Large language models often produce biased or stereotypical outputs. One way to reduce this possibility is to be more inclusive in our prompts, but doing so may not come naturally to most users. Therefore, we designed a tool that coaches users to write more inclusive prompts—a strategy that leverages design friction to provide a media literacy intervention. Data from a user study (N=344) show that compared to no coaching, inclusive prompt coaching directly increased users’ awareness of algorithmic bias and their perceived prompting efficacy. It also indirectly enhanced their trust in the system and perceived trust calibration through cognitive elaboration. However, inclusive prompt coaching resulted in a less satisfying user experience. These findings have implications for ethical interventions in prompting for better communicating and combating algorithmic bias. We discuss the benefits and limitations of inclusive prompt coaching, as well as ways to balance usability for long-term adoption of generative AI systems.

Suhas Bn, Yash Mahajan, Dominik O. Mattioli, Andrew M. Sherrill, Rosa I. Arriaga, Christopher Wiese, Saeed Abdullah. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

The advancement of AI systems for mental health support is hindered by limited access to therapeutic conversation data, particularly for trauma treatment. We present Thousand Voices of Trauma, a synthetic benchmark dataset of 3,000 therapy conversations based on Prolonged Exposure therapy protocols for Post-traumatic Stress Disorder (PTSD). The dataset comprises 500 unique cases, each explored through six conversational perspectives that mirror the progression of therapy from initial anxiety to peak distress to emotional processing. We incorporated diverse demographic profiles (ages 18-80, M=49.3, 49.4\% male, 44.4\% female, 6.2\% non-binary), 20 trauma types, and 10 trauma-related behaviors using deterministic and probabilistic generation methods. Analysis reveals realistic distributions of trauma types (witnessing violence 10.6\%, bullying 10.2\%) and symptoms (nightmares 23.4\%, substance abuse 20.8\%). Clinical experts validated the dataset's therapeutic fidelity, highlighting its emotional depth while suggesting refinements for greater authenticity. We also developed an emotional trajectory benchmark with standardized metrics for evaluating model responses. This privacy-preserving dataset addresses critical gaps in trauma-focused mental health data, offering a valuable resource for advancing both patient-facing applications and clinician training tools.

The mental health crisis in the United States spotlights the need for more scalable training for mental health workers. While present-day AI systems have sparked hope for addressing this problem, we must not be too quick to incorporate or solely focus on technological advancements. We must ask empirical questions about how to ethically collaborate with and integrate autonomous AI into the clinical workplace. For these Human-Autonomy Teams (HATs), poised to make the leap into the mental health domain, special consideration around the construct of trust is in order. A reflexive look toward the multidisciplinary nature of such HAT projects illuminates the need for a deeper dive into varied stakeholder considerations of ethics and trust. In this paper, we investigate the impact of domain---and the ranges of expertise within domains---on ethics- and trust-related considerations for HATs in mental health. We outline our engagement of 23 participants in two speculative activities: design fiction and factorial survey vignettes. Grounded by a video storyboard prototype, AI- and Psychotherapy-domain experts and novices alike imagined TEAMMAIT, a prospective AI system for psychotherapy training. From our inductive analysis emerged 10 themes surrounding ethics, trust, and collaboration. Three can be seen as substantial barriers to trust and collaboration, where participants imagined they would not work with an AI teammate that didn't meet these ethical standards. Another five of the themes can be seen as interrelated, context-dependent, and variable factors of trust that impact collaboration with an AI teammate. The final two themes represent more explicit engagement with the prospective role of an AI teammate in psychotherapy training practices. We conclude by evaluating our findings through the lens of Mayer et al.'s Integrative Model of Organizational Trust to discuss the risks of HATs and adapt models of ability-, benevolence-, and integrity-based trust. These updates motivate implications for the design and integration of HATs in mental health work.

Abstract Human cognitive performance affects a wide range of aspects of our daily lives. Numerous factors influence our cognitive performance, and cognitive performance in turn impacts our capabilities. Partial sleep deprivation in particular negatively affects vigilance, a key factor in many work tasks. Sleep in general plays a large role in physiological recovery and our capability to perform mental tasks. In this work, we focus on two research questions. First, we investigate how fluctuations in sleep quality influence cognitive vigilance. Second, we study how smartphone typing can be leveraged as a continuous measurement for cognitive vigilance and can thus be an indicator of decline in cognitive capabilities and sleep quality. We report on a 2-month field study in which we collected cognitive performance data using the Psychomotor Vigilance Task (PVT), mobile keyboard typing metrics from participants’ personal smartphones, and sleep quality metrics through a wearable sleep-tracking ring. Our findings highlight that individual sleep metrics such as night-time heart rate, sleep latency, sleep timing, sleep restfulness, and overall sleep quantity significantly influence vigilance. Long sleep latencies can reduce reaction times up to 30 ms, abnormal sleep durations up to 20 ms, and night-time awake time up to 10 ms. Heart rate is a well-known indicator of recovery quality, and improvements in both heart rate and heart rate variability (HRV) show positive variations of 15–20 ms in reaction test performance. To expand the current research on cognitive computing, we introduce smartphone typing metrics as a proxy or a complementary method for continuous passive measurement of cognitive vigilance and report on statistically significant correlations in PVT performance and typing speed and error rates. Together, our findings contribute to ubiquitous computing via a longitudinal case study with a novel wearable device, the resulting findings on the association between sleep and cognitive function, and the introduction of smartphone keyboard typing as a proxy of cognitive function.

Financial stability is a key challenge for individuals living with bipolar disorder (BD). Symptomatic periods in BD are associated with poor financial decision-making, contributing to a negative cycle of worsening symptoms and an increased risk of bankruptcy. There has been an increased focus on designing supportive financial technologies (fintech) to address varying and intermittent needs across different stages of BD. However, little is known about this population’s expectations and privacy preferences related to financial data sharing for longitudinal care management. To address this knowledge gap, we have deployed a factorial vignette survey using the Contextual Integrity framework. Our data from individuals with BD (N=480) shows that they are open to sharing financial data for long term care management. We have also identified significant differences in sharing preferences across age, gender, and diagnostic subtype. We discuss the implications of these findings in designing equitable fintech to support this marginalized community.

This paper aims to establish relationships between conversational markers and health outcomes using data from cardiopulmonary rehabilitation sessions.Specifically, we used speech and text data from conversations between patients and researchers to assess exercise compliance and psychological wellbeing.We trained a Multimodal Transformer (MMT) on speech, transcript, and ground-truth labels.We further evaluate MMT's predictive performance by using session summaries generated by three Large Language Models (LLMs), which focused on dialogue characteristics (e.g., sentiment, thematic content, and future planning).Our findings establish the feasibility of augmenting speech and language processing of clinical sessions to improve decision-making and health outcomes.

Bipolar disorder (BD) is a serious mental illness that requires life-long management. Manic and depressive mood episodes in BD are characterized by idiosyncratic behavioral changes. Identifying these early-warning signs is critical for effective illness management. However, there are unique design constraints for technologies focusing on preemptive assessment and intervention in BD given the need for data-intensive monitoring and balancing user agency. In this paper, we aim to establish acceptance, needs, and concerns regarding a preemptive assessment and intervention system to support longitudinal BD management. We interviewed 10 individuals living with BD. To ground the findings in lived experiences, we used a hypothetical assessment and intervention system focusing on online behaviors. Based on the data, we have identified requirements for effective behavioral monitoring across illness episodes. We have also established design recommendations to support dynamic, longitudinal interventions that can address the evolving user needs for life-long BD management.

Is the current state of fitness applications effective at motivating and satisfying the needs of Hispanic users? With most mHealth research conducted with a predominantly white population, the answer to this question is lacking. In this study, we address this question through a survey study with Hispanic users of fitness applications (N= 211) and use the Motivational Technology Model (MTM) and Self-Determination Theory (SDT) as theoretical frameworks. We found that using interactivity features is essential to inspire more autonomous forms of motivation to use fitness applications. This is because interactivity helps satisfy users’ needs for relatedness. However, interactivity also decreased autonomy and competence suggesting the need to design fitness applications that increase relatedness without compromising autonomy. Implications for the design of fitness applications for the population at large and Hispanics, in particular, are discussed.

Recent studies have used speech signals to assess depression. However, speech features can lead to serious privacy concerns. To address these concerns, prior work has used privacy-preserving speech features. However, using a subset of features can lead to information loss and, consequently, non-optimal model performance. Furthermore, prior work relies on a centralized approach to support continuous model updates, posing privacy risks. This paper proposes to use Federated Learning (FL) to enable decentralized, privacy-preserving speech analysis to assess depression. Using an existing dataset (DAIC-WOZ), we show that FL models enable a robust assessment of depression with only 4–6% accuracy loss compared to a centralized approach. These models also outperform prior work using the same dataset. Furthermore, the FL models have short inference latency and small memory footprints while being energy-efficient. These models, thus, can be deployed on mobile devices for real-time, continuous, and privacy-preserving depression assessment at scale.

Active listening is a well-known skill applied in human communication to build intimacy and elicit self-disclosure to support a wide variety of cooperative tasks. When applied to conversational UIs, active listening from machines can also elicit greater self-disclosure by signaling to the users that they are being heard, which can have positive outcomes. However, it takes considerable engineering effort and training to embed active listening skills in machines at scale, given the need to personalize active-listening cues to individual users and their specific utterances. A more generic solution is needed given the increasing use of conversational agents, especially by the growing number of socially isolated individuals. With this in mind, we developed an Amazon Alexa skill that provides privacy-preserving and pseudo-random backchanneling to indicate active listening. User study (N = 40) data show that backchanneling improves perceived degree of active listening by smart speakers. It also results in more emotional disclosure, with participants using more positive words. Perception of smart speakers as active listeners is positively associated with perceived emotional support. Interview data corroborate the feasibility of using smart speakers to provide emotional support. These findings have important implications for smart speaker interaction design in several domains of cooperative work and social computing.

"Always-on" smart speakers have raised privacy and security concerns, to address which vendors have introduced customizable privacy settings. But, does the act of customizing one's privacy preferences have any effects on user experience and trust? To address this question, we developed an app for Amazon Alexa and conducted a user study (N = 90). Our data show that the affordance to customize privacy settings enhances trust and usability for regular users, while it has adverse effects on power users. In addition, only enabling privacy-setting customization without allowing content customization negatively affects trust among users with higher privacy concerns. When they can customize both content and privacy settings, user trust is highest. That is, while privacy customization may cause reactance among power users, allowing privacy-concerned individuals to simultaneously customize content can help to alleviate the resultant negative effect on trust. These findings have implications for designing more privacy-sensitive and trustworthy smart speakers.

Public speaking anxiety is one of the most common social phobias. We explore the feasibility of using a conversational agent to reduce this anxiety. We developed a public-speaking tutor on the Amazon Alexa platform that enables users to engage in cognitive reconstruction exercises. We also investigated how the sociability of the agent might affect its performance as a tutor. A user study of 53 college students with fear of public speaking showed that the interaction with the agent served to assuage pre-speech state anxiety. Agent sociability improved the sense of interpersonal closeness, which was associated with lower pre-speech anxiety. Moreover, sociability of the agent increased participants' satisfaction and their willingness to continue engagement. Our findings, thus, have implications not only for addressing public speaking anxiety in a scalable way but also for the design of future conversational agents using smart speaker platforms.

Alertness is a crucial component of our cognitive performance. Reduced alertness can negatively impact memory consolidation, productivity and safety. As a result, there has been an increasing focus on continuous assessment of alertness. The existing methods usually require users to wear sensors, fill out questionnaires, or perform response time tests periodically, in order to track their alertness. These methods may be obtrusvie to some users, and thus have limited capability. In this work, we propose AlertnessScanner, a computer-vision-based system that collects in-situ pupil information to model alertness in the wild. We conducted two in-the-wild studies to evaluate the effectiveness of our solution, and found that AlertnessScanner passively and unobtrusively assess alertness. We discuss the implications of our findings and present opportunities for mobile applications that measure and act upon changes in alertness.

Early detection of mental health changes in individuals with serious mental illness is critical for effective intervention. CrossCheck is the first step towards the passive monitoring of mental health indicators in patients with schizophrenia and paves the way towards relapse prediction and early intervention. In this paper, we present initial results from an ongoing randomized control trial, where passive smartphone sensor data is collected from 21 outpatients with schizophrenia recently discharged from hospital over a period ranging from 2-8.5 months. Our results indicate that there are statistically significant associations between automatically tracked behavioral features related to sleep, mobility, conversations, smart-phone usage and self-reported indicators of mental health in schizophrenia. Using these features we build inference models capable of accurately predicting aggregated scores of mental health indicators in schizophrenia with a mean error of 7.6% of the score range. Finally, we discuss results on the level of personalization that is needed to account for the known variations within people. We show that by leveraging knowledge from a population with schizophrenia, it is possible to train accurate personalized models that require fewer individual-specific data to quickly adapt to new users.

Given the importance of creativity for both personal and societal achievements, there have been consistent efforts to stimulate creative ability. But an important environmental factor --- blue (i.e., short wavelength) light --- has been relatively unexplored to date. Blue light improves a number of cognitive processes (e.g., attention, working memory and sleep) known to influence our creative abilities. In this study, we investigate the effects of blue light on enhancing creativity in tasks and compare it to the effects of walking, which has been shown to stimulate creative ability. Based on data from 21 participants over 2 weeks, we found that blue light resulted in a 24.3% increase in convergent thinking ability, while walking improved divergent thinking by 18%. We discuss the implications of the findings within the context of UbiComp research. To the best of our knowledge, this is the first systematic examination of the impact of blue light on convergent and divergent thinking ability.

Throughout the day, our alertness levels change and our cognitive performance fluctuates. The creation of technology that can adapt to such variations requires reliable measurement with ecological validity. Our study is the first to collect alertness data in the wild using the clinically validated Psychomotor Vigilance Test. With 20 participants over 40 days, we find that alertness can oscillate approximately 30% depending on time and body clock type and that Daylight Savings Time, hours slept, and stimulant intake can influence alertness as well. Based on these findings, we develop novel methods for unobtrusively and continuously assessing alertness. In estimating response time, our model achieves a root-mean-square error of 80.64 milliseconds, which is significantly lower than the 500ms threshold used as a standard indicator of impaired cognitive ability. Finally, we discuss how such real-time detection of alertness is a key first step towards developing systems that are sensitive to our biological variations.

Our body clock causes considerable variations in our behavioral, mental, and physical processes, including alertness, throughout the day. While much research has studied technology usage patterns, the potential impact of underlying biological processes on these patterns is under-explored. Using data from 20 participants over 40 days, this paper presents the first study to connect patterns of mobile application usage with these contributing biological factors. Among other results, we find that usage patterns vary for individuals with different body clock types, that usage correlates with rhythms of alertness, that app use features such as duration and switching can distinguish periods of low and high alertness, and that app use reflects sleep interruptions as well as sleep duration. We conclude by discussing how our findings inform the design of biologically-friendly technology that can better support personal rhythms of performance.

OBJECTIVE: To evaluate the feasibility of automatically assessing the Social Rhythm Metric (SRM), a clinically-validated marker of stability and rhythmicity for individuals with bipolar disorder (BD), using passively-sensed data from smartphones. METHODS: Seven patients with BD used smartphones for 4 weeks passively collecting sensor data including accelerometer, microphone, location, and communication information to infer behavioral and contextual patterns. Participants also completed SRM entries using a smartphone app. RESULTS: We found that automated sensing can be used to infer the SRM score. Using location, distance traveled, conversation frequency, and non-stationary duration as inputs, our generalized model achieves root-mean-square-error of 1.40, a reasonable performance given the range of SRM score (0-7). Personalized models further improve performance with mean root-mean-square-error of 0.92 across users. Classifiers using sensor streams can predict stable (SRM score ≥3.5) and unstable (SRM score <3.5) states with high accuracy (precision: 0.85 and recall: 0.86). CONCLUSIONS: Automatic smartphone sensing is a feasible approach for inferring rhythmicity, a key marker of wellbeing for individuals with BD.

The increasing adoption of social media provides unprecedented opportunities to gain insight into human nature at vastly broader scales. Regarding the study of population-wide sentiment, prior research commonly focuses on text-based analyses and ignores a treasure trove of sentiment-laden content: images. In this paper, we make methodological and computational contributions by introducing the Smile Index as a formalized measure of societal happiness. Detecting smiles in 9 million geo-located tweets over 16 months, we validate our Smile Index against both text-based techniques and self-reported happiness. We further make observational contributions by applying our metric to explore temporal trends in sentiment, relate public mood to societal events, and predict economic indicators. Reflecting upon the innate, language-independent aspects of facial expressions, we recommend future improvements and applications to enable robust, global-level analyses. We conclude with implications for researchers studying and facilitating the expression of collective emotion through socio-technical systems.

MoodLight is an interactive ambient lighting system that responds to biosensor input related to an individual's current level of arousal. Changes in levels of arousal correspond to fluctuations in the color of light provided by the system, altering the immediate environment in ways intimately related to the user's private internal state. We use this intervention to explore personal and social implications of the ambient display of biosensor data. A design probe study conducted with university students provided the opportunity to observe MoodLight being used by individuals and dyads. Discussion of findings highlights key tensions associated with the dialectics of technology-mediated self-awareness and automated disclosure of personal information, addressing issues of agency, skepticism and uncertainty. This study provides greater understanding of the ways in which the representations of personal informatics, with a focus on ambient feedback, influence our perceptions of ourselves and those around us.

By nature, we are circadian creatures whose bodies' biological clocks drive numerous physiological, mental, and behavioral rhythms. Simultaneously, we are social beings. Accordingly, our internal circadian timings experience interference from externally determined factors such as work schedules and social engagements, and digital connectivity imports additional social constraints that can further misalign our individual body clocks. Misalignment between biological and social time causes social jet lag [50], which has serious physical and mental health consequences. It particularly impacts our sleep processes and neurobehavioral functioning. Examining the interplay between biological rhythms and technology-mediated social interactions, we find that technology may both modulate and reflect circadian rhythms. We also leverage such social-sensor data to infer sleep-related behaviors and disruptions and to analyze variations in attention, cognitive performance, and mood following (in)adequate sleep. We conclude with recommendations for designing technologies attuned to our innate biological traits.

In this paper, we argue that atypical cognitive, perceptual and behavioral characteristics associated with serious mental illnesses should be taken into consideration when designing health technologies. While applications have been developed to assist in the treatment of these illnesses, the specific psychological characteristics of these disorders have rarely been considered extensively in the design process. Here, we explore how an understanding of the low-level characteristics of bipolar disorder, combined with a clinically-validated treatment and patients' lived experience, can inform mHealth design. We present a novel method -- in situ design -- to support ecologically valid design, and demonstrate its use through the co-development with 9 individuals with bipolar disorder of MoodRhythm, a mobile application designed to track and stabilize daily routines. We provide evidence that mHealth design elements tailored to the characteristics and needs of individuals with bipolar disorder can result in engaging interactions.

We often think of ourselves as individuals with steady capabilities. However, converging strands of research indicate that this is not the case. Our biochemistry varies significantly over the course of a 24 hour period. Consequently our levels of alertness, productivity, physical activity, and even sensitivity to pain fluctuate throughout the day. This offers a considerable opportunity for the UbiComp community to identify novel measurements and interventions that can leverage these daily variations. To illustrate this potential, we present results from an empirical study with 9 participants over 97 days investigating whether such variations manifest in low-level smartphone use, focusing on daily rhythms related to sleep. Our findings demonstrate that phone usage patterns can be used to detect and predict individual daily variations indicative of temporal preference, sleep duration, and deprivation. We also identify opportunities and challenges for measuring and enhancing well-being using these simple and effective markers of circadian rhythms.

The rising popularity of the sensor-equipped smartphone is changing the possible scale and scope of human activity inference. The diversity in user population seen in large user bases can overwhelm conventional one-size-fits-all classification approaches. Although personalized models are better able to handle population diversity, they often require increased effort from the end user during training and are computationally expensive. In this paper, we propose an activity classification framework that is scalable and can tractably handle an increasing number of users. Scalability is achieved by maintaining distinct groups of similar users during the training process, which makes it possible to account for the differences between users without resorting to training individualized classifiers. The proposed framework keeps user burden low by leveraging crowd-sourced data labels, where simple natural language processing techniques in combination with multi-instance learning are used to handle labeling errors introduced by low-commitment everyday users. Experiment results on a large public dataset demonstrate that the framework can cope with population diversity irrespective of population size.

In this paper, we describe a novel approach to understand and explain news spreading dynamics on Twitter by using well-known epidemic models. Our underlying hypothesis is that the information diffusion on Twitter is analogous to the spread of a disease. As mathematical epidemiology has been extensively studied, being able to express news spreading as an epidemic model enables us to use a wide range of tools and procedures which have been proven to be both analytically rich and operationally useful. To further emphasize this point, we also show how we can readily use one of such tools - a procedure for detection of influenza epidemics, to detect change of trend dynamics on Twitter.