Digital Phenotypes of Mobile Keyboard Backspace Rates and Their Associations With Symptoms of Mood Disorder: Algorithm Development and Validation (Preprint) DOI
Qimin Liu, Emma Ning, Mindy K. Ross

et al.

Published: July 26, 2023

BACKGROUND Passive sensing through smartphone keyboard data can be used to identify and monitor symptoms of mood disorders with low participant burden. Behavioral phenotyping based on mobile keystroke aid in clinical decision-making provide insights into the individual disorders. OBJECTIVE This study aims derive digital phenotypes backspace use among 128 community adults across 2948 observations using a Bayesian mixture model. METHODS Eligible participants completed virtual screening visit where all eligible were instructed download custom-built BiAffect (University Illinois). The unobtrusively captures dynamics. All consenting this exclusively for up 4 weeks real life, participants’ compliance was checked at 2 follow-up visits week 4. As part research protocol, every underwent evaluations by psychiatrist during each visit. RESULTS We found that derived associated not only diagnoses severity depression mania but also specific symptoms. Using linear mixed-effects model random intercepts accounting nested structure from daily data, rates continuous scale did differ between healthy control groups (<i>P</i>=.11). 3-class had mean 0.112, 0.180, 0.268, respectively, SD 0.048. In total, 3 classes, estimated comprise 37.5% (n=47), 54.4% (n=72), 8.1% (n=9) sample. grouped individuals Low, Medium, High rate groups. Individuals unipolar disorder predominantly Medium group (n=54), some Low (n=27) few (n=6). group, compared significantly higher ratings (<i>b</i>=2.32, <i>P</i>=.008). (<i>P</i>=.88) or without (<i>P</i>=.27) adjustment medication diagnoses. both nonzero (<i>b</i>=1.91, <i>P</i>=.02) (<i>b</i>=1.46, <i>P</i>&lt;.001). showed odds elevated (<i>P</i>=.03), motor activity (<i>P</i>=.04), irritability (<i>P</i>&lt;.05). CONCLUSIONS demonstrates promise typing kinematics practice. Monitoring single kinematic feature, is, rates, passive imposes burden participants. Based real-life our feature useful researchers practitioners distinguish those CLINICALTRIAL

Language: Английский

Childhood neglect is associated with alterations in neural prediction error signaling and the response to novelty DOI Creative Commons
Joseph Aloi,

Kathleen I. Crum,

Karina S. Blair

et al.

Psychological Medicine, Journal Year: 2024, Volume and Issue: 54(14), P. 3930 - 3938

Published: Oct. 1, 2024

Abstract Background One in eight children experience early life stress (ELS), which increases risk for psychopathology. ELS, particularly neglect, has been associated with reduced responsivity to reward. However, little work investigated the computational specifics of this disrupted reward response – respect neural Reward Prediction Errors (RPE) a critical signal successful instrumental learning and extent they are augmented novel stimuli. The goal current study was investigate associations abuse representation RPE non-novel Methods hundred seventy-eight participants (aged 10–18, M = 14.9, s.d. 2.38) engaged Novelty task while undergoing functional magnetic resonance imaging. In task, learn choose or stimuli win monetary rewards varying from $0 $0.30 per trial. Levels neglect were measured using Childhood Trauma Questionnaire. Results Adolescents exposed high levels showed RPE-modulated blood oxygenation level dependent within medial lateral frontal cortices when exploring ( p < 0.05, corrected multiple comparisons) relative adolescents lower neglect. Conclusions These data expand on previous by indicating that but not abuse, is impairments cortices. there no association between behavioral suggesting these differences do necessarily translate into context task.

Language: Английский

Citations

1
Digital Phenotypes of Mobile Keyboard Backspace Rates and Their Associations With Symptoms of Mood Disorder: Algorithm Development and Validation DOI Creative Commons
Qimin Liu, Emma Ning, Mindy K. Ross

et al.

Journal of Medical Internet Research, Journal Year: 2024, Volume and Issue: 26, P. e51269 - e51269

Published: Oct. 29, 2024

Background Passive sensing through smartphone keyboard data can be used to identify and monitor symptoms of mood disorders with low participant burden. Behavioral phenotyping based on mobile keystroke aid in clinical decision-making provide insights into the individual disorders. Objective This study aims derive digital phenotypes backspace use among 128 community adults across 2948 observations using a Bayesian mixture model. Methods Eligible participants completed virtual screening visit where all eligible were instructed download custom-built BiAffect (University Illinois). The unobtrusively captures dynamics. All consenting this exclusively for up 4 weeks real life, participants’ compliance was checked at 2 follow-up visits week 4. As part research protocol, every underwent evaluations by psychiatrist during each visit. Results We found that derived associated not only diagnoses severity depression mania but also specific symptoms. Using linear mixed-effects model random intercepts accounting nested structure from daily data, rates continuous scale did differ between healthy control groups (P=.11). 3-class had mean 0.112, 0.180, 0.268, respectively, SD 0.048. In total, 3 classes, estimated comprise 37.5% (n=47), 54.4% (n=72), 8.1% (n=9) sample. grouped individuals Low, Medium, High rate groups. Individuals unipolar disorder predominantly Medium group (n=54), some Low (n=27) few (n=6). group, compared significantly higher ratings (b=2.32, P=.008). (P=.88) or without (P=.27) adjustment medication diagnoses. both nonzero (b=1.91, P=.02) (b=1.46, P<.001). showed odds elevated (P=.03), motor activity (P=.04), irritability (P<.05). Conclusions demonstrates promise typing kinematics practice. Monitoring single kinematic feature, is, rates, passive imposes burden participants. Based real-life our feature useful researchers practitioners distinguish those

Language: Английский

Citations

1
From Classical Methods to Generative Models: Tackling the Unreliability of Neuroscientific Measures in Mental Health Research DOI Open Access
Nathaniel Haines, Holly Sullivan‐Toole, Thomas M. Olino

et al.

Published: Jan. 2, 2023

Advances in computational statistics and corresponding shifts funding initiatives over the past few decades have led to a proliferation of neuroscientific measures being developed context mental health research. Although such undoubtedly deepened our understanding neural mechanisms underlying cognitive, affective, behavioral processes associated with various conditions, clinical utility remains underwhelming. Recent commentaries point toward poor reliability partially explain this lack translation. Here, we: (1) provide concise theoretical overview how unreliability impedes translation measures; (2) discuss modeling principles—including those from hierarchical structural equation frameworks—can help improve reliability; then (3) demonstrate combine principles within generative framework achieve more reliable, generalizable brain-behavior relationships for use

Language: Английский

Citations

3
Individual Differences in Computational Psychiatry: A Review of Current Challenges DOI Open Access
Povilas Karvelis, Martin P. Paulus, Andreea O. Diaconescu

et al.

Published: Dec. 11, 2022

Bringing precision to the understanding and treatment of mental disorders requires instruments for studying clinically relevant individual differences. One promising approach is development computational assays: integrating models with cognitive tasks infer latent patient-specific disease processes in brain computations. While recent years have seen many methodological advancements modelling cross-sectional patient studies, much less attention has been paid basic psychometric properties (reliability construct validity) measures provided by assays. In this review, we assess extent issue examining emerging empirical evidence. To contextualize this, also provide a more general perspective on key developments that are needed translating assays clinical practice. Emerging evidence suggests most show poor-to-moderate reliability often little improvement over simple behavioral measures. Furthermore, used test accounts lack convergent validity, which compromises their interpretability. Taken together, these issues pose risk invalidating previous findings undermining ongoing research efforts using study (and even group) We suggest single-task designs, currently dominate landscape, partly blame problems therefore not suitable solving them. Instead, validity need be studied systematically longitudinal designs batteries tasks. Finally, enable applications, it will necessary establish predictive make efficient burdensome.

Language: Английский

Citations

5
Key challenges in neurocognitive assessment of individuals with antisocial personality disorder and psychopathy DOI Creative Commons
Julia Griem, Nathan J. Kolla, John Tully

et al.

Frontiers in Behavioral Neuroscience, Journal Year: 2022, Volume and Issue: 16

Published: Sept. 2, 2022

OPINION article Front. Behav. Neurosci., 02 September 2022Sec. Pathological Conditions https://doi.org/10.3389/fnbeh.2022.1007121

Language: Английский

Citations

4
Selective Amygdala Hypoactivity to Fear in Boys With Persistent Conduct Problems After Parent Training DOI Creative Commons
Arjun Sethi,

Suzanne O’Brien,

James Blair

et al.

Biological Psychiatry, Journal Year: 2022, Volume and Issue: 94(1), P. 50 - 56

Published: Oct. 8, 2022

BackgroundParenting interventions reduce antisocial behaviour (ASB) in some children with conduct problems (CP), but not others. Understanding the neural basis for this disparity is important because persistent ASB associated lifelong morbidity and places a huge burden on our health criminal justice systems. One of most highly replicated correlates amygdala hypoactivity to another person's fear. We aimed assess whether fear CP remediated following reduction after successful treatment, and/or if it marker ASB.MethodsWe conducted prospective, case-control, study typically developing (TD) boys. Both groups (aged 5-10 years) completed two MRI sessions (18±5.8 weeks apart) assessed at each visit. Participants included boys referral parenting intervention group, TD recruited from same schools geographical regions. Final functional data was available 36 57 were divided into those whose improved (n=27) or persisted (n=30) intervention. Functional assessing reactivity then analysed using longitudinal group (TD/improving CP/persistent CP) x timepoint (pre/post) design.ResultsAmygdala only observed absent intervention.ConclusionsOur findings suggest that resistant change intervention, putative target future treatments. Parenting ASB. design. Amygdala Our

Language: Английский

Citations

4
Multimodal active subspace analysis for computing assessment oriented subspaces from neuroimaging data DOI
Ishaan Batta, Anees Abrol, Vince D. Calhoun

et al.

Journal of Neuroscience Methods, Journal Year: 2024, Volume and Issue: 406, P. 110109 - 110109

Published: March 15, 2024

Language: Английский

Citations

0
Using Machine Learning to Determine a Functional Classifier of Retaliation and Its Association With Aggression DOI Creative Commons
James Blair, Johannah Bashford‐Largo, Ahria Dominguez

et al.

JAACAP Open, Journal Year: 2024, Volume and Issue: 3(1), P. 137 - 146

Published: June 13, 2024

Methods to determine integrity of integrated neural systems engaged in functional processes have proven elusive. This study sought the extent which a machine learning retaliation classifier (retaliation vs unfair offer) developed from sample typically developing (TD) adolescents could be applied an independent clinically concerning youth and classifier-determined for was associated with antisocial behavior proactive reactive aggression. Blood oxygen level-dependent response data were collected 82 TD 120 while they performed task. The support vector algorithm tested on (adolescents externalizing internalizing diagnoses). able distinguish offer phase after training (accuracy = 92.48%, sensitivity 89.47%, specificity 93.18%) that comparably successful distinguishing function test sample. Increasing distance hyperplane decreasing conduct problems current provides preliminary importance endophenotype whose is reported levels We worked ensure sex gender balance recruitment human participants. race, ethnic, and/or other types diversity questionnaires prepared inclusive way. selection non-human subjects. experimental samples through cell lines. genomic datasets. Diverse lines datasets not available. One or more authors this paper self-identifies as member one historically underrepresented racial ethnic groups science. actively promote our author group. inclusion science While citing references scientifically relevant work, we also reference list. list includes contributors location community where research conducted who participated collection, design, analysis, interpretation work.

Language: Английский

Citations

0
Neuro-cognitive systems that, when dysfunctional, increase aggression risk and the potential for translation into clinical tools DOI Creative Commons
James Blair

Aggression and Violent Behavior, Journal Year: 2024, Volume and Issue: unknown, P. 102007 - 102007

Published: Sept. 1, 2024

Language: Английский

Citations

0
Neurobiology of temperament, personality and psychopathology: what’s next? DOI
Ирина Трофимова, Alan D. Pickering

Current Opinion in Behavioral Sciences, Journal Year: 2022, Volume and Issue: 44, P. 101112 - 101112

Published: March 3, 2022

Language: Английский

Citations

1