The exact text we saved

This is the version exactly as it was saved, character for character. The SHA-256 hash below is a 64-character code worked out from that text. Run the same text through SHA-256 and you get the same code; change a single letter and the code comes out completely different. Each version also stores the previous version’s hash, which chains them together so no version can be swapped out quietly.

TRUVACE RECORD VERSION
record: TRV-2026-2617
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-11-12T08:04:35.484326Z
status: published
lens: p_space
sector: health
headline: How Consumers Evaluate Human, AI , and Hybrid Therapists: Psychological Pathways to Expected Service Satisfaction
dek: ABSTRACT As generative artificial intelligence becomes prominent in digital therapy, understanding how consumers evaluate human, AI, and hybrid therapists is increasingly important. Drawing on insights from signaling, construal level, stimulus‐organism‐response, and perceived risk research, this paper examines how therapy delivery mode shapes expected service satisfaction through capabilities‐based, relational‐based, and risk‐based pathways. Across two experiments, consumers reported higher expected service sati…
gain_title: (none)
problem_title: In two experiments with 395 participants combined, consumers expected to be less satisfied with AI or hybrid AI/human therapists than with human therapists.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: In two experiments with 395 participants combined, consumers expected to be less satisfied with AI or hybrid AI/human therapists than with human therapists.
problem_evidence: consumers reported higher expected service satisfaction for human therapists than for AI or hybrid (AI/human) therapists
quick_read: Two experiments found that consumers expected to be more satisfied with a human therapist than with an AI therapist or a hybrid AI/human therapist. Study 1 had 200 participants and Study 2 had 195.

In Study 1, the gap was partly explained by how competent and expert participants judged the therapist to be, and by how psychologically close they felt to it. In Study 2, the preference for human over hybrid therapists was fully explained by skepticism of the therapist.

Privacy concerns and psychological closeness did not significantly explain the human-over-hybrid preference in Study 2. The authors say the findings bear on consumer protection, transparency and the design of responsible digital therapy.
limitation: The studies measured expected service satisfaction, not satisfaction or outcomes after actual therapy with an AI or hybrid therapist.
tag: Evidence-backed problem
key_points: Study 1 (200 participants) found higher expected satisfaction for human therapists than for AI or hybrid therapists. | In Study 1, perceived competence, perceived expertise and psychological closeness partially mediated the effect. | Study 2 (195 participants) found the preference for human over hybrid therapists was fully mediated by skepticism of the therapist. | In Study 2, privacy concerns and psychological closeness were not significant mediators. | The paper draws on signaling, construal level, stimulus-organism-response and perceived risk research.
rundown: The paper tests how therapy delivery mode (human, AI or hybrid AI/human) shapes expected service satisfaction. It frames three routes: capabilities-based, relational-based and risk-based.

Study 1 (N = 200) compared human, AI and hybrid therapists and found the human advantage was partly carried by perceived competence, perceived expertise and psychological closeness. Study 2 (N = 195) found that skepticism fully accounted for the preference for human over hybrid therapists, while privacy concerns did not.

The authors conclude that consumer responses depend on inferred capability, relational connection and skepticism. They point to implications for consumer protection, transparency and responsible digital therapy design.
sources:
- peer_reviewed | Journal of Consumer Affairs | https://doi.org/10.1111/joca.70061 | 2026-05-26
prev: 0000000000000000000000000000000000000000000000000000000000000000

SHA-256 hash

31252d8a7785f873ac1be3ab3d41682d6e86c8fb6aea659aafae90a494cf3dbc

Previous version’s hash

0000000000000000000000000000000000000000000000000000000000000000