SASamir Akre

Samir Akre

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Samir Akre has a total of 97 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Correspondence in emotionality between spoken words and ecological momentary assessment in individuals varying in depression and anhedonia

    by Olivia M. Losiewicz, Tomislav D. Zbozinek, Zachary D. Cohen, Samir Akre-Bhide, Alex A.T. Bui, Michelle G. Craske

  • Open Access

    How US Youth Use AI Chatbots: Conversation Patterns From Naturalistic Keystroke Observations (Preprint)

    by Anne J Maheux, Debra Boeldt, Samir Akre-Bhide, Jessica Flannery, Allison Paige, Giavanna Villella, Kaitlyn Burnell, Eva H Telzer, Scott H Kollins

    BACKGROUND Youth are increasingly engaging with generative AI (GenAI) chatbots; yet, little is known about the nature of their interactions. OBJECTIVE This study used naturalistic behavioral observations of keystrokes typed into GenAI mobile apps to characterize the topics youth discuss with chatbots, describing themes by age and app, word count by theme (a proxy for quantity of engagement), and co-occurring themes. METHODS Participants were 3363 US youth aged 1-17 years using the Aura parental monitoring app, a commercially available service that guardians install on children’s mobile devices for digital safety. Participants included only those who opted into the overlay Aura keyboard (to collect keystroke data) and who engaged in some degree of GenAI use. Participants’ keystrokes for distinct GenAI apps by day (ie, user-app-days) were coded using a large language model with human review. RESULTS Usage patterns were highly variable, with most users accessing only 1 to 2 GenAI apps (range 1-15) across a median of 4 days (IQR 1-11, range 1-247; 0.4%-90% of days during the study period), sending a median of 6 (IQR 2-27, range 2-2865) messages per app-day. Results identified 11 mutually inclusive themes, including functional tool use; social practice; role-play, including general, romantic, and sexual; emotional support; friend-like interactions; and violence. Most user-app-days involved using AI as a tool (26,343/42,355, 62%), but notable proportions included violence (6460/42,355, 15%), sexual role-play (6393/42,355, 15%), romantic role-play (3538/42,355, 8%), general role-play (3707/42,355, 9%), and friend-like interaction (2624/42,355, 6%). Theme prevalence varied by age: tool use was more common among older adolescents, whereas violence and sexual or romantic role-play were relatively more common among younger users. Themes also varied by app, with general-purpose apps most often used as tools and companionship or role-play apps showing higher usage of sexual, romantic, general role-play, and violent content; however, ChatGPT appeared across all themes due to high overall use. Word counts were highest for user-app-days involving violence, followed by sexual role-play, romantic role-play, and emotional support, suggesting greater engagement when using GenAI in these ways. Co-occurrence analyses showed that tool use often appeared alone, whereas violence, sexual role-play, and romantic role-play frequently co-occurred within a user-app-day, including instances suggestive of violent sexual role-play. CONCLUSIONS Findings suggest that youth engagement with GenAI varies substantially across users. Tool use is most common, requiring more nuanced investigation into the precise ways youth use GenAI to potentially support or thwart their development. GenAI engagement also may involve developmentally salient and potentially risky themes of sexuality, romance, and violence, especially among younger users. More research is needed that pairs GenAI engagement with psychological measures of well-being to examine the impacts of AI interactions and promote safety-focused design features.

  • Open Access

    How US Youth Use AI Chatbots: Conversation Patterns from Naturalistic Keystroke Observations (Preprint)

    by Anne Maheux, Debra Boeldt, Samir Akre-Bhide, Jessica Flannery, Allison Paige, Giavanna Villella, Kaitlyn Burnell, Eva H Telzer, Scott Kollins

    Background: Youth are increasingly engaging with generative AI (GenAI) chatbots; yet, little is known about the nature of their interactions. Objective: This study used naturalistic behavioral observations of keystrokes typed into GenAI mobile apps to characterize the topics youth discuss with chatbots, describing themes by age and app, word count by theme (a proxy for quantity of engagement), and co-occurring themes. Methods: Participants were 3363 US youth aged 1-17 years using the Aura parental monitoring app, a commercially available service that guardians install on children’s mobile devices for digital safety. Participants included only those who opted into the overlay Aura keyboard (to collect keystroke data) and who engaged in some degree of GenAI use. Participants’ keystrokes for distinct GenAI apps by day (ie, user-app-days) were coded using a large language model with human review. Results: Usage patterns were highly variable, with most users accessing only 1 to 2 GenAI apps (range 1-15) across a median of 4 days (IQR 1-11, range 1-247; 0.4%-90% of days during the study period), sending a median of 6 (IQR 2-27, range 2-2865) messages per app-day. Results identified 11 mutually inclusive themes, including functional tool use; social practice; role-play, including general, romantic, and sexual; emotional support; friend-like interactions; and violence. Most user-app-days involved using AI as a tool (26,343/42,355, 62%), but notable proportions included violence (6460/42,355, 15%), sexual role-play (6393/42,355, 15%), romantic role-play (3538/42,355, 8%), general role-play (3707/42,355, 9%), and friend-like interaction (2624/42,355, 6%). Theme prevalence varied by age: tool use was more common among older adolescents, whereas violence and sexual or romantic role-play were relatively more common among younger users. Themes also varied by app, with general-purpose apps most often used as tools and companionship or role-play apps showing higher usage of sexual, romantic, general role-play, and violent content; however, ChatGPT appeared across all themes due to high overall use. Word counts were highest for user-app-days involving violence, followed by sexual role-play, romantic role-play, and emotional support, suggesting greater engagement when using GenAI in these ways. Co-occurrence analyses showed that tool use often appeared alone, whereas violence, sexual role-play, and romantic role-play frequently co-occurred within a user-app-day, including instances suggestive of violent sexual role-play. Conclusions: Findings suggest that youth engagement with GenAI varies substantially across users. Tool use is most common, requiring more nuanced investigation into the precise ways youth use GenAI to potentially support or thwart their development. GenAI engagement also may involve developmentally salient and potentially risky themes of sexuality, romance, and violence, especially among younger users. More research is needed that pairs GenAI engagement with psychological measures of well-being to examine the impacts of AI interactions and promote safety-focused design features.


  • Open Access

    How US Youth Use AI Chatbots: Conversation Patterns from Naturalistic Keystroke Data

    by Anne J. Maheux, Debra Boeldt, Samir Akre-Bhide, Jessica E. Flannery, Allison Paige, Giavanna Villella, Kaitlyn Burnell, Eva H. Telzer, Scott Kollins

    Youth are increasingly engaging with generative artificial intelligence (genAI) chatbots, yet little is known about the nature of their interactions. This study used naturalistic behavioral observations of keystrokes typed into genAI mobile apps to characterize the topics youth discuss with chatbots. Participants were 3,363 U.S. youth ages ≤9–17 years using the Aura parental monitoring app who engaged in some degree of genAI use. Text input was coded using a large language model with human review. Most youth used genAI as a functional tool for problem-solving and information seeking. However, a substantial minority engaged in conversations involving violence, romantic and sexual role-play, and friendship. Younger users showed higher rates of potentially inappropriate content than older teens. Although general-purpose applications (e.g., ChatGPT) were mostly used for task assistance, they were also popular for socioemotional purposes. More research is needed examining psychosocial impacts of AI interactions and safety-focused design features.

  • Open Access

    Detecting momentary reward and affect with real-time passive digital sensor data

    by Samir Akre-Bhide, Zachary D Cohen, Amelia Welborn, Tomislav D Zbozinek, Michelle G Craske, Alex Bui

    Objectives: This study explores the capability of passive digital sensor data from smartphones and smartwatches to predict self-reported ecological momentary assessments (EMA) of affect, motivation, interest, and pleasure in activities in an unseen test sample. Materials and Methods: Data were collected from 245 depressed participants with high-to-low anhedonia (195 train, 50 test) generating 23 812 EMA sessions. Machine learning models were used to assess the ability of behavioral and physiological features, aggregated over windows of 15 minutes to 3 hours, to predict momentary subjective states. Results: For 12 of 15 EMA questions asked, machine learning models exceeded random chance in the fully-held-out test sample, suggesting detectable signals between passive measures and subjective states. Dependent on the sensor type, the optimal aggregation periods ranged from 15 minutes to 3 hours, with generally at least two hours of data being required. Subgroup analyses revealed variations in model performance by demographics, depression severity, and anhedonia severity. Conclusion: This study establishes the feasibility of using passive digital sensing to detect momentary subjective states, providing a baseline for scalable, non-invasive mental health monitoring


  • Open Access

    Technology Effects and Child Health: Wellness Impact and Social Effects (TECHWISE): Protocol for a Prospective, Observational, Real-World Study

    by Scott H Kollins, Jessica Flannery, Karen Goetz, Samir Akre-Bhide

    Background: There has been controversy over the extent to which technology use in general—and social media exposure specifically—may be associated with adverse youth mental health outcomes. To date, studies have generally been small and exploratory, relying on youth self-reports to characterize technology and social media use patterns. The goal of this study is to explore the associations between objectively collected data, gathered through a commercially available app, and youth mental health outcomes. Objective: The broad goal of this study is to characterize the association between objectively measured technology/social media use and a range of mental health–related outcomes. Three specific aims will be initially addressed. First, we aim to quantify the association between objectively measured smartphone use and measures of well-being. Second, we aim to differentiate types of engagement—specifically, the types of content consumed versus the overall time spent on the device—and examine their association with outcomes. Third, we aim to identify moderating factors, such as age, gender, and socioeconomic status, that might influence these relationships. A secondary broad objective of this research is to establish a freely available data resource that can be accessed by qualified investigators to address a much wider range of questions in the future. Methods: Up to 1000 male, female, and nonbinary youth aged 8-17 years (inclusive), along with their primary caregivers, will be enrolled in the study. Youth participants must have their own dedicated iOS- or Android-based smartphone or tablet, and both they and their parents must be willing and able to download and install the data collection app on their devices. The study is open to all US-based participants who meet these 2 criteria. Following electronic consent (eConsent), participants and their caregivers will complete a range of baseline measures electronically, including assessments of psychiatric and social functioning, as well as measures of loneliness, digital stress, and disordered eating. Caregivers will be asked to provide information on the participant’s health and mental health history. Youth and caregivers will then complete a similar battery of assessments 1, 2, and 3 months after baseline. Youth participants will also respond to daily questions about their mood, stress, physical activity, and sleep. Both youth and their caregivers will be compensated for completing measures at each time point. The data collection app gathers a wide range of daily data from the participant’s device, including temporal patterns of use, the number and frequency of various app usage, social interactions within apps, and keystroke data. A variety of analytic methods will be used to address key questions related to how technology and social media use are associated with mental health and wellness outcomes. Results: Enrollment for this study began on November 13, 2024. As of May 20, 2025, a total of 106 participants and their caregivers had consented to participate and provided baseline data. An additional 203 children and parents have consented and are currently undergoing eligibility verification and enrollment. Initial data analysis is anticipated to begin in late autumn 2025 or winter 2026, with the expected publication of initial findings in spring 2026. Conclusions: This study will be among the largest to date to collect both objective device usage data and validated, clinically relevant outcome measures. In accordance with our data-sharing policies, any qualified investigator will be able to access the study data, provided appropriate steps are followed. Trial Registration: ClinicalTrials.gov NCT06664944; https://clinicaltrials.gov/ct2/show/NCT06664944 International Registered Report Identifier (IRRID): DERR1-10.2196/69358


  • Open Access

    Comparing self reported and physiological sleep quality from consumer devices to depression and neurocognitive performance

    by Samir Akre, Zachary D. Cohen, Amelia Welborn, Tomislav D. Zbozinek, Brunilda Balliu, Michelle G. Craske, Alex A. T. Bui

    This study examines the relationship between self-reported and physiologically measured sleep quality and their impact on neurocognitive performance in individuals with depression. Using data from 249 participants with medium to severe depression monitored over 13 weeks, sleep quality was assessed via retrospective self-report and physiological measures from consumer smartphones and smartwatches. Correlations between self-reported and physiological sleep measures were generally weak. Machine learning models revealed that self-reported sleep quality could detect all depression symptoms measured using the Patient Health Questionnaire-14, whereas physiological sleep measures detected “sleeping too much” and low libido. Notably, only self-reported sleep disturbances correlated significantly with neurocognitive performance, specifically with processing speed. Physiological sleep was able to detect changes in self-reported sleep, medication use, and sleep latency. These findings emphasize that self-reported and physiological sleep quality are not measuring the same construct, and both are important to monitor when studying sleep quality in relation to depression.


  • Open Access

    Technology Effects and Child Health: Wellness Impact and Social Effects (TECHWISE): Protocol for a Prospective, Observational, Real-World Study (Preprint)

    by Scott H Kollins, Jessica Flannery, Karen Goetz, Samir Akre-Bhide

    BACKGROUND There has been controversy over the extent to which technology use in general—and social media exposure specifically—may be associated with adverse youth mental health outcomes. To date, studies have generally been small and exploratory, relying on youth self-reports to characterize technology and social media use patterns. The goal of this study is to explore the associations between objectively collected data, gathered through a commercially available app, and youth mental health outcomes. OBJECTIVE The broad goal of this study is to characterize the association between objectively measured technology/social media use and a range of mental health–related outcomes. Three specific aims will be initially addressed. First, we aim to quantify the association between objectively measured smartphone use and measures of well-being. Second, we aim to differentiate types of engagement—specifically, the types of content consumed versus the overall time spent on the device—and examine their association with outcomes. Third, we aim to identify moderating factors, such as age, gender, and socioeconomic status, that might influence these relationships. A secondary broad objective of this research is to establish a freely available data resource that can be accessed by qualified investigators to address a much wider range of questions in the future. METHODS Up to 1000 male, female, and nonbinary youth aged 8-17 years (inclusive), along with their primary caregivers, will be enrolled in the study. Youth participants must have their own dedicated iOS- or Android-based smartphone or tablet, and both they and their parents must be willing and able to download and install the data collection app on their devices. The study is open to all US-based participants who meet these 2 criteria. Following electronic consent (eConsent), participants and their caregivers will complete a range of baseline measures electronically, including assessments of psychiatric and social functioning, as well as measures of loneliness, digital stress, and disordered eating. Caregivers will be asked to provide information on the participant’s health and mental health history. Youth and caregivers will then complete a similar battery of assessments 1, 2, and 3 months after baseline. Youth participants will also respond to daily questions about their mood, stress, physical activity, and sleep. Both youth and their caregivers will be compensated for completing measures at each time point. The data collection app gathers a wide range of daily data from the participant’s device, including temporal patterns of use, the number and frequency of various app usage, social interactions within apps, and keystroke data. A variety of analytic methods will be used to address key questions related to how technology and social media use are associated with mental health and wellness outcomes. RESULTS Enrollment for this study began on November 13, 2024. As of May 20, 2025, a total of 106 participants and their caregivers had consented to participate and provided baseline data. An additional 203 children and parents have consented and are currently undergoing eligibility verification and enrollment. Initial data analysis is anticipated to begin in late autumn 2025 or winter 2026, with the expected publication of initial findings in spring 2026. CONCLUSIONS This study will be among the largest to date to collect both objective device usage data and validated, clinically relevant outcome measures. In accordance with our data-sharing policies, any qualified investigator will be able to access the study data, provided appropriate steps are followed. CLINICALTRIAL ClinicalTrials.gov NCT06664944; https://clinicaltrials.gov/ct2/show/NCT06664944 INTERNATIONAL REGISTERED REPORT DERR1-10.2196/69358

  • Open Access

    Comparison of self-reported and physiological sleep quality from consumer devices to depression and neurocognitive performance

    by Samir Akre, Zachary Cohen, Amelia Welborn, Tomislav Zbozinek, Brunilda Balliu, Michelle Craske, Alex Bui

    This study examines the relationship between self-reported and physiologically measured sleep quality in individuals with depression and its impact on neurocognitive performance. Using data from 249 participants with medium to high depression monitored over 13 weeks, sleep quality was assessed via retrospective self-report and physiological measures from consumer smartphones and smartwatches. Correlations between self-reported and physiological sleep measures were generally weak. Machine learning models revealed that self-reported sleep quality could detect all depression symptoms measured on the Patient Health Questionnaire-14, whereas physiological measures only detected “sleeping too much” and low libido. Notably, only self-reported sleep disturbances correlated significantly with neurocognitive performance. Physiological sleep was able to detect changes in the self-reported sleep quality domains of sleep medication use and sleep latency. These findings emphasize that self-reported and physiological sleep quality are not measuring the same construct, and both are important to monitor when studying sleep quality in relation to depression.

  • Open Access

    A Template and Tutorial for Preregistering Studies Using Passive Smartphone Measures

    by Anna M Langener, Björn S. Siepe, Mahmoud Medhat Elsherif, Koen Niemeijer, Pia K. Andresen, Samir Akre-Bhide, Laura Francina Bringmann, Zachary Daniel Cohen, Nathaniel R Choukas, Konstantin Drexl, et al.

    Passive smartphone measures hold significant potential and are increasingly employed in psychological and biomedical research to capture an individual's behavior. These measures involve the near-continuous and unobtrusive collection of data from smartphones without requiring active input from participants. For example, GPS sensors are used to determine the (social) context of a person, and accelerometers to measure movement. However, utilizing passive smartphone measures presents methodological challenges during data collection and analysis. Researchers must make multiple decisions when working with such measures, which can result in different conclusions. Unfortunately, the transparency of these decision-making processes is often lacking. The implementation of open science practices is only beginning to emerge in digital phenotyping studies and varies widely across studies. Well-intentioned researchers may fail to report on some decisions due to the variety of choices that must be made. To address this issue and enhance reproducibility in digital phenotyping studies, we propose the adoption of preregistration as a way forward. Although there have been some attempts to preregister digital phenotyping studies, a template for registering such studies is currently missing. This could be problematic due to the high level of complexity that requires a well-structured template. Therefore, our objective was to develop a preregistration template that is easy to use and understandable for researchers. Additionally, we explain this template and provide resources to assist researchers in making informed decisions regarding data collection, cleaning, and analysis. Overall, we aim to make researchers' choices explicit, enhance transparency, and elevate the standards for studies utilizing passive smartphone measures.