective and objective measures associated with depression severity. The goals of this pilot study are to develop an engaging user interface for high task adherence through user-centered design; test the quality of collected data from passive sensors; start building clinically relevant behavioral measures (features) from passive sensors and active inputs; and preliminarily explore connections between these features and depression severity. METHODS: A total of 600 participants were asked to download the study app to join this fully remote, observational 12-week study. The app passively collected 20 sensor data streams (eg, ambient audio level, location, and inertial measurement units), and participants were asked to complete daily survey tasks, weekly voice diaries, and the clinically validated Patient Health Questionnaire (PHQ-9) self-survey. Pairwise correlations between derived behavioral features (eg, weekly minutes spent at home) and PHQ-9 were computed. Using these behavioral features, we also constructed an elastic net penalized multivariate logistic regression model predicting depressed versus nondepressed PHQ-9 scores (ie, dichotomized PHQ-9). RESULTS: A total of 415 individuals logged into the app. Over the course of the 12-week study, these participants completed 83.35% (4151/4980) of the PHQ-9s. Applying data sufficiency rules for minimally necessary daily and weekly data resulted in 3779 participant-weeks of data across 384 participants. Using a subset of 34 behavioral features, we found that 11 features showed a significant (P<.001 Benjamini-Hochberg adjusted) Spearman correlation with weekly PHQ-9, including voice diary-derived word sentiment and ambient audio levels. Restricting the data to those cases in which all 34 behavioral features were present, we had available 1013 participant-weeks from 186 participants. The logistic regression model predicting depression status resulted in a 10-fold cross-validated mean area under the curve of 0.656 (SD 0.079). CONCLUSIONS: This study finds a strong proof of concept for the use of a smartphone-based assessment of depression outcomes. Behavioral features derived from passive sensors and active tasks show promising correlations with a validated clinical measure of depression (PHQ-9). Future work is needed to increase scale that may permit the construction of more complex (eg, nonlinear) predictive models and better handle data missingness.

Transforming Psychiatry into Data-Driven Medicine with Digital Measurement Tools.

Psychiatry has been limited by historically rooted practices centered primarily on subjective observation. Fields such as oncology have progressed toward data-driven clinical decision-making that combines subjective clinical assessment of symptoms and preferences with biological measures such as genetics, biomarkers, imaging, and integrative physiology to derive quantitative risk scores and decision support. In contrast, psychiatry has just begun to scratch the surface of measurement-based care with validated clinical questionnaires. An opportunity exists to improve modern psychiatric care with novel data streams from digital sensors combined with clinical observation and subjective self-report. The prospect of integrating this complex information with modern computational and analytical methods could advance the field, both in research and clinical practice. Here we discuss this possibility and propose some key priorities to enable these innovations toward improving clinical outcomes in the future.

Enhanced change detection performance reveals improved strategy use in avid action video game players.

Recent research has shown that avid action video game players (VGPs) outperform non-video game players (NVGPs) on a variety of attentional and perceptual tasks. However, it remains unknown exactly why and how such differences arise; while some prior research has demonstrated that VGPs' improvements stem from enhanced basic perceptual processes, other work indicates that they can stem from enhanced attentional control. The current experiment used a change-detection task to explore whether top-down strategies can contribute to VGPs' improved abilities. Participants viewed alternating presentations of an image and a modified version of the image and were tasked with detecting and localizing the changed element. Consistent with prior claims of enhanced perceptual abilities, VGPs were able to detect the changes while requiring less exposure to the change than NVGPs. Further analyses revealed this improved change detection performance may result from altered strategy use; VGPs employed broader search patterns when scanning scenes for potential changes. These results complement prior demonstrations of VGPs' enhanced bottom-up perceptual benefits by providing new evidence of VGPs' potentially enhanced top-down strategic benefits.

Generalized "satisfaction of search": adverse influences on dual-target search accuracy.

The successful detection of a target in a radiological search can reduce the detectability of a second target, a phenomenon termed satisfaction of search (SOS). Given the potential consequences, here we investigate the generality of SOS with the goal of simultaneously informing radiology, cognitive psychology, and nonmedical searches such as airport luggage screening. Ten experiments utilizing nonmedical searches and untrained searchers suggest that SOS is affected by a diverse array of factors, including (1) the relative frequency of different target types, (2) external pressures (reward and time), and (3) expectations about the number of targets present. Collectively, these experiments indicate that SOS arises when searchers have a biased expectation about the low likelihood of specific targets or events, and when they are under pressure to perform efficiently. This first demonstration of SOS outside of radiology implicates a general heuristic applicable to many kinds of searches. In an example like airport luggage screening, the current data suggest that the detection of an easy-to-spot target (e.g., a water bottle) might reduce detection of a hard-to-spot target (e.g., a box cutter).

Staying in bounds: Contextual constraints on object-file coherence.

Coherent visual perception necessitates the ability to track distinct objects as the same entities over time and motion. Calculations of such object persistence appear to be fairly automatic and constrained by specific rules. We explore the nature of object persistence here within the object-file framework; object files are mid-level visual representations that track entities over time and motion as the same persisting objects and store and update information about the objects. We present three new findings. First, objects files are constrained by the principle of "boundedness"; persisting entities should maintain a single closed contour. Second, object files are constrained by the principle of "containment"; all the parts and properties of a persisting object should reside within, and be connected to, the object itself. Third, object files are sensitive to the context in which an object appears; the very same physical entity that can instantiate object-file formation in one experimental context cannot in another. This contextual influence demonstrates for the first time that object files are sensitive to more than just the physical properties contained within any given visual display.

Que PASA? The posterior-anterior shift in aging.

A consistent finding from functional neuroimaging studies of cognitive aging is an age-related reduction in occipital activity coupled with increased frontal activity. This posterior-anterior shift in aging (PASA) has been typically attributed to functional compensation. The present functional magnetic resonance imaging sought to 1) confirm that PASA reflects the effects of aging rather than differences in task difficulty; 2) test the compensation hypothesis; and 3) investigate whether PASA generalizes to deactivations. Young and older participants were scanned during episodic retrieval and visual perceptual tasks, and age-related changes in brain activity common to both tasks were identified. The study yielded 3 main findings. First, inconsistent with a difficulty account, the PASA pattern was found across task and confidence levels when matching performance among groups. Second, supporting the compensatory hypothesis, age-related increases in frontal activity were positively correlated with performance and negatively correlated with the age-related occipital decreases. Age-related increases and correlations with parietal activity were also found. Finally, supporting the generalizability of the PASA pattern to deactivations, aging reduced deactivations in posterior midline cortex but increased deactivations in medial frontal cortex. Taken together, these findings demonstrate the validity, function, and generalizability of PASA, as well as its importance for the cognitive neuroscience of aging.

Rare targets are rarely missed in correctable search.

Failing to find a tumor in an x-ray scan or a gun in an airport baggage screening can have dire consequences, making it fundamentally important to elucidate the mechanisms that hinder performance in such visual searches. Recent laboratory work has indicated that low target prevalence can lead to disturbingly high miss rates in visual search. Here, however, we demonstrate that misses in low-prevalence searches can be readily abated. When targets are rarely present, observers adapt by responding more quickly, and miss rates are high. Critically, though, these misses are often due to response-execution errors, not perceptual or identification errors: Observers know a target was present, but just respond too quickly. When provided an opportunity to correct their last response, observers can catch their mistakes. Thus, low target prevalence may not be a generalizable cause of high miss rates in visual search.

The medial temporal lobe distinguishes old from new independently of consciousness.

Although it is widely accepted that the medial temporal lobes (MTLs) are critical for becoming aware that something happened in the past, there is virtually no evidence whether MTL sensitivity to event oldness also depends on conscious awareness. Using event-related functional magnetic resonance imaging, we show that activity in posterior MTL tracks whether an item is actually old (true oldness), regardless of participants' awareness of oldness (perceived oldness). Confirming its sensitivity to the objective nature of the stimulus, activity in this region was strongly correlated with individual memory performance (r = 0.74). At the same time, we found that memory errors (misses) were associated with activity in an anterior MTL region, which signaled whether an item was consciously experienced as new (perceived novelty). Logistic regression analyses based on individual trial activity indicated that the two MTL regions showed opposing relationships with behavior, and that memory performance was determined by their joint activity. Furthermore, functional connectivity analyses showed that perceived novelty activity in the ANTERIOR [corrected] MTL inhibited true oldness activity in the POSTERIOR [corrected] MTL. These findings indicate that participants' behavior reflected the combined effects of multiple MTL regions. More generally, our results show that parts of MTL can distinguish old from new independently of consciousness.

Role of prefrontal and anterior cingulate regions in decision-making processes shared by memory and nonmemory tasks.

In the episodic retrieval (ER) domain, activations in right dorsolateral prefrontal cortex (DLPFC) are often attributed to postretrieval monitoring. Yet, right DLPFC activations are also frequently found during nonmemory tasks. To investigate the role of this region across different cognitive functions, we directly compared brain activity during ER and visual perception (VP) using event-related functional magnetic resonance imaging. In the ER task, participants decided whether words were old or new, whereas in the VP task, they decided which of the two colored screen areas was larger. In both tasks, each decision was followed by a confidence rating. The main finding was that right DLPFC (Brodmann area 46/10) activity was greater for low- than for high-confidence decisions in both tasks, demonstrating a general role in decision making. Even when reaction times (RTs) were included in the model, confidence remained the significant predictor of activity, suggesting that right DLPFC is involved in discontinuous evaluation rather than in continuous monitoring. In contrast, activity in anterior cingulate cortex was not only greater for low-confidence decisions but also increased with RT, reflecting a role in continuous conflict monitoring. Overall, the results demonstrate how direct cross-function comparisons clarify the generality and specificity of the functions of various brain regions.

Effects of healthy aging on hippocampal and rhinal memory functions: an event-related fMRI study.

Event-related functional magnetic resonance imaging was used to study the effects of healthy aging on hippocampal and rhinal memory functions. Memory for past events can be based on retrieval accompanied by specific contextual details (recollection) or on the feeling that an event is old or new without the recovery of contextual details (familiarity). There is evidence that recollection is more dependent on hippocampus, whereas familiarity is more dependent on the rhinal cortex, and that healthy aging has greater effects on recollection than on familiarity. However, little evidence is available about the neural correlates of these effects. Here, we isolated activity associated with recollection and familiarity by distinguishing between linear and quasi-exponential "perceived oldness" functions derived from recognition confidence levels. The main finding was a double dissociation within the medial temporal lobes between recollection-related activity in hippocampus, which was reduced by aging, and familiarity-related activity in rhinal cortex, which was increased by aging. In addition, age dissociations were found within parietal and posterior midline regions. Finally, aging reduced functional connectivity within a hippocampal-retrosplenial/parietotemporal network but increased connectivity within a rhinal-frontal network. These findings indicate that older adults compensate for hippocampal deficits by relying more on rhinal cortex, possibly through a top-down frontal modulation. This finding has important clinical implications because early Alzheimer's disease impairs both hippocampus and rhinal cortex.