{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T17:55:03Z","timestamp":1790790903859,"version":"4.1.0"},"reference-count":42,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Mhealth Uhealth"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Ubiquitous smartphone access and statistical advances offer opportunities to continuously track affect intensity, which is central to various psychological processes and behaviors. Research demonstrated the potential of personalized predictions of momentary negative affect (NA) and positive affect (PA) using passive sensing. However, studies typically incorporated all available data sources without differentiating their added value, nor did they investigate whether refining location features with self-reported semantic location (eg, workplaces) improved personalized predictions.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n We evaluated three specific aims: (1) how different combinations of data sources improved performance compared to personalized baseline models, (2) whether model predictions differed across passive data aggregation timescales, and (3) whether incorporating self-reported semantic locations improved model predictions.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Adults (final n=133) completed a 14-day ecological momentary assessment (EMA) protocol reporting emotional experiences 5 times daily alongside smartphone sensing. Testing data (n=532 EMAs) used the last 4 surveys for each individual, with the remaining used for training and validation (n=6805 [NA]\/6800 [PA] EMAs). We evaluated whether combinations of personalization, passive sensing, and affect history improved baseline prediction, and how full-information temporal fusion transformers (TFTs) performed across 6 timescales (1, 3, 6, 12, 24, and 48 hours), with or without self-reported semantic location features.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n The baseline model, using each individual\u2019s mean affect in the training set, demonstrated moderate predictive performance for NA (mean absolute error [MAE]=0.66, 95% CI 0.60\u20100.73;\n R<\/jats:italic>\n \u00b2=40.2%) and PA (MAE=0.71, 95% CI 0.65\u20100.78;\n R<\/jats:italic>\n \u00b2=36.1%). Full-information TFTs improved NA prediction (MAE=0.62, 95% CI 0.56\u20100.69;\n R\u00b2<\/jats:italic>\n =45.2%; \u0394MAE=\u22120.04, 95% CI \u22120.06 to \u22120.01;\n P<\/jats:italic>\n values \u2264.004; Cohen\n d<\/jats:italic>\n =\u22120.27) but not PA prediction (MAE=0.70, 95% CI 0.64\u20100.78;\n R<\/jats:italic>\n \u00b2=32.5%; \u0394MAE=\u22120.01, 95% CI -0.03 to 0.02;\n P<\/jats:italic>\n values >.10; Cohen\n d<\/jats:italic>\n =\u22120.04). No pairwise timescale comparison survived false discovery rate (FDR) correction (NA:\n P<\/jats:italic>\n FDR<\/jats:sub>\n =.05-.98; PA:\n P<\/jats:italic>\n FDR<\/jats:sub>\n =.08-.99). Adding self-reported locations did not improve NA prediction (\u0394MAE=0.02, 95% CI \u22120.01 to 0.04;\n P<\/jats:italic>\n values >.20; Cohen\n d<\/jats:italic>\n =0.11) or PA prediction (\u0394MAE=0.01, 95% CI \u22120.01 to 0.03;\n P<\/jats:italic>\n values>.33; Cohen\n d<\/jats:italic>\n =0.07). However, incorporating self-reported semantic locations changed the composition and relative ranking of important inputs, with these changes varying across NA and PA and between past and future inputs.\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Incorporating smartphone features provided a modest and significant improvement in momentary NA prediction, but not PA prediction. Model performance did not vary across passive data aggregation timescales. While adding self-report semantic locations did not improve prediction accuracy, it changed variable-importance patterns and may provide additional context for interpreting digital behavioral markers. Future personalized predictions should incorporate person-mean affect as an essential benchmark. These findings support passive smartphone sensing as a valuable supplement to, rather than a replacement for, active EMA.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/90970","type":"journal-article","created":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T16:58:05Z","timestamp":1790787485000},"page":"e90970-e90970","source":"Crossref","is-referenced-by-count":0,"title":["Incremental Value of Smartphone Sensing for Monitoring Momentary Affect Intensity in Adults Using Transformer-Based Models: Observational Study"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9456-2375","authenticated-orcid":false,"given":"Yiqin","family":"Zhu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7625-810X","authenticated-orcid":false,"given":"Yuyi","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4479-096X","authenticated-orcid":false,"given":"Renee J","family":"Thompson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,9,28]]},"reference":[{"issue":"3","key":"R1","doi-asserted-by":"publisher","first-page":"150","DOI":"10.1080\/17439760600566016","article-title":"The experience of positive emotion is associated with the automatic processing of positive emotional words","volume":"1","author":"Strauss","journal-title":"J Posit Psychol"},{"issue":"2","key":"R2","doi-asserted-by":"publisher","first-page":"332","DOI":"10.1080\/02699931.2018.1451306","article-title":"Negative affect varying in motivational intensity influences scope of memory","volume":"33","author":"Threadgill","journal-title":"Cogn Emot"},{"issue":"1","key":"R3","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1146\/annurev-psych-010213-115043","article-title":"Emotion and decision making","volume":"66","author":"Lerner","journal-title":"Annu Rev Psychol"},{"key":"R4","doi-asserted-by":"crossref","unstructured":"Diagnostic and Statistical Manual of Mental Disorders: DSM-5. American Psychiatric Association; 2013. [doi: 10.1176\/appi.books.9780890425596]","DOI":"10.1176\/appi.books.9780890425596"},{"issue":"1","key":"R5","doi-asserted-by":"publisher","first-page":"285","DOI":"10.1146\/annurev.clinpsy.121208.131305","article-title":"Cognition and depression: current status and future directions","volume":"6","author":"Gotlib","journal-title":"Annu Rev Clin Psychol"},{"issue":"1","key":"R6","doi-asserted-by":"publisher","first-page":"122","DOI":"10.1037\/0033-2909.133.1.122","article-title":"Autobiographical memory specificity and emotional disorder","volume":"133","author":"Williams","journal-title":"Psychol Bull"},{"key":"R7","doi-asserted-by":"publisher","DOI":"10.2196\/65710","article-title":"Readdressing the ongoing challenge of missing data in youth ecological momentary assessment studies: meta-analysis update","volume":"27","author":"Drexl","journal-title":"J Med Internet Res"},{"issue":"1","key":"R8","doi-asserted-by":"publisher","DOI":"10.1093\/jamiaopen\/ooag005","article-title":"Detecting momentary reward and affect with real-time passive digital sensor data","volume":"9","author":"Akre-Bhide","journal-title":"JAMIA Open"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","DOI":"10.2196\/39425","article-title":"Objective prediction of next-day\u2019s affect using multimodal physiological and behavioral data: algorithm development and validation study","volume":"7","author":"Jafarlou","journal-title":"JMIR Form Res"},{"issue":"11","key":"R10","doi-asserted-by":"publisher","first-page":"5146","DOI":"10.1017\/S0033291722002161","article-title":"Predicting states of elevated negative affect in adolescents from smartphone sensors: a novel personalized machine learning approach","volume":"53","author":"Ren","journal-title":"Psychol Med"},{"key":"R11","unstructured":"Mobile fact sheet. Pew Research Center. Nov 20, 2025. URL: https:\/\/www.pewresearch.org\/internet\/fact-sheet\/mobile\/ [Accessed 04-09-2026]"},{"issue":"12","key":"R12","doi-asserted-by":"publisher","DOI":"10.3390\/s20123572","article-title":"Passive sensing of prediction of moment-to-moment depressed mood among undergraduates with clinical levels of depression sample using smartphones","volume":"20","author":"Jacobson","journal-title":"Sensors (Basel)"},{"issue":"1","key":"R13","doi-asserted-by":"publisher","DOI":"10.1038\/s41398-021-01445-0","article-title":"Personalized machine learning of depressed mood using wearables","volume":"11","author":"Shah","journal-title":"Transl Psychiatry"},{"issue":"10","key":"R14","doi-asserted-by":"publisher","first-page":"665","DOI":"10.1038\/s44159-024-00355-4","article-title":"Positive affect and reward processing in the treatment of depression, anxiety and trauma","volume":"3","author":"Craske","journal-title":"Nat Rev Psychol"},{"issue":"5","key":"R15","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1037\/ccp0000396","article-title":"Positive affect treatment for depression and anxiety: a randomized clinical trial for a core feature of anhedonia","volume":"87","author":"Craske","journal-title":"J Consult Clin Psychol"},{"issue":"4","key":"R16","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2026.7403","article-title":"Positive affect treatment for depression, anxiety, and low positive affect: a randomized clinical trial","volume":"9","author":"Meuret","journal-title":"JAMA Netw Open"},{"issue":"2","key":"R17","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1037\/0033-2909.98.2.219","article-title":"Toward a consensual structure of mood","volume":"98","author":"Watson","journal-title":"Psychol Bull"},{"issue":"4","key":"R18","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1037\/0022-3514.38.4.668","article-title":"Influence of extraversion and neuroticism on subjective well-being: happy and unhappy people","volume":"38","author":"Costa","journal-title":"J Pers Soc Psychol"},{"issue":"1-2","key":"R19","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1016\/j.biopsycho.2004.03.008","article-title":"What does the prefrontal cortex \u201cdo\u201d in affect: perspectives on frontal EEG asymmetry research","volume":"67","author":"Davidson","journal-title":"Biol Psychol"},{"issue":"6","key":"R20","doi-asserted-by":"publisher","first-page":"304","DOI":"10.1093\/geronb\/60.6.p304","article-title":"Understanding the relative importance of positive and negative social exchanges: examining specific domains and appraisals","volume":"60","author":"Newsom","journal-title":"J Gerontol B Psychol Sci Soc Sci"},{"issue":"4","key":"R21","doi-asserted-by":"publisher","first-page":"1748","DOI":"10.1016\/j.ijforecast.2021.03.012","article-title":"Temporal Fusion Transformers for interpretable multi-horizon time series forecasting","volume":"37","author":"Lim","journal-title":"Int J Forecast"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/0092-6566(87)90023-7","article-title":"Affect intensity as an individual difference characteristic: a review","volume":"21","author":"Larsen","journal-title":"J Res Pers"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","DOI":"10.1177\/25152459231202677","article-title":"It\u2019s all about timing: exploring different temporal resolutions for analyzing digital-phenotyping data","volume":"7","author":"Langener","journal-title":"Adv Methods Pract Psychol Sci"},{"issue":"3","key":"R24","doi-asserted-by":"publisher","first-page":"355","DOI":"10.1037\/0033-2909.130.3.355","article-title":"Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research","volume":"130","author":"Dickerson","journal-title":"Psychol Bull"},{"issue":"2","key":"R25","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1177\/1754073912468165","article-title":"Appraisal theories of emotion: state of the art and future development","volume":"5","author":"Moors","journal-title":"Emotion Review"},{"issue":"5","key":"R26","doi-asserted-by":"publisher","first-page":"805","DOI":"10.1037\/0022-3514.76.5.805","article-title":"Core affect, prototypical emotional episodes, and other things called emotion: dissecting the elephant","volume":"76","author":"Russell","journal-title":"J Pers Soc Psychol"},{"issue":"4","key":"R27","doi-asserted-by":"publisher","first-page":"677","DOI":"10.1037\/a0037250","article-title":"The Situational Eight DIAMONDS: a taxonomy of major dimensions of situation characteristics","volume":"107","author":"Rauthmann","journal-title":"J Pers Soc Psychol"},{"issue":"8","key":"R28","doi-asserted-by":"publisher","DOI":"10.2196\/38495","article-title":"Predicting multiple sclerosis outcomes during the COVID-19 stay-at-home period: observational study using passively sensed behaviors and digital phenotyping","volume":"9","author":"Chikersal","journal-title":"JMIR Ment Health"},{"issue":"7","key":"R29","doi-asserted-by":"publisher","DOI":"10.2196\/13209","article-title":"Identifying behavioral phenotypes of loneliness and social isolation with passive sensing: statistical analysis, data mining and machine learning of smartphone and Fitbit data","volume":"7","author":"Doryab","journal-title":"JMIR Mhealth Uhealth"},{"issue":"6","key":"R30","doi-asserted-by":"publisher","first-page":"1516","DOI":"10.1037\/emo0001510","article-title":"Interoceptive awareness and clarity of one\u2019s emotions and goals: a naturalistic investigation","volume":"25","author":"Gross","journal-title":"Emotion"},{"issue":"2","key":"R31","doi-asserted-by":"publisher","first-page":"301","DOI":"10.1080\/02699931.2025.2504560","article-title":"When and why people do NOT regulate their emotions: examining the reasons and contexts","volume":"40","author":"Lai","journal-title":"Cogn Emot"},{"issue":"7","key":"R32","doi-asserted-by":"publisher","first-page":"7691","DOI":"10.3758\/s13428-024-02445-w","article-title":"SEMA3: a free smartphone platform for daily life surveys","volume":"56","author":"O\u2019Brien","journal-title":"Behav Res"},{"key":"R33","doi-asserted-by":"publisher","DOI":"10.3389\/fict.2015.00006","article-title":"AWARE: Mobile context instrumentation framework","volume":"2","author":"Ferreira","journal-title":"Front ICT"},{"key":"R34","doi-asserted-by":"crossref","unstructured":"Canzian L Musolesi M . Trajectories of depression: unobtrusive monitoring of depressive states by means of smartphone mobility traces analysis. Presented at: UbiComp \u201915: The 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing Association for Computing Machinery; Sep 7-11, 2015:1293-1304; Osaka, Japan. URL: https:\/\/dl.acm.org\/doi\/proceedings\/10.1145\/2750858 [Accessed 21-09-2026] [doi: 10.1145\/2750858.2805845]","DOI":"10.1145\/2750858.2805845"},{"key":"R35","doi-asserted-by":"crossref","unstructured":"Saeb S Zhang M Kwasny M Karr CJ Kording K Mohr DC . The relationship between clinical, momentary, and sensor-based assessment of depression. Presented at: 9th International Conference on Pervasive Computing Technologies for Healthcare; May 20-23, 2015:229-232; Istanbul, Turkey. URL: http:\/\/eudl.eu\/proceedings\/PervasiveHealth\/2015 [Accessed 04-09-2026] [doi: 10.4108\/icst.pervasivehealth.2015.259034]","DOI":"10.4108\/icst.pervasivehealth.2015.259034"},{"key":"R36","unstructured":"Ringwald WR . Refining behavioral phenotypes for binge drinking from broad liabilities to proximal predictors with multiple raters of personality and smartphone sensor data [Dissertation]. University of Pittsburgh; 2024. URL: https:\/\/d-scholarship.pitt.edu\/45398 [Accessed 08-05-2025]"},{"key":"R37","unstructured":"Ester M Kriegel HP Sander J Xu X . A density-based algorithm for discovering clusters in large spatial databases with noise. Presented at: KDD\u201996: Second International Conference on Knowledge Discovery and Data Mining; Aug 2-4, 1996:226-231; Portland, Oregon, USA. URL: https:\/\/aaai.org\/papers\/kdd96-037-a-density-based-algorithm-for-discovering-clusters-in-large-spatial-databases-with-noise\/ [Accessed 20-09-2026]"},{"key":"R38","unstructured":"Clemens K . Geocoding with OpenStreetMap data. Presented at: GEOProcessing; Feb 22-27, 2015; Lisbon, Portugal. URL: https:\/\/www.thinkmind.org\/download_full.php?instance=GEOProcessing+2015 [Accessed 20-09-2026]"},{"key":"R39","volume":"12","author":"Pedregosa","journal-title":"J Mach Learn Res"},{"issue":"2","key":"R40","doi-asserted-by":"publisher","first-page":"301","DOI":"10.1111\/j.1467-9868.2005.00503.x","article-title":"Regularization and variable selection via the elastic net","volume":"67","author":"Zou","journal-title":"J R Stat Soc Series B Stat Methodol"},{"key":"R41","doi-asserted-by":"crossref","unstructured":"Chen T Guestrin C . XGBoost: a scalable tree boosting system. Presented at: KDD \u201916: The 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining Association for Computing Machinery; Aug 13-17, 2016:785-794; San Francisco, CA, USA. URL: https:\/\/dl.acm.org\/doi\/proceedings\/10.1145\/2939672 [Accessed 20-09-2026] [doi: 10.1145\/2939672.2939785]","DOI":"10.1145\/2939672.2939785"},{"issue":"10","key":"R42","doi-asserted-by":"publisher","first-page":"78","DOI":"10.1145\/2347736.2347755","article-title":"A few useful things to know about machine learning","volume":"55","author":"Domingos","journal-title":"Commun ACM"}],"container-title":["JMIR mHealth and uHealth"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T16:58:09Z","timestamp":1790787489000},"score":1,"resource":{"primary":{"URL":"https:\/\/mhealth.jmir.org\/2026\/1\/e90970"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,28]]},"references-count":42,"URL":"https:\/\/doi.org\/10.2196\/90970","relation":{},"ISSN":["2291-5222"],"issn-type":[{"value":"2291-5222","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,28]]},"article-number":"v14i3e90970"}}