Pith. sign in

REVIEW 2 major objections 5 minor 57 references

Machine Learning-based Context-Aware EMAs: An Offline Feasibility Study

T0 review · 2 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read EMA scheduling can be recast as maximizing a weighted combination of predicted response likelihood and emotion-prediction uncertainty, and offline evidence on two datasets indicates this raises receptivity while capturing less-common…

desk verdict Sensible idea and solid label-grounded feasibility evidence in RQ1/RQ2, but RQ3's headline gains are circular—computed from the same models the trigger optimizes—so treat those numbers as artifacts until validated on real responses. read the letter →

arxiv 2506.15834 v1 pith:GKQJEQ5P submitted 2025-06-18 cs.HC

Zachary D King , Maryam Khalid , Han Yu , Kei Shibuya , Khadija Zanna , Marzieh Majd , Ryan L Brown , Yufei Shen
show 4 more authors
Thomas Vaessen George Kypriotakis Christopher P Fagundes Akane Sano
This is my paper · ORCID
classification cs.HC
keywords ecologicalmomentaryassessmentEMAcompliancereceptivitypredictionmodeluncertaintyemotionrecognitionmulti-objectiveoptimizationmobilehealthMonteCarlodropout
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Mobile health studies rely on Ecological Momentary Assessments (EMAs), short in-the-moment surveys, and low response rates starve emotion-recognition models of ground truth. Machine-learned triggers that fire only when a response is likely can nudge sampling toward emotions that happen to co-occur with responsiveness, narrowing the emotional range collected. This paper proposes a multi-objective score $J = \max_{t} w_u U(t)^2 + w_r R(t)^2$ that combines predicted response likelihood $R(t)$ with uncertainty $U(t)$ of an emotion-prediction model, so prompts are sent when people are both likely to answer and experiencing emotions the model rarely sees. In offline evaluation on 91 spousal caregivers of people with Alzheimer's disease and related dementias (73 with usable labels) and 45 healthy participants, the score is significantly higher at actual responses, higher when reported positive affect deviates from a participant's typical range, and a simulated trigger outperforms random scheduling on compliance and within-person emotion variance. A reader should care because this is a modular way to improve the quality of the subjective data that downstream emotion models depend on, without changing survey content or incentives.

What carries the argument

The central object is the multi-objective function $J=\max_{t\in T} w_u U(t)^2 + w_r R(t)^2$, a weighted sum of squared outputs from two learned models: a binary receptivity classifier (neural network for ADRD, random forest for Healthy) and a regression neural network with Monte Carlo dropout whose prediction variance serves as the uncertainty term $U(t)$. The paper's key design choice is using uncertainty rather than the emotion prediction itself: because positive-affect scores cluster near the participant mean, high-uncertainty moments are assumed to coincide with underrepresented emotional states, counteracting the bias a purely receptivity-driven trigger would introduce. The evaluation machinery is semi-personalized cross-validation, which trains on other participants first and then adds the target participant's earlier days, mimicking how a deployed system would adapt; the offline trigger evaluation compares the time point that maximizes $J$ against a randomly chosen time point within each three-hour window.

What would settle it

Run a randomized field experiment comparing the Smart Trigger with a random trigger for one to two weeks per participant, tracking actual EMA response rates and the variance of the positive-affect scores actually reported; the central claim collapses if the measured compliance difference is not significant or if the smart-trigger group does not show greater within-person emotion variance, and it is also weakened if participants' actual responses at smart-selected times are systematically lower than the model predicted.

Watch

Extended reading notes

Core claim

The paper's claim is that EMA delivery can be treated as a maximization over candidate times of a weighted combination of receptivity and emotion-prediction uncertainty. The receptivity model outputs the probability $R(t)$ that the participant answers a prompt at time $t$; the emotion model, a dropout-based neural network regression over positive affect (PA), outputs a mean prediction and variance $U(t)$ from 200 stochastic forward passes. The function $J = \max_{t\in T} w_u U(t)^2 + w_r R(t)^2$ then selects the moment that balances the two goals. In both datasets, mixed-effects models show $J$ is a significant positive predictor of actual response labels (RQ1) and of the absolute participant-specific z-score of PA (RQ2), meaning higher $J$ aligns with reported emotions farther from the participant's mean. The offline simulation (RQ3) predicts average receptivity rates of 0.93 vs 0.82 (ADRD) and 0.94 vs 0.87 (Healthy) for Smart Trigger vs random, with statistically larger within-participant variance in predicted emotion scores, indicating the trigger would collect a broader emotional range while improving compliance.

Load-bearing premise

The load-bearing premise is that the receptivity model's predicted response labels and the emotion model's predicted scores at unobserved time points are faithful stand-ins for what participants would actually do and feel; if those predictions drift from reality, the simulated compliance gains and emotion-coverage gains are artifacts of evaluating the trigger on the same model that chose the trigger's times.

Editorial extensions

If this is right

  • Deployed in an EMA app, the rule would schedule prompts without changing survey length, frequency, or incentives, sidestepping the compliance trade-offs of those alternatives.
  • The weights $w_u$ and $w_r$ let researchers favor compliance in less-adherent populations or emotional coverage in emotion-model training studies, and the paper suggests adaptively tuning them per participant.
  • If the offline response-rate estimates hold prospectively, a study using the Smart Trigger could expect receptivity gains on the order of 0.82 to 0.93 (caregivers) and 0.87 to 0.94 (healthy adults) relative to random timing.
  • Capturing more extreme positive-affect scores would give emotion-recognition models training labels at the edges of the distribution, where they currently have little supervision.
  • Because the function is modular, new constructs could be added as additional weighted terms, extending the same scheduling logic to other target variables or populations.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: the same uncertainty-seeking objective could generalize to other skewed constructs such as stress or negative affect, or to intervention delivery, provided the prediction model is accurate enough; the paper reports that negative affect was too skewed to model well, so this extension is conditional on model quality.
  • Editorial inference: the RQ3 comparison is an in-silico upper bound because the Smart Trigger selects times using the receptivity model and is then evaluated on those same predicted labels; a prospective randomized trial is needed to confirm the rate gap.
  • Editorial inference: a third term measuring the gap between predicted and participant-average emotion could be added to the function to counter social-desirability bias, since the paper identifies that bias as a reason negative affect lacks high-intensity reports.
  • Editorial inference: because the function is evaluated only through model outputs, it can be re-derived for just-in-time adaptive interventions where the trigger must decide within minutes rather than across a scheduled window.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. This paper proposes a multi-objective function J = max_{t in T} w_u U(t)^2 + w_r R(t)^2 for deciding when to deliver ecological momentary assessments (EMAs), where R(t) is a receptivity model's predicted response probability and U(t) is an emotion recognition model's predictive uncertainty. The intended effect is to prompt when participants are likely to respond and when their emotional state is underrepresented. The authors evaluate the idea offline on two datasets (73 ADRD caregivers and 45 healthy participants) through three research questions: RQ1 tests whether J is higher at times with actual EMA responses using mixed-effects models; RQ2 tests whether J is associated with participant-specific absolute z-scores of positive affect; RQ3 simulates a 'Smart Trigger' versus a random trigger and compares predicted response rates and predicted PA distributions. The paper concludes that the multi-objective function would improve receptivity and capture a broader emotional range.

Significance. The paper addresses a real problem: ML-based EMA scheduling that maximizes receptivity alone can bias collected emotions, because receptivity is correlated with affect. The proposed balance of receptivity and uncertainty is sensible, and the authors provide a genuinely useful offline evaluation framework for RQ1 and RQ2 using actual EMA response labels and PA labels. Modeling comparisons across algorithms and two populations add credibility. If the RQ3 simulation were replaced by an evaluation using actual labels, the contribution would be relevant to mHealth and affective computing. However, the current RQ3 results cannot be taken as evidence of improved compliance or broader emotion capture.

major comments (2)
  1. [§3.3.6, §4.2.3] In RQ3, the Smart Trigger selects the time point that maximizes J = w_u U(t)^2 + w_r R(t)^2 (Eq. 1), and the response rate is then computed from the same receptivity model's predicted response values at that time point, not from actual EMA responses. Because J is increasing in R(t), the selected time point will have a higher predicted response probability than a random time point by construction, so the reported gains (0.93 vs 0.82 for ADRD; 0.94 vs 0.87 for Healthy) are largely a tautological artifact. Similarly, the PA-distribution and within-participant variance comparisons rely on emotion predictions at the selected times, so they do not establish that real emotion reports would be more diverse. The offline limitation is acknowledged in §5.3.1, but this is not merely speculation: the comparison is structurally guaranteed to favor the Smart Trigger. Please validate the simulation against actual EMA labels (e.g., limiting to prompt occasions where a real response outcome exists) or reframe RQ3 explicitly as 'expected behavior under the fitted models' with appropriate caveats and a sensitivity analysis.
  2. [§3.1.4, Eq. (1)] The weights w_u and w_r are never reported, even though every result in RQ1–RQ3 depends on J constructed with specific values. Without these values, the analyses cannot be reproduced and the results cannot be checked for sensitivity to the weighting. Please report the weights used in all experiments, and ideally a sensitivity analysis over a range of weights (the paper already suggests adaptive weights in §5.2).
minor comments (5)
  1. [§3.3.6] Please state the threshold used to convert the receptivity model's predicted probability into binary response/non-response for the simulated response rate.
  2. [§3.1.3, §3.1.4] Define U(t) explicitly; the text says the variance across 200 stochastic forward passes is used, but U(t) is introduced only in Eq. 1 without a formal definition.
  3. [§4.2.1] The repeated-measures ANOVA is non-significant for the Healthy study; please report the effect size rather than attributing the null result to sample size.
  4. [Figure 5] The caption says 'PS' where 'PA' is meant.
  5. [§7 and reproducibility] No data or code availability statement is included; given the reproducibility concerns about the weights, an explicit statement is needed.

Circularity Check

2 steps flagged · score 6.0 of 10

RQ3's simulated Smart Trigger gains are by construction: J maximizes the same R(t) that is then used to compute the simulated response rates.

  1. fitted input called prediction [Section 3.3.6 (RQ3), Eq. (1); results in Section 4.2.3]
    "In contrast, the Smart Trigger selected the time point that maximized the value of 𝐽 (Eq. 1). We utilize our receptivity and emotion recognition models to obtain predicted values for responsiveness (response or non-response) and emotion (PA score) at each selected time point. We then calculate each participant’s response rate using both triggers."

    RQ3's Smart Trigger maximizes J = w_u U(t)^2 + w_r R(t)^2, where R(t) is the receptivity model's predicted probability of response. The simulated 'response rate' is then computed from that same model's predicted response/non-response at the selected time points. Choosing the time with maximal R(t) guarantees that the selected points' mean predicted response probability is no lower than the mean over randomly selected points; the reported gains (ADRD 0.93 vs 0.82; Healthy 0.94 vs 0.87) are therefore largely a mathematical consequence of optimizing the evaluation metric itself, not evidence about actual EMA response behavior. The manuscript acknowledges the offline setting is 'speculative' (§5.3.1), but the structural issue is stronger: the comparison is forced by construction.

  2. fitted input called prediction [Section 3.3.6 (RQ3) and Section 4.2.3 (emotion-distribution comparison)]
    "Additionally, we analyze the distribution of predicted emotional states captured by each strategy. For both the Smart Trigger and the random trigger, we predict emotion scores at each selected time point to compare the distributions of participants’ predicted emotions captured by each method."

    The emotion half of the same simulation compares 'predicted emotional states' produced by the emotion-recognition model whose uncertainty U(t) is explicitly maximized by the Smart Trigger. The within-participant predicted-PA variance differences (ADRD 3.04 vs 2.83; Healthy 2.76 vs 2.15) are computed from the model's own outputs, not from EMA responses. Because the trigger deliberately seeks high-uncertainty points, and the model's uncertainty is a property of the same fitted model used to generate the predicted PA scores, the comparison is partly self-referential: the Smart Trigger is rewarded for selecting points that the model itself regards as unusual. This does not establish that the trigger would capture a broader range of actual emotions.

full rationale

RQ1 and RQ2 are grounded in actual EMA response labels and reported PA scores, so their positive findings are not circular: J's association with actual receptivity and with participant-specific absolute z-scores of PA is an empirical correlation, not a definitional consequence. The circularity is concentrated in the RQ3 simulation. The Smart Trigger is defined (Eq. 1) as the time point maximizing w_u U(t)^2 + w_r R(t)^2; the simulated receptivity rate is then computed from 'predicted values for responsiveness' produced by the same receptivity model. Selecting the maximum of R(t) and then averaging R(t)-derived response predictions over the selected points guarantees an advantage over random selection, so the reported 0.93 vs 0.82 and 0.94 vs 0.87 differences are largely by construction. The emotion-range comparison is less strictly forced but also uses the same emotion model's predicted PA values, rather than actual EMA reports, as the outcome, so the reported variance increases are not independent evidence. The manuscript's own caveat that the findings 'were validated offline and are therefore speculative' (§5.3.1) understates this structural issue. Weights w_u and w_r are never reported, making Eq. 1 non-reproducible (a reporting gap, not a circularity). The citation to the authors' prior King et al. [22] is motivational rather than load-bearing for the current derivation, since RQ1 and RQ2 test the bias claim on actual labels. Because the central feasibility claim retains independent support from RQ1 and RQ2, the overall circularity score is 6 rather than higher.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The objective function's behavior rests on several unverified assumptions about model uncertainty, the validity of simulated responses, and standard statistical procedures. No new entities are introduced.

free parameters (4)
  • weights w_u and w_r = not reported in paper
    Eq. 1 defines J as a weighted combination, and the paper states weights allow prioritizing receptivity or uncertainty, but the evaluation never reports the values used. Results depend on this choice.
  • segment window length = 30 minutes
    The 30-minute window was selected because, after comparing 10, 15, 30 and 60 minutes (Tables S5 and S7), it gave the best performance on both datasets; this is post-hoc selection on the evaluation data.
  • dropout rate = 0.3
    The emotion model's uncertainty estimates use MC dropout with rate 0.3, a hand-chosen hyperparameter that controls the scale of U(t) and therefore J.
  • number of stochastic forward passes = 200
    The emotion model's uncertainty is estimated from 200 Monte Carlo forward passes, a hand-chosen number that affects the stability of U(t).
assumptions (4)
  • domain assumption MC dropout variance U(t) is higher for less frequently observed emotional states, so maximizing U(t) improves emotional diversity.
    Invoked throughout RQ2 and Section 3.1.4; the paper tests J rather than U alone against PA z-scores.
  • ad hoc to paper Predicted responses from the receptivity model at arbitrary time points are a valid proxy for actual EMA responses in the simulation.
    Load-bearing for RQ3's compliance results; not validated against held-out actual responses.
  • domain assumption Emotional state systematically affects EMA receptivity, so receptivity-only triggers bias the sampled emotions.
    Borrowed from cited prior work, especially King et al. (2024); motivates the multi-objective design.
  • standard math Standard statistical assumptions for mixed-effects models, ANOVA, and paired t-tests hold in these datasets.
    Used throughout Section 4; no diagnostics are reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Machine Learning-based Context-Aware EMAs: An Offline Feasibility Study." pith.science (2026). https://pith.science/paper/GKQJEQ5P

@misc{pith2026250615834,
  author       = {Pith},
  title        = {Pith review of: Machine Learning-based Context-Aware EMAs: An Offline Feasibility Study},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GKQJEQ5P}},
  note         = {Machine review of arXiv:2506.15834}
}
read the original abstract

Mobile health (mHealth) systems help researchers monitor and care for patients in real-world settings. Studies utilizing mHealth applications use Ecological Momentary Assessment (EMAs), passive sensing, and contextual features to develop emotion recognition models, which rely on EMA responses as ground truth. Due to this, it is crucial to consider EMA compliance when conducting a successful mHealth study. Utilizing machine learning is one approach that can solve this problem by sending EMAs based on the predicted likelihood of a response. However, literature suggests that this approach may lead to prompting participants more frequently during emotions associated with responsiveness, thereby narrowing the range of emotions collected. We propose a multi-objective function that utilizes machine learning to identify optimal times for sending EMAs. The function identifies optimal moments by combining predicted response likelihood with model uncertainty in emotion predictions. Uncertainty would lead the function to prioritize time points when the model is less confident, which often corresponds to underrepresented emotions. We demonstrate that this objective function would result in EMAs being sent when participants are responsive and experiencing less commonly observed emotions. The evaluation is conducted offline using two datasets: (1) 91 spousal caregivers of individuals with Alzheimer's Disease and Related dementias (ADRD), (2) 45 healthy participants. Results show that the multi-objective function tends to be higher when participants respond to EMAs and report less commonly observed emotions. This suggests that using the proposed objective function to guide EMA delivery could improve receptivity rates and capture a broader range of emotions.

Figures

Figures reproduced from arXiv: 2506.15834 by the authors.

Figure 1
Figure 1. Framework for an EMA Trigger that would utilize the proposed multi-objective function. [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Receptivity Labeling Methodology for Responses (top) and Non-Responses (bottom). In the figure, [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Distribution of Positive Affect (PA) and Negative Affect (NA) among all participants. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Differences in 𝐽 (Equation 1) during responses and non-responses for each population. 4.2.2 Relationship between the output of the multi-objective function 𝐽 and Reported Emotional State (RQ2). Figure 5a illustrates the relationship between 𝐽 and reported PA for the AD…
Figure 5
Figure 5. Figure 5: The dashed line represents the distribution of collected positive affect (PA) labels during the day. Note that the y-axis does not [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 56 canonical work pages

  1. [1]

    MAH Akhand, Shuvendu Roy, Nazmul Siddique, Md Abdus Samad Kamal, and Tetsuya Shimamura. 2021. Facial emotion recognition using transfer learning in the deep CNN.Electronics10, 9 (2021), 1036

    work page 2021

  2. [2]

    Alzheimer’s Association. 2019. 2019 Alzheimer’s disease facts and figures.Alzheimer’s & dementia15, 3 (2019), 321–387

    work page 2019

  3. [3]

    Rummana Bari, Md Mahbubur Rahman, Nazir Saleheen, Megan Battles Parsons, Eugene H Buder, and Santosh Kumar. 2020. Automated detection of stressful conversations using wearable physiological and inertial sensors.Proceedings of the ACM on interactive, mobile, wearable and ubiquitous technologies4, 4 (2020), 1–23

    work page 2020

  4. [4]

    Everything is perfect, and we have no problems

    Nicole Bergen and Ronald Labonté. 2020. “Everything is perfect, and we have no problems”: detecting and limiting social desirability bias in qualitative research.Qualitative health research30, 5 (2020), 783–792

    work page 2020

  5. [5]

    Niranjan Bidargaddi, Daniel Almirall, Susan Murphy, Inbal Nahum-Shani, Michael Kovalcik, Timothy Pituch, Haitham Maaieh, and Victor Strecher

    work page

  6. [6]

    Carlos Carreiras, Ana Priscila Alves, André Lourenço, Filipe Canento, Hugo Silva, Ana Fred, et al. 2015–. BioSPPy: Biosignal Processing in Python. https://github.com/PIA-Group/BioSPPy/ [Online; accessed <today>]

    work page 2015

  7. [7]

    Yu-Chun Chen, Yu-Jen Lee, Kuei-Chun Kao, Jie Tsai, En-Chi Liang, Wei-Chen Chiu, Faye Shih, and Yung-Ju Chang. 2023. Are you killing time? Predicting smartphone users’ time-killing moments via fusion of smartphone sensor data and screenshots. InProceedings of the 2023 CHI Conference on Human Factors in Computing Systems. 1–19

    work page 2023

  8. [8]

    Joana De Calheiros Velozo, Thomas Vaessen, Stephan Claes, and Inez Myin-Germeys. 2024. Investigating adverse daily life effects following a psychosocial laboratory stress task, and the moderating role of Psychopathology.StreSS27, 1 (2024), 2380403. 18 King et al

    work page 2024

Show all 57 references
  1. [9]

    Joana De Calheiros Velozo, Thomas Vaessen, Ginette Lafit, Stephan Claes, and Inez Myin-Germeys. 2023. Is daily-life stress reactivity a measure of stress recovery? An investigation of laboratory and daily-life stress.Stress and Health39, 3 (2023), 638–650

    2023

  2. [10]

    Gudrun Eisele, Hugo Vachon, Ginette Lafit, Peter Kuppens, Marlies Houben, Inez Myin-Germeys, and Wolfgang Viechtbauer. 2020. The effects of sampling frequency and questionnaire length on perceived burden, compliance, and careless responding in experience sampling data in a stu...

    2020

  3. [11]

    Denzil Ferreira, Vassilis Kostakos, and Anind K Dey. 2015. AWARE: mobile context instrumentation framework.Frontiers in ICT2 (2015), 6

    2015

  4. [12]

    Filipe Galvão, Soraia M Alarcão, and Manuel J Fonseca. 2021. Predicting exact valence and arousal values from EEG.Sensors21, 10 (2021), 3414

    2021

  5. [13]

    Cheng He, Yun-jin Yao, and Xue-song Ye. 2017. An emotion recognition system based on physiological signals obtained by wearable sensors. In Wearable sensors and robots. Springer, 15–25

    2017

  6. [14]

    Joyce Ho and Stephen S Intille. 2005. Using context-aware computing to reduce the perceived burden of interruptions from mobile devices. In Proceedings of the SIGCHI conference on Human factors in computing systems. 909–918

    2005

  7. [15]

    Simon Ho. 2018. Sensor Motion. https://github.com/sho-87/sensormotion [Online; accessed <today>]

    2018

  8. [16]

    Karen Hovsepian, Mustafa Al’Absi, Emre Ertin, Thomas Kamarck, Motohiro Nakajima, and Santosh Kumar. 2015. cStress: towards a gold standard for continuous stress assessment in the mobile environment. InProceedings of the 2015 ACM international joint conference on pervasive and ...

    2015

  9. [17]

    Long Hu, Jun Yang, Min Chen, Yongfeng Qian, and Joel JPC Rodrigues. 2018. SCAI-SVSC: Smart clothing for effective interaction with a sustainable vital sign collection.Future Generation Computer Systems86 (2018), 329–338

    2018

  10. [18]

    Lam Huynh, Tri Nguyen, Thu Nguyen, Susanna Pirttikangas, and Pekka Siirtola. 2021. Stressnas: Affect state and stress detection using neural architecture search. InAdjunct Proceedings of the 2021 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proc...

    2021

  11. [19]

    2016.𝜇EMA: Microinteraction-based ecological momentary assessment (EMA) using a smartwatch

    Stephen Intille, Caitlin Haynes, Dharam Maniar, Aditya Ponnada, and Justin Manjourides. 2016.𝜇EMA: Microinteraction-based ecological momentary assessment (EMA) using a smartwatch. InProceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computin...

    2016

  12. [20]

    Leila Jameel, Lucia Valmaggia, Georgina Barnes, and Matteo Cella. 2022. mHealth technology to assess, monitor and treat daily functioning difficulties in people with severe mental illness: A systematic review.Journal of psychiatric research145 (2022), 35–49

    2022

  13. [21]

    Andrew Jones, Danielle Remmerswaal, Ilse Verveer, Eric Robinson, Ingmar HA Franken, Cheng K Fred Wen, and Matt Field. 2019. Compliance with ecological momentary assessment protocols in substance users: a meta-analysis.Addiction114, 4 (2019), 609–619

    2019

  14. [22]

    Zachary D King, Han Yu, Thomas Vaessen, Inez Myin-Germeys, and Akane Sano. 2024. Investigating Receptivity and Affect Using Machine Learning: Ecological Momentary Assessment and Wearable Sensing Study.JMIR mHealth and uHealth12, 1 (2024), e46347

    2024

  15. [23]

    Florian Künzler, Varun Mishra, Jan-Niklas Kramer, David Kotz, Elgar Fleisch, and Tobias Kowatsch. 2019. Exploring the state-of-receptivity for mhealth interventions.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies3, 4 (2019), 1–27

    2019

  16. [24]

    Weicheng Kuo, Christian Häne, Esther Yuh, Pratik Mukherjee, and Jitendra Malik. 2018. Cost-sensitive active learning for intracranial hemorrhage detection. InMedical Image Computing and Computer Assisted Intervention–MICCAI 2018: 21st International Conference, Granada, Spain, ...

    2018

  17. [25]

    Jixin Li, Aditya Ponnada, Wei-Lin Wang, Genevieve Dunton, and Stephen Intille. 2024. Ask Less, Learn More: Adapting Ecological Momentary Assessment Survey Length by Modeling Question-Answer Information Gain.Proceedings of the ACM on interactive, mobile, wearable and ubiquitous...

    2024

  18. [26]

    David D Luxton, Russell A McCann, Nigel E Bush, Matthew C Mishkind, and Greg M Reger. 2011. mHealth for mental health: Integrating smartphone technology in behavioral healthcare.Professional Psychology: Research and Practice42, 6 (2011), 505

    2011

  19. [27]

    Andrew Mackinnon, Anthony F Jorm, Helen Christensen, Ailsa E Korten, Patricia A Jacomb, and Bryan Rodgers. 1999. A short form of the Positive and Negative Affect Schedule: Evaluation of factorial validity and invariance across demographic variables in a community sample.Person...

    1999

  20. [28]

    Marek Malik and A John Camm. 1993. Components of heart rate variability—what they really mean and what we really measure.The American journal of cardiology72, 11 (1993), 821–822

    1993

  21. [29]

    Abhinav Mehrotra, Mirco Musolesi, Robert Hendley, and Veljko Pejovic. 2015. Designing content-driven intelligent notification mechanisms for mobile applications. InProceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing. 813–824

    2015

  22. [30]

    Varun Mishra, Florian Künzler, Jan-Niklas Kramer, Elgar Fleisch, Tobias Kowatsch, and David Kotz. 2021. Detecting Receptivity for mHealth Interventions in the Natural Environment.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies5, 2 (2021), 1–24

    2021

  23. [31]

    Varun Mishra, Byron Lowens, Sarah Lord, Kelly Caine, and David Kotz. 2017. Investigating contextual cues as indicators for EMA delivery. In Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2017 ACM Internat...

    2017

  24. [32]

    Anna Mitenkova, Jean Kossaifi, Yannis Panagakis, and Maja Pantic. 2019. Valence and arousal estimation in-the-wild with tensor methods. In2019 14th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2019). IEEE, 1–7

    2019

  25. [33]

    Leanne G Morrison, Charlie Hargood, Veljko Pejovic, Adam WA Geraghty, Scott Lloyd, Natalie Goodman, Danius T Michaelides, Anna Weston, Mirco Musolesi, Mark J Weal, et al. 2017. The effect of timing and frequency of push notifications on usage of a smartphone-based stress manag...

    2017

  26. [34]

    Aja Louise Murray, Ruth Brown, Xinxin Zhu, Lydia Gabriela Speyer, Yi Yang, Zhouni Xiao, Denis Ribeaud, and Manuel Eisner. 2023. Prompt-level predictors of compliance in an ecological momentary assessment study of young adults’ mental health.Journal of Affective Disorders322 (2...

    2023

  27. [35]

    Inez Myin-Germeys, Jim van Os, Joseph E Schwartz, Arthur A Stone, and Philippe A Delespaul. 2001. Emotional reactivity to daily life stress in psychosis.Archives of general psychiatry58, 12 (2001), 1137–1144

    2001

  28. [36]

    Supriya Nagesh, Alexander Moreno, Stephanie M Carpenter, Jamie Yap, Soujanya Chatterjee, Steven Lloyd Lizotte, Neng Wan, Santosh Kumar, Cho Lam, David W Wetter, et al. 2021. Transformers for prompt-level EMA non-response prediction.arXiv preprint arXiv:2111.01193(2021)

    2021 arXiv

  29. [37]

    Grzegorz J Nalepa, Krzysztof Kutt, Barbara Giżycka, Paweł Jemioło, and Szymon Bobek. 2019. Analysis and use of the emotional context with wearable devices for games and intelligent assistants.Sensors19, 11 (2019), 2509

    2019

  30. [38]

    Charlotte Ottenstein and Linda Werner. 2022. Compliance in ambulatory assessment studies: Investigating study and sample characteristics as predictors.Assessment29, 8 (2022), 1765–1776

    2022

  31. [39]

    Veljko Pejovic and Mirco Musolesi. 2014. InterruptMe: designing intelligent prompting mechanisms for pervasive applications. InProceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing. 897–908

    2014

  32. [40]

    Andrew JK Phillips, William M Clerx, Conor S O’Brien, Akane Sano, Laura K Barger, Rosalind W Picard, Steven W Lockley, Elizabeth B Klerman, and Charles A Czeisler. 2017. Irregular sleep/wake patterns are associated with poorer academic performance and delayed circadian and sle...

    2017

  33. [41]

    Vincent Pichot, Frédéric Roche, Sébastien Celle, Jean-Claude Barthélémy, and Florian Chouchou. 2016. HRVanalysis: a free software for analyzing cardiac autonomic activity.Frontiers in physiology7 (2016), 557

    2016

  34. [42]

    Martin Pielot, Bruno Cardoso, Kleomenis Katevas, Joan Serrà, Aleksandar Matic, and Nuria Oliver. 2017. Beyond interruptibility: Predicting opportune moments to engage mobile phone users.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies1, 3 (20...

    2017

  35. [43]

    Daniel S Quintana, Adam J Guastella, Tim Outhred, Ian B Hickie, and Andrew H Kemp. 2012. Heart rate variability is associated with emotion recognition: Direct evidence for a relationship between the autonomic nervous system and social cognition.International journal of psychop...

    2012

  36. [44]

    Haroon Rashid, Sanjana Mendu, Katharine E Daniel, Miranda L Beltzer, Bethany A Teachman, Mehdi Boukhechba, and Laura E Barnes. 2020. Predicting subjective measures of social anxiety from sparsely collected mobile sensor data.Proceedings of the ACM on Interactive, Mobile, Weara...

    2020

  37. [45]

    Aki Rintala, Martien Wampers, Inez Myin-Germeys, and Wolfgang Viechtbauer. 2020. Momentary predictors of compliance in studies using the experience sampling method.Psychiatry research286 (2020), 112896

    2020

  38. [46]

    Simon P Rowland, J Edward Fitzgerald, Thomas Holme, John Powell, and Alison McGregor. 2020. What is the clinical value of mHealth for patients? NPJ digital medicine3, 1 (2020), 4

    2020

  39. [47]

    Akane Sano and Rosalind W Picard. 2013. Stress recognition using wearable sensors and mobile phones. In2013 Humaine association conference on affective computing and intelligent interaction. IEEE, 671–676

    2013

  40. [48]

    Hillol Sarker, Moushumi Sharmin, Amin Ahsan Ali, Md Mahbubur Rahman, Rummana Bari, Syed Monowar Hossain, and Santosh Kumar. 2014. Assessing the availability of users to engage in just-in-time intervention in the natural environment. InProceedings of the 2014 ACM international ...

    2014

  41. [49]

    Joseph E Schwartz and Arthur A Stone. 1998. Strategies for analyzing ecological momentary assessment data.Health Psychology17, 1 (1998), 6

    1998

  42. [50]

    Sara Taylor, Natasha Jaques, Weixuan Chen, Szymon Fedor, Akane Sano, and Rosalind Picard. 2015. Automatic identification of artifacts in electrodermal activity data. In2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEE...

    2015

  43. [51]

    Akke Vellinga, Colum Devine, Min Yun Ho, Colin Clarke, Patrick Leahy, Jane Bourke, Declan Devane, Sinead Duane, and Patricia Kearney. 2020. What do patients value as incentives for participation in clinical trials? A pilot discrete choice experiment.Research Ethics16, 1-2 (2020), 1–12

    2020

  44. [52]

    Cheng K Fred Wen, Stefan Schneider, Arthur A Stone, and Donna Spruijt-Metz. 2017. Compliance with mobile ecological momentary assessment protocols in children and adolescents: a systematic review and meta-analysis.Journal of medical Internet research19, 4 (2017), e132

    2017

  45. [53]

    Cornelia Wrzus and Andreas B Neubauer. 2023. Ecological momentary assessment: A meta-analysis on designs, samples, and compliance across research fields.Assessment30, 3 (2023), 825–846

    2023

  46. [54]

    Han Yu and Akane Sano. 2023. Semi-Supervised Learning for Wearable-based Momentary Stress Detection in the Wild.Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies7, 2 (2023), 1–23

    2023

  47. [55]

    Bobo Zhao, Zhu Wang, Zhiwen Yu, and Bin Guo. 2018. EmotionSense: Emotion recognition based on wearable wristband. In2018 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computing, Scalable Computing & Communications, Cloud & Big Data Computing, Interne...

    2018

  48. [57]

    The HRV features are only obtained when participants are asleep and are relatively still while sleeping

    and calculated by Fitbit. The HRV features are only obtained when participants are asleep and are relatively still while sleeping. RMSSD (root mean square of successive differences between normal heartbeats) is one such feature, reflecting parasympathetic nervous system activi...

  49. [2018]

    To prompt or not to prompt? A microrandomized trial of time-varying push notifications to increase proximal engagement with a mobile health app.JMIR mHealth and uHealth6, 11 (2018), e10123

    2018

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.