conduct ecological momentary assessment

Existing mobile sensing data collection methods neglect subjective feedback (e.g., questionnaires, self-reports), leading to fragmented contextual understanding and inaccurate behavioral modeling. To address this, we propose a human-in-the-loop collaborative sensing framework built upon the iLog platform. Our approach introduces three key innovations: (1) a dual-dimensional “context–time” modeling paradigm; (2) a calendar-style real-time monitoring dashboard; and (3) a dynamic acquisition plan revision mechanism. Leveraging context-aware modeling, real-time visual analytics, an adaptive experimental workflow engine, and purposeful human–system interaction design, the framework enhances controllability for researchers, participants, and the system itself. Evaluated with 350 university students, our method significantly improves semantic richness, contextual completeness, and overall data quality—enabling more accurate behavioral modeling and fine-grained personalized analysis.

This study addresses the challenge of modeling longitudinal stress dynamics in Ecological Momentary Assessment (EMA) due to irregular self-report timing and sparse data. We propose Ema2Vec—a novel, learnable time-encoding mechanism specifically designed for irregular, self-reported timestamps—the first of its kind. Integrated with sequence modeling frameworks, Ema2Vec jointly leverages heterogeneous multimodal data, including EMA text reports, mobile sensing signals, and wearable device measurements, to enable fine-grained, continuous prediction of individual stress and affective states. Experiments demonstrate statistically significant performance gains (p < 0.01) over fixed-window baselines and time-agnostic models on longitudinal stress prediction. Crucially, Ema2Vec mitigates temporal dependency modeling bias induced by non-uniform sampling and missing observations. By unifying irregular temporal structure with multimodal behavioral signals, this work establishes a new paradigm for digital phenotyping in mental health research.

Existing stressor assessment methods—such as diaries, end-of-day interviews, and conventional ecological momentary assessment (EMA)—rely on sparse sampling and structured responses, leading to biased estimation of the true frequency of daily stress events. To address this, we propose a wearable-triggered, free-text EMA paradigm coupled with an asymptotic modeling framework, enabling the first unbiased, population-level estimation of frequencies across multiple categories of daily stressors. Our model explicitly corrects for selection bias inherent in sparse sampling, thereby establishing a theoretical benchmark for stress exposure. Leveraging large-scale empirical data from a diverse adult sample, we estimate a mean of 5.39 stress events per person per day, with work-related (1.76), health-related (0.59), and transportation-related (0.55) stressors ranking highest—establishing, for the first time, empirically grounded, real-world stress baselines.

Traditional ecological momentary assessment (EMA) analyses often rely on simplified metrics such as means, which inadequately capture individual longitudinal dynamics. This study moves beyond mean-based approaches by integrating sequence analysis, principal component analysis, and K-means clustering to directly identify latent subgroups exhibiting similar behavioral patterns from full time-series data, while effectively accommodating heterogeneity in sample size and observation frequency. Validated against latent class analysis (LCA) and latent transition analysis (LTA), the proposed method successfully uncovers distinct stress trajectory subgroups in real-world EMA stress data and substantially enhances explanatory power for variability in cognitive performance.

This study addresses the opacity of mandatory wearable sensing systems in high-stakes institutional settings, where users struggle to understand how their behavioral data are translated into consequential judgments. Drawing on in-depth interviews and behavioral observations of 24 individuals under electronic monitoring in China’s community correction system, the research introduces the concept of “sensor literacy” to elucidate how users actively construct risk awareness, probe system boundaries, and adapt their conduct under opaque surveillance. Two adaptive patterns emerge: limited behavioral flexibility when rules are predictable, and excessive activity contraction under high uncertainty. Notably, computational habits persist even after device removal. These findings offer critical insights for enhancing transparency and human-centered design in institutional sensing systems.

This study addresses the gap in understanding real-world ear-worn device usage by integrating survey responses from 330 adults, multi-year Apple Health audio exposure logs from 90 participants, and psychosocial assessments. It presents the first empirical linkage between long-term passive usage data and psychological factors, revealing that ear-worn device use is highly intermittent: daily usage increased from an average of 37 minutes in 2020 to 64 minutes in 2024, yet nearly half of all days recorded no use, and high-volume listening remained rare (only 4% of weeks exceeded WHO-recommended limits). These findings establish a crucial empirical foundation for ear-worn computing and offer design implications centered on intermittent wear patterns, context-aware adaptation, and personalized auditory health interventions.