The effect of personalizing a psychotherapy conversational agent on therapeutic bond and usage intentions

Опубликована: Март 18, 2024

While 33.6% of college students suffer from mental health problems, only 24.6% these with symptoms would seek professional help due to their personal attitudes or costs associated therapy. Psychotherapy chatbots may offer a solution as they are always available, anonymous, and cost-effective. Research has shown that can significantly reduce anxiety depression. However, there is lack understanding about the personalization preferences users effects on outcomes. To investigate this, we developed personalizable psychotherapy chatbot designed provide personalized help. In randomized controlled trial (n = 54), participants were either assigned condition non-personalizable control condition. After 1 week usage, had higher therapeutic bond version compared baseline. fact, was similar between psychologist his client. This promising result, high been linked success in psychotherapy. Participants reported therapy style, personality, avatar most important aspects chatbot. also liked chatbot's usage name transparency what learned them. These features likely for establishing strong users. ability personalize no impact intentions participants. be explained by fact both conditions equally able them health. 53 indicated willing use when integrated human therapist. findings indicate potential need further research integration traditional

Язык: Английский

---

The Role of Humanization and Robustness of Large Language Models in Conversational Artificial Intelligence for Individuals With Depression: A Critical Analysis

JMIR Mental Health, Год журнала: 2024, Номер 11, С. e56569 - e56569

Опубликована: Апрель 27, 2024

Abstract Large language model (LLM)–powered services are gaining popularity in various applications due to their exceptional performance many tasks, such as sentiment analysis and answering questions. Recently, research has been exploring potential use digital health contexts, particularly the mental domain. However, implementing LLM-enhanced conversational artificial intelligence (CAI) presents significant ethical, technical, clinical challenges. In this viewpoint paper, we discuss 2 challenges that affect of CAI for individuals with issues, focusing on case patients depression: tendency humanize lack contextualized robustness. Our approach is interdisciplinary, relying considerations from philosophy, psychology, computer science. We argue humanization hinges reflection what it means simulate “human-like” features LLMs role these systems should play interactions humans. Further, ensuring contextualization robustness requires considering specificities production depression, well its evolution over time. Finally, provide a series recommendations foster responsible design deployment therapeutic support depression.

Язык: Английский

---

Does realizing strengths, insight, and behavioral practice through a psychological intervention promote personality change? An intensive longitudinal study

European Journal of Personality, Год журнала: 2024, Номер unknown

Опубликована: Янв. 15, 2024

The mechanisms of change underlying the effectiveness personality interventions are largely unclear. In this study, we used data from a three-month digital intervention with an intensive longitudinal design to test whether greater realization general factors is partly responsible for change. Participants ( N = 679, 53.0% female; age: M 25.3 years, SD 7.1) seeking increase either Emotional Stability, Conscientiousness, or Extraversion provided self-ratings on their weekly states and three generic strengths, insights, behavioral practice. We found single-factor structure within between individuals. Results showed within-person increases in Extraversion, Conscientiousness as well across intervention. Changes were coupled changes factors. Finally, results provide initial support hypothesis that effects Within-person associated subsequent Stability during following week. present findings highlight need better understand how why people result interventions.

Язык: Английский

---

Me and My AI Bot: Exploring the 'AIholic' Phenomenon and University Students' Dependency on Generative AI Chatbots - Is This the New Academic Addiction?

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Май 23, 2024

Amidst the buzz of technological advancement in education, our study unveils a more disconcerting narrative surrounding student chatbot interactions. Our investigation has found that students, primarily driven by intrinsic motivations like competence and relatedness, increasingly lean on chatbots. This dependence is not just preference but borders an alarming reliance, magnified exponentially their individual risk perceptions. While celebrating AI's rapid integration education tempting, results raise urgent red flags. Many hypotheses were supported, pointing toward potential over-dependence Nevertheless, unpredictable outcomes most revealing, exposing terrain role education. It no longer matter if how deep rabbit hole dependency goes. As we stand cusp educational revolution, caution urgently needed. Before wholly embrace chatbots as primary educators, it imperative to understand repercussions replacing human touch with AI serves stark wake-up call, urging stakeholders reconsider unchecked learning environments. The future may very well be digital, at what cost connection autonomy?

Язык: Английский

---

Comparing button-based chatbots with webpages for presenting fact-checking results: A case study of health information

Information Processing & Management, Год журнала: 2022, Номер 60(2), С. 103203 - 103203

Опубликована: Дек. 13, 2022

Язык: Английский

---

Determinants of Chatbot adoption among older adults: An extended TAM approach using PLS-SEM

Information Development, Год журнала: 2025, Номер unknown

Опубликована: Фев. 6, 2025

This study investigates factors influencing chatbot adoption among older adults by extending the Technology Acceptance Model (TAM) with computer anxiety (CANX) and self-efficacy (CSE) to address demographic-specific barriers. Data from 150 participants in Taoyuan City, who engaged two chatbots for information services, were analyzed using partial least squares structural equation modeling. Results show that perceived ease of use (PEOU) positively influences both usefulness (PU) attitude (ATT), PU being strongest predictor behavioral intention (BI), followed ATT PEOU. CSE was found enhance PEOU, while CANX negatively affected it. Explaining 69.3% BI variance, findings validate TAM’s relevance emphasize significance addressing psychological barriers technology adoption. The underscore need interfaces prioritize accessibility cater specific needs elderly users. Additionally, they highlight importance designing provide functional benefits engagement adults.

Язык: Английский

---

Creating AI Persona

SpringerBriefs in computer science, Год журнала: 2025, Номер unknown, С. 49 - 59

Опубликована: Янв. 1, 2025

Язык: Английский

---

Similarity attracts, or does it? studying personality-based convergence and sense of engagement with a digital health assistant

Telematics and Informatics, Год журнала: 2025, Номер unknown, С. 102262 - 102262

Опубликована: Март 1, 2025

Язык: Английский

---

Artificial intelligence vs. human coaches: examining the development of working alliance in a single session

Frontiers in Psychology, Год журнала: 2025, Номер 15

Опубликована: Апрель 15, 2025

The collaborative relationship, or working alliance, between a client and their coach is well-recognized factor that contributes to the effectiveness of coaching. rise artificial intelligence (AI) challenges us explore whether human-to-human relationships can extend AI, potentially reshaping future Our presumption skills professional human coaches surpass AI in forging effective stands untested - but we really claim this advantage? purpose study was examine perceptions being coached by simulated coach, who embodied as conversational vocal live-motion avatar, compared partnering with coach. mixed methods randomized controlled trial explored if how ratings alliance coaching process aligned two types an alternative treatments design. Both treatment groups identified personal goal pursue had one 60-min session guided CLEAR (contract, listen, explore, action, review) model. Quantitative data were captured through surveys qualitative input open-ended survey questions debrief interviews. To sidestep rapid obsolescence technology, engineered using Wizard Oz approach facilitate advanced experience, participants unknowingly interacting expert coaches. aim glean insights into reactions future, fully autonomous capabilities results showed built similar moderately high levels both types, no significant difference treatments. Qualitative themes indicated client's connection existed within context wherein guide used variety techniques support plan towards goal. Overall, believed they engaging assigned type, while five coaches, acting confederates, blinded roles. Clients are willing appreciate building partnerships which has research practical implications.

Язык: Английский

---

AI-Enabled Conversational Journaling for Advancing Parkinson's Disease Symptom Tracking

Опубликована: Апрель 24, 2025

Язык: Английский

---