The Opportunities and Risks of Large Language Models in Mental Health.

Unlabelled: Global rates of mental health concerns are rising, and there is increasing realization that existing models of mental health care will not adequately expand to meet the demand. With the emergence of large language models (LLMs) has come great optimism regarding their promise to create novel, large-scale solutions to support mental health. Despite their nascence, LLMs have already been applied to mental health-related tasks. In this paper, we summarize the extant literature on efforts to use LLMs to provide mental health education, assessment, and intervention and highlight key opportunities for positive impact in each area. We then highlight risks associated with LLMs' application to mental health and encourage the adoption of strategies to mitigate these risks. The urgent need for mental health support must be balanced with responsible development, testing, and deployment of mental health LLMs. It is especially critical to ensure that mental health LLMs are fine-tuned for mental health, enhance mental health equity, and adhere to ethical standards and that people, including those with lived experience with mental health concerns, are involved in all stages from development through deployment. Prioritizing these efforts will minimize potential harms to mental health and maximize the likelihood that LLMs will positively impact mental health globally.

A metric for characterizing the arm nonuse workspace in poststroke individuals using a robot arm.

An overreliance on the less-affected limb for functional tasks at the expense of the paretic limb and in spite of recovered capacity is an often-observed phenomenon in survivors of hemispheric stroke. The difference between capacity for use and actual spontaneous use is referred to as arm nonuse. Obtaining an ecologically valid evaluation of arm nonuse is challenging because it requires the observation of spontaneous arm choice for different tasks, which can easily be influenced by instructions, presumed expectations, and awareness that one is being tested. To better quantify arm nonuse, we developed the bimanual arm reaching test with a robot (BARTR) for quantitatively assessing arm nonuse in chronic stroke survivors. The BARTR is an instrument that uses a robot arm as a means of remote and unbiased data collection of nuanced spatial data for clinical evaluations of arm nonuse. This approach shows promise for determining the efficacy of interventions designed to reduce paretic arm nonuse and enhance functional recovery after stroke. We show that the BARTR satisfies the criteria of an appropriate metric for neurorehabilitative contexts: It is valid, reliable, and simple to use.

Correlation between performance and quantity/variability of leg exploration in a contingency learning task during infancy.

Quantity and quality of motor exploration are proposed to be fundamental for infant motor development. However, it is still not clear what types of motor exploration contribute to learning. To determine whether changes in quantity of leg movement and/or variability of leg acceleration are related to performance in a contingency learning task, twenty 6-8-month-old infants with typical development participated in a contingency learning task. During this task, a robot provided reinforcement when the infant's right leg peak acceleration was above an individualized threshold. The correlation coefficient between the infant's performance and the change in quantity of right leg movement, linear variability, and nonlinear variability of right leg movement acceleration from baseline were calculated. Simple linear regression and multiple linear regression were calculated to explain the contribution of each variable to the performance individually and collectively. We found significant correlation between the performance and the change in quantity of right leg movement (r = 0.86, p < 0.001), linear variability (r = 0.71, p < 0.001), and nonlinear variability (r = 0.62, p = 0.004) of right leg movement acceleration, respectively. However, multiple linear regression showed that only quantity and linear variability of leg movements were significant predicting factors for the performance ratio (p < 0.001, adjusted R2 = 0.94). These results indicated that the quantity of exploration and variable exploratory strategies could be critical for the motor learning process during infancy.

Modeling engagement in long-term, in-home socially assistive robot interventions for children with autism spectrum disorders.

Socially assistive robotics (SAR) has great potential to provide accessible, affordable, and personalized therapeutic interventions for children with autism spectrum disorders (ASD). However, human-robot interaction (HRI) methods are still limited in their ability to autonomously recognize and respond to behavioral cues, especially in atypical users and everyday settings. This work applies supervised machine-learning algorithms to model user engagement in the context of long-term, in-home SAR interventions for children with ASD. Specifically, we present two types of engagement models for each user: (i) generalized models trained on data from different users and (ii) individualized models trained on an early subset of the user's data. The models achieved about 90% accuracy (AUROC) for post hoc binary classification of engagement, despite the high variance in data observed across users, sessions, and engagement states. Moreover, temporal patterns in model predictions could be used to reliably initiate reengagement actions at appropriate times. These results validate the feasibility and challenges of recognition and response to user disengagement in long-term, real-world HRI settings. The contributions of this work also inform the design of engaging and personalized HRI, especially for the ASD community.

Toward Predicting Infant Developmental Outcomes From Day-Long Inertial Motion Recordings.

As improvements in medicine lower infant mortality rates, more infants with neuromotor challenges survive past birth. The motor, social, and cognitive development of these infants are closely interrelated, and challenges in any of these areas can lead to developmental differences. Thus, analyzing one of these domains - the motion of young infants - can yield insights on developmental progress to help identify individuals who would benefit most from early interventions. In the presented data collection, we gathered day-long inertial motion recordings from N = 12 typically developing (TD) infants and N = 24 infants who were classified as at risk for developmental delays (AR) due to complications at or before birth. As a first research step, we used simple machine learning methods (decision trees, k-nearest neighbors, and support vector machines) to classify infants as TD or AR based on their movement recordings and demographic data. Our next aim was to predict future outcomes for the AR infants using the same simple classifiers trained from the same movement recordings and demographic data. We achieved a 94.4% overall accuracy in classifying infants as TD or AR, and an 89.5% overall accuracy predicting future outcomes for the AR infants. The addition of inertial data was much more important to producing accurate future predictions than identification of current status. This work is an important step toward helping stakeholders to monitor the developmental progress of AR infants and identify infants who may be at the greatest risk for ongoing developmental challenges.

Socially assistive robotics: Human augmentation versus automation.

Intelligent, interactive systems provide assistance by facilitating social interactions rather than by automating physical tasks.

Assessing the quality and quantity of social interaction in a socially assistive robot-guided therapeutic setting.

We present an application of a socially assistive robotics (SAR) system in a therapeutic setting. We examine the amount of interaction elicited by the robot in a therapeutic setting with individuals post-stroke. We examine the role of various communication modalities, and their affects on the participants' responses. Seven participants of mild to moderate functional impairment due to stroke interacted with our SAR system during three sessions of motor task practice. The robot guided the users as they performed a wire puzzle task, while providing them with feedback about their performance. We evaluated the amount of verbalization and eye contact made with the robot. Our results indicate that users make eye contact more often than they verbalize when interacting with the robot. Further, user interactions are most frequent at the beginning of a practice session, and occur less frequently as the session progresses. When a user observes that the robot is not responding to a certain type of communication, the user limits the use of that communication modality. These insights should be useful in the design of future robot-based therapeutic interventions.

Socially assistive robotics for post-stroke rehabilitation.

BACKGROUND: Although there is a great deal of success in rehabilitative robotics applied to patient recovery post stroke, most of the research to date has dealt with providing physical assistance. However, new rehabilitation studies support the theory that not all therapy need be hands-on. We describe a new area, called socially assistive robotics, that focuses on non-contact patient/user assistance. We demonstrate the approach with an implemented and tested post-stroke recovery robot and discuss its potential for effectiveness. RESULTS: We describe a pilot study involving an autonomous assistive mobile robot that aids stroke patient rehabilitation by providing monitoring, encouragement, and reminders. The robot navigates autonomously, monitors the patient's arm activity, and helps the patient remember to follow a rehabilitation program. We also show preliminary results from a follow-up study that focused on the role of robot physical embodiment in a rehabilitation context. CONCLUSION: We outline and discuss future experimental designs and factors toward the development of effective socially assistive post-stroke rehabilitation robots.