R0024/2026-03-25/Q003/SRC03/E01

Research	R0024 — Sycophancy and Addiction
Run	2026-03-25
Query	Q003
Source	SRC03
Evidence	SRC03-E01
Type	Reported

Synthesis of three peer-reviewed studies on AI chatbot addiction and dopamine mechanisms

URL: https://www.techpolicy.press/ai-chatbots-and-addiction-what-does-the-research-say/

Extract

The article synthesizes three peer-reviewed studies:

1. Shen & Yoon (CHI 2025): Identified four addiction pathways including non-deterministic responses as "reward uncertainty, which tends to increase dopamine release, similar to playing a slot machine" and empathetic/agreeable responses that "let users feel understood and validated."

2. Zhang et al. (IJHCI, August 2025): "Investigating AI Chatbot Dependence: Associations with Internet and Smartphone Dependence, Mental Health Outcomes, and the Moderating Role of Usage Purposes." Found correlations between chatbot usage and mental health outcomes, moderated by usage purpose.

3. Fang et al. (arXiv preprint, March 2025): "How AI and Human Behaviors Shape Psychosocial Effects of Chatbot Use: A Longitudinal Randomized Controlled Study." A longitudinal RCT examining psychosocial effects of chatbot use. Authors include researchers from MIT and OpenAI safety team.

The review explicitly links chatbot notifications to social media patterns: "similar trends have been observed in the context of notifications from social media applications, which research has found to be a major contributor to addiction."

Relevance to Hypotheses

Hypothesis	Relationship	Strength
H1	Supports	Three separate peer-reviewed studies demonstrate research activity in this area
H2	Contradicts	Multiple studies exist at major venues
H3	Supports	Studies are recent (2025) and use different methodologies, suggesting an emerging field

Context

The inclusion of a longitudinal RCT (Fang et al.) is notable — it represents methodological advancement beyond cross-sectional surveys and self-report. However, even this study does not directly measure dopamine levels.