The New School is a private research university in New York City. It was founded in 1919 as The New School for Social Research with an original mission dedicated to academic freedom and intellectual inquiry and a home for progressive thinkers. Since then, the school has grown to house five divisions within the university. These include the Parsons School of Design, the Eugene Lang College of Liberal Arts, the College of Performing Arts (which itself consists of the Mannes School of Music, the School of Drama, and the School of Jazz and Contemporary Music), The New School for Social Research, and the Schools of Public Engagement.In addition, the university maintains the Parsons Paris campus and has also launched or housed a range of institutions, such as the international research institute World Policy Institute, the Philip Glass Institute, the Vera List Center for Art and Politics, the India China Institute, the Observatory on Latin America, and the Center for New York City Affairs. It is classified among "R2: Doctoral Universities – High research activity". Approximately 10,000 students are enrolled in undergraduate and postgraduate programs and disciplines including design, social sciences, music, liberal arts, humanities, architecture, fine arts, design, drama, finance, psychology, and public policy.Its faculty and alumni include numerous notable businesspeople, political figures, fashion designers, journalists, musicians, and artists. Notable students and alumni who have achieved prominence in political and business fields include economist Heather Boushey, a member of President Joe Biden's Council of Economic Advisers; Chris Hughes, co-founder of Facebook; Douglas Cliggott, former chief investment strategist for JPMorgan Chase; entrepreneur Stewart Krentzman, formerly an executive at Oki Electric Industry; former President of Israel Shimon Peres; and former Brazilian Minister of Finance Nelson Barbosa. In the arts, notable alumni include the artist-activist Harry Belafonte; fashion designers Alexander Wang, Marc Jacobs, and Donna Karan; actors Marlon Brando, Tony Curtis, Rod Steiger, Bea Arthur, and Bradley Cooper; jazz pianist Bill Evans; trumpeter/composer Roy Hargrove; songwriter Burt Bacharach; soprano Frederica von Stade; playwright Lorraine Hansberry; composer John Cage; and artists Norman Rockwell, Ai Weiwei, Jasper Johns and Julie Umerle.
Recent advances in multimodal wearable biosensing enable continuous, noninvasive or minimally invasive monitoring of physical, physiological, and biochemical states in daily life. In parallel, multidomain AI architectures are increasingly capable of fusing heterogeneous streams, creating new opportunities for scalable, patient-specific health analytics. Yet many sensor-AI systems remain narrow, tracking limited parameters, and often emphasize real-time signals while underutilizing longitudinal clinical context and structured medical knowledge that could strengthen clinical reasoning. Here, we propose a pathway to decentralized healthcare that unites multimodal wearable biosensing with multidomain AI. We review recent progress across wearable sensing modalities and summarize how multisensory fusion can improve patient profiling, enhance diagnostic discrimination, and enable earlier risk prediction. We then describe AI pipelines that integrate biosensor measurements with electronic health records and curated medical literature and knowledge graphs to support evidence-grounded decision support. Finally, we discuss remaining challenges, including data quality and cross-modality alignment, privacy and governance for cross-domain data sharing, and robust generalization under real-world heterogeneity. We highlight emerging opportunities in continual learning, retrieval-augmented reasoning, and closed-loop therapeutics. This "from biosignals to decisions" framework advances AI-enabled decentralized healthcare by shifting actionable insights from the clinic into everyday environments.
Attachment theory, with its core concepts, perspectives, and insights developed over the past five decades, is influential for professionals working with young children. However, practitioners face challenges translating attachment theory and research into practical applications. This manifests in attachment myths, theoretical misinterpretations, and inconsistency of application. This state-of-the-art review is authored by 47 attachment researchers and practitioners and examines key insights from attachment theory to facilitate attachment-aware practice for professionals working with children and their caregivers. Following the ongoing debate on practical relevance in attachment theory, we present both 'strict' and 'expansive' translational perspectives on applications for addressing preventative or clinical attachment concerns. We first review core attachment propositions, based on replicated research of attachment and caregiving. We next address common misconceptions that hinder adequate practical applications. We present measures of attachment and sensitive parenting that might be helpful for practitioners. We also review evidence-based and promising attachment interventions, discussing core components of (preventative) support for parents or caregivers and the children in their care. We emphasize that attachment theory's clinical value lies not in assigning attachment classifications, but rather in understanding crucial insights into caregiving and early socioemotional development (e.g., secure base phenomena; the value of safe, stable, and shared good-enough care), developed in attachment research over the past 50 years, that may inform policy and clinical reasoning and areas for prevention and intervention.
Previous network analyses have demonstrated that the connection between positivity avoidance strategies strengthens as depression increases. However, the connection between emotion regulation strategies for positive emotions, that are employed to either downplay or upregulate positive affect responsively, have not been explored across depression severity. Using moderated network analyses, we examined the associations between positive emotion regulation strategies (i.e., dampening and amplifying) and affect (i.e., positive and negative) at depression levels ranging from none to severe (N = 754; women = 517, men = 233; Mage = 19.30, SDage = 3.34). We conducted 6 iterative moderated networks with individual positive emotion regulation strategies and the subscale of the opposite valence (e.g., dampening items and the amplifying subscale). Results indicate that as depression increased, associations between dampening strategies and amplifying became increasingly more negative. Additionally, the positive associations between dampening items became more strongly connected as depression increased. Severe depressive symptoms are associated with a strong negative relation between dampening and amplifying, suggesting a tendency to dampen positivity and difficulty upregulating positive affect. Future work should examine the longitudinal relationship between depression and positive emotion regulation to better understand the course of these response strategies.
Importance:Large language models (LLMs) are increasingly applied to mental health contexts, yet their capacity to generate responses that align with evidence-based psychotherapy remains uncertain. Motivational interviewing (MI), a structured counseling approach, provides an empirically grounded setting for evaluating alignment between LLM-generated and human therapist responses. Objective:To evaluate how closely an LLM's responses align with therapist responses in MI sessions, using automated similarity metrics. Design, Setting, and Participants:This cross-sectional study used high-fidelity therapist-client transcripts annotated with the Motivational Interviewing Treatment Integrity system. Transcripts were sourced from publicly available counseling videos. For each therapist turn, the GPT-4o LLM generated a response using a standardized, MI-informed prompt based on the preceding conversation context. Analyses were conducted between March and May 2025. Main Outcomes and Measures:Alignment between LLM-generated and therapist responses was assessed using (1) cosine similarity based on sentence embeddings to capture semantic overlap and (2) DeepEval, a contextual deep-learning-based metric assessing coherence and contextual appropriateness. A therapist topic-consistency index quantified within-session thematic coherence and was examined as a moderator of alignment. Results:A total of 3706 therapist turns from 154 MI sessions were evaluated. Mean (SD) DeepEval scores were higher than mean (SD) cosine similarity scores (0.72 [0.31] vs 0.29 [0.20]; P < .001), suggesting limited semantic overlap despite greater contextual appropriateness. Therapist topic consistency significantly moderated similarity, where cosine similarity was higher in high-consistency than low-consistency sessions (mean [SD] difference, 0.027 [0.007]; t3706 = 3.987; P < .001), as was DeepEval score (mean [SD] difference, 0.038 [0.010]; t3706 = 3.747; P < .001). Correlation between metrics was negligible (Spearman ρ, -0.01), indicating that they captured distinct aspects of response alignment. LLM performance declined slightly across longer conversations (mean [SD] slope reduction for cosine similarity, -0.0005 [0.0016], and for DeepEval, -0.0005 [0.0022]), with increased verbosity and signs of reduced contextual grounding. Conclusions and Relevance:In this cross-sectional study of 154 MI sessions, prompted LLMs showed general alignment with therapist responses in MI-oriented conversations, as judged by automated similarity metrics. However, limitations in long-range coherence, stylistic alignment, and the use of indirect proxies for therapeutic quality highlight the need for improved prompt design, MI-specific evaluation methods, and clinical validation before integration into mental health care.
This paper advances the concept of data worlds as a critical lens for understanding both the making of the field of global health, and the ethos of philanthrocapitalism. Constituted through particular forms of expertise, infrastructure and techniques, data worlds are not mere repositories of information. Rather, they actively constitute and circumscribe the field of global health by privileging specific problems and interventions. Frequently constructed through the work of philanthropic foundations, these data worlds shape how global health challenges are rendered visible and actionable. Attending to their selective exclusions and inclusions allows us to see like a philanthrocapitalist. Drawing on an analysis of the Gates Foundation's global health work in India, the paper demonstrates how data worlds can serve as an archive for understanding how particular forms of knowledge and governance get authorized in global health.
