Andrew M. Sherrill

Treatments like Cognitive Behavioral Therapy (CBT) have high clinical efficacy for a variety of mental health disorders. Their effectiveness, however, is often limited by low patient engagement with clinical homework. Novel technologies are accordingly being integrated into clinical mental health workflows, but this is a delicate endeavor and studies of in-the-wild clinical deployments remain rare in the CSCW and HCI communities. Lab-based and theoretical design studies are valuable first steps toward integration, but they struggle to disambiguate the real-world impacts of novel technologies on ecologically sensitive clinical environments as the relationships between human actors and technologies are co-constituted. We developed a collective sensing system meant to help contextualize patient homework exercises within a two-week CBT intensive outpatient program (IOP) for PTSD. We then conducted an in-the-wild field deployment study involving 20 patients and 10 clinicians over 15 months. The system received strong performance ratings for usefulness, usability, and preliminary efficacy, and, in interviews, patients and clinicians reported that it facilitated collaborative clinical homework review and accountability throughout the therapy process. We also found four themes related to its unexpected real-world impacts on CBT practice surrounding emotional processing, patient agency, accountability, and contextual trust. Given the observed complexity of relationships developed between patients, clinicians, and the collective sensing system, we ultimately draw from the philosophical tradition of postphenomenology and conduct a “variational cross-examination” of our findings to discuss implications for future deployments. Our work highlights the importance of clinical deployments and the subsequent value of postphenomenological analysis to the CSCW and HCI communities.

The emergence of generative artificial intelligence (GenAI) presents new opportunities for enhancing exposure therapy for anxiety- and trauma-related disorders. GenAI is not a replacement for clinical expertise but a powerful ancillary tool. This article offers practical guidance for clinicians and patients on how to use GenAI to develop and conduct personalized exposure exercises. To help orient clinicians to these tools, we provide an overview of ethical considerations and describe the fundamentals of how to use GenAI between and within sessions. We discuss how GenAI tools can assist in generating exposure stimuli, such as images, videos, and scripts, as well as conversational agents. We also provide guidance on how these tools can assist in brainstorming and organizing exposure hierarchies. Finally, we discuss how exposure therapy can continue to flourish using GenAI through clinician training, the exchange of GenAI stimuli and procedures, and clear messaging from leadership and organizations.

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.