Self-administered interventions based on natural language processing models for reducing depressive and anxious symptoms: Systematic review and meta-analysis (Preprint)

Опубликована: Апрель 15, 2024

The introduction of Natural Language Processing (NLP) technologies has significantly enhanced the potential self-directed interventions for treating anxiety and depression by improving human-computer interactions. Despite these advancements, particularly in AI Large Models (LLMs), robust evidence validating their effectiveness remains sparse. To determine whether based on NLP models can reduce depressive symptoms. Our study was a systematic review, protocol registered PROSPERO (CRD42023472120). databases we used review are Web Science, SCOPUS, MEDLINE (via PubMed), PsycINFO EBSCO), IEEE Xplore, EMBASE Cochrane Library. quality included studies assessed using JBI Critical Appraisal Tools. 21 articles were selected 16 meta-analysis each outcome. overall showed that self-administered more effective reducing symptoms (SMD=0.819; 95%CI: 0.389-1.250; p<0.001) (SMD=0.272; 95% CI: 0.116-0.428; p=0.001) compared with various control conditions. In subgroup analysis, AI-based shown to be (SMD=1.059 [0.520 1.597]; (SMD=0.302 [0.073 0.532]; p=0.010) pooled Also, NLP-based outperform psychoeducation bibliotherapy both (SMD=1.481 [0.368 2.594]; p=0.009) (SMD=0.561 [0.195 0.927]; p=0.003). addition, than waitlist or no intervention anxious (SMD=0.196 [0.042 0.351]; p=0.013). findings support usefulness self-applied alleviating widely prevalent mental health problems such as Protocol (CRD42023472120)

Язык: Английский

---

Current Landscape and Future Directions for Mental Health Conversational Agents (CAs) for Youth: Scoping Review (Preprint)

Опубликована: Май 30, 2024

Conversational Agents (CAs, chatbots) are systems enabled with the ability to interact users using natural human dialogue. They increasingly used support interactive knowledge discovery of sensitive topics such as mental health topics. While much research on CAs for has focused adult populations, insights from may not apply youth. The aim study was comprehensively evaluate state-of-the-art Following Preferred Reporting Items Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we synthesized 39 peer-reviewed studies specific designed We conducted a scoping review literature characteristics youth, design/computational considerations evaluation outcomes reported in found that most were older peers provide therapeutic and/or educational content promote youth well-being. Most based expert knowledge, few incorporated inputs technical maturity its infancy, focusing building prototypes rule-based models deliver pre-written content, limited safety features respond imminent risk. Research findings suggest while appreciate 24/7 availability friendly/empathetic conversation CAs, they provided by be limited. Finally, concerning trend reviewed did address ethical aspects concerned about privacy confidentiality their data. Our highlights need researchers continue work together align evidence-based lessons learned how best these technologies brings light needing further development evaluation. new large language (LLMs) can make more feasible. Yet, should prioritized. Although preliminary evidence shows positive trends long-term evaluative larger sample sizes robust designs is needed validate efficacy. More importantly, collaborative efforts clinical experts early design phases summative stages build safe, effective, youth-centered CAs. practices risk mitigation N/A

Язык: Английский

---

Exploring Digital Affordances in Online Mental Health Resources: A Scoping Review of Methodologies and Populations

International Journal of Human-Computer Interaction, Год журнала: 2024, Номер unknown, С. 1 - 17

Опубликована: Июль 11, 2024

Язык: Английский

---

Examining the Effects of Cognitive Behavioural Therapy With a Virtual Agent on User Motivation and Improvement in Psychological Distress and Anxiety: Two-Session Experimental Study (Preprint)

JMIR Formative Research, Год журнала: 2024, Номер 8, С. e55234 - e55234

Опубликована: Авг. 14, 2024

Cognitive behavioral therapy (CBT) is a valuable treatment for mood disorders and anxiety. CBT methods, such as cognitive restructuring, are employed to change automatic negative thoughts more realistic ones.

Язык: Английский

---

Disrupted self, therapy, and the limits of conversational AI

Philosophical Psychology, Год журнала: 2024, Номер unknown, С. 1 - 27

Опубликована: Сен. 3, 2024

Conversational agents (CA) are thought to be promising for psychotherapy because they give the impression of being able engage in conversations with human users. However, given high risk therapy patients who already a vulnerable situation, there is need investigate extent which CA contribute goals and discuss CA's limitations, especially complex cases. In this paper, we understand as way dealing existential situations position CAs context therapeutic experience patients. This determined by patient's unique personal specific goals. We suggest that fundamentally dialogical activity, it crucially involves work on self one's self-narrative. brings us our central question: possible productive dialogue, their limitations epistemic agents? will several those show how these undermine possibility engaging illustrate through discussions cases grief abuse.

Язык: Английский

---