The research team, led by Hong Wu, Quzhi Liu, and colleagues at Hohai University in Nanjing, designed a cross-sectional online survey around two parallel 11-item experience scales. Each item probed a distinct dimension of perceived support quality: acceptance, understanding, goal alignment, collaboration, professional competence, response efficiency, safety and privacy, trust, reduced concern about being judged, willingness for deep self-disclosure, and disclosure of negative emotions. Because every participant rated both modalities on identical items, the design enabled paired statistical comparisons that control for the stable personality traits and life circumstances each individual brings to any evaluation. This within-person approach matters enormously in this field, because people who choose to use chatbots may differ systematically from people who seek therapy, and those pre-existing differences can masquerade as modality effects in cruder between-group studies.

The statistical machinery behind the study was deliberately layered. The primary analysis used a paired-samples t test on a 10-item composite score, constructed after the researchers discovered that the response-efficiency item was not strictly measurement-equivalent across the two modalities, meaning that a chatbot’s instant reply and a therapist’s considered response may simply not measure the same underlying quality. To guard against distortion from this problematic item, the team removed it from both modality scores and re-ran the comparison. They also applied paired Wilcoxon signed-rank tests with Monte Carlo two-tailed p values and Holm correction to the item-level data, a procedure that controls the inflated risk of false positives when many comparisons are tested simultaneously. Finally, a repeated-measures linear mixed-effects model served as a sensitivity analysis, explicitly modeling the within-person paired structure and testing whether the size of the modality difference varied according to individual characteristics such as attachment style.

The headline finding was a small but statistically robust overall advantage for AI-based support. On the 10-item composite, AI support scored a mean of 57.77 with a standard deviation of 6.55, while human counseling scored 56.80 with a standard deviation of 6.75. The mean difference, calculated as human minus AI, was −0.97 points, with a 95 percent confidence interval spanning −1.73 to −0.21, a paired t value of −2.53 on 232 degrees of freedom, and a p value of .012. The standardized effect size was modest, dz = 0.17, which the researchers emphasize is a small difference. Yet its persistence after the potentially non-equivalent response-efficiency item was excluded suggests the pattern is not merely an artifact of comparing a chatbot’s instant replies with a therapist’s slower turnaround. At the item level, AI support received higher ratings on five exploratory indicators, while human counseling won on exactly two: mutual understanding and professional competence.

That two-item advantage for human counselors is arguably the most psychologically meaningful part of the results. Mutual understanding and professional competence are precisely the qualities that define the therapeutic relationship in classical accounts of counseling, and the study suggests they remain the territory where human practitioners are perceived to excel, even among people who also use and generally rate AI support favorably. Conversely, the domains where AI scored higher map onto what the researchers describe as low-threshold and low-social-threat experiences. Talking to a chatbot appears to reduce the fear of being judged, makes deep self-disclosure feel safer, and lowers the barrier to seeking support in the first place. For people who find the prospect of sitting across from a professional intimidating, a machine that never sighs, never frowns, and never files a clinical note about your worst thoughts may feel like a genuinely safer opening move.

Attachment theory provided the study’s central individual-difference framework, and the regression results revealed a coherent pattern. Attachment anxiety, characterized by a heightened need for reassurance and fear of rejection, positively predicted perceived experience quality in both modalities, meaning anxiously attached participants rated both AI support and human counseling more favorably. Attachment avoidance, characterized by discomfort with closeness and interdependence, negatively predicted both outcomes, so avoidantly attached participants rated both forms of support less favorably. In other words, the same relational dispositions color how people experience support whether it comes from a person or a machine, suggesting that chatbots do not escape the attachment dynamics that shape human relationships. Notably, specialized mental health and therapy chatbot use was associated with higher AI-experience scores, while associations involving shorter AI-use duration were less consistent, hinting that purpose-built tools may deliver a better experience than general-purpose assistants pressed into therapeutic duty.

The mixed-effects sensitivity analysis added a subtler layer to the attachment findings. The interaction between modality and attachment avoidance was statistically significant, indicating that the relative gap between AI-based support and human counseling was not uniform across participants but varied according to how avoidant they were in close relationships. Although the study’s published summary does not detail the direction of this interaction, its existence implies that attachment avoidance may shape which modality feels more comfortable, a possibility with direct clinical implications for matching people to support formats. Meanwhile, the hierarchical regression on the human-minus-AI difference score showed improved model fit after the attachment variables were added, but the final full model was not significant at the omnibus level, so the researchers treat the difference-score analysis as suggestive rather than conclusive.

The authors are unusually explicit about the limits of their evidence, and these caveats deserve as much attention as the findings. The design was cross-sectional and entirely self-report, so the data capture perceived experiences rather than objectively measured therapeutic outcomes, and no causal claims about which modality actually helps more can be sustained. The distributions showed restricted upper-end variability, with medians generally high and often identical across modalities, which compresses the statistical room to detect differences. The AI tools participants reported using were heterogeneous, spanning whatever chatbots they happened to try, and the 11-item experience indicators were newly developed for this study and remain exploratory rather than validated instruments. The researchers also caution that the sensitivity analysis relied on a composite score derived from ordinal item ratings, a measurement choice that warrants care when interpreting the small mean difference that survived.

What the study ultimately delivers is a differentiated map rather than a verdict. Among dual users, perceived experiences differed by domain rather than showing a uniform preference for either machines or humans. AI-based psychological support appears to excel as a low-barrier, judgment-free channel that encourages disclosure, while human counseling retains perceived advantages in mutual understanding and professional competence, the relational core of formal therapy. For clinicians, policymakers, and the rapidly growing industry building mental health chatbots, the implication is that the two modalities may be complements rather than competitors: AI could serve as an accessible first step or an adjunct between sessions, while human professionals remain the reference point for depth and expertise. As conversational AI becomes more capable and more widely adopted, studies that compare experiences within the same individuals, as this one does, will be essential for understanding not just whether people use these tools, but how the tools make them feel, and for whom they work best.

Subject of Research: Perceived experiences of AI-based psychological support compared with human counseling among adults who have used both modalities

Article Title: Differences in perceived experiences of AI-based versus human psychological support among dual users: paired comparisons, hierarchical regression, and repeated-measures sensitivity analysis

Article References: Wu, H., Liu, Q., Shi, Y., Yang, M., Zhang, J., & Zhang, Y. (2026). Differences in perceived experiences of AI-based versus human psychological support among dual users: paired comparisons, hierarchical regression, and repeated-measures sensitivity analysis. BMC Psychology. https://doi.org/10.1186/s40359-026-05676-y

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05676-y

Keywords: conversational AI, mental health chatbots, human counseling, attachment theory, self-disclosure, user experience, paired comparison, hierarchical regression, mixed-effects model, psychological support, trust, BMC Psychology

The replacement question, the study suggests, treats psychological care as if it were a single activity that could be performed either by a person or by a machine. In reality, someone seeking help may need many different things at once: distress that is understood without being reduced too quickly to a label, practical support to change patterns of avoidance or sleep, a relationship in which speaking feels possible, time to remain with experiences that are not yet fully nameable, a fresh account of what has happened to them, and connections to schools, workplaces, welfare services, medical care, or community support. These forms of work overlap, but they are not identical, and AI enters each of them in different ways.

To capture this complexity, the study drew on a systematic analysis of the review literature. PubMed/MEDLINE and Scopus were searched for English-language reviews combining terms for mental health, AI technologies, and review research. After deduplication, 267 reviews met eligibility conditions and formed a candidate pool. Using predefined adequacy criteria, nine reviews were selected as the initial derivation corpus. From each, the researcher extracted five elements linked to a specific table, passage, or primary study: input, AI transformation, output, immediate recipient, and affected action. These operation anchors were then split into elementary relations, each representing one output changing one immediate action for one recipient, yielding 84 relations in the initial corpus.

The comparison of these relations, organized by the action that changed immediately after an AI output rather than by technology name, produced twelve first-order operation clusters. Ten of them reached a direct action by a service user or care provider, covering current-state detection, future prediction, information extraction and summarization, interpretive inquiry, care planning, structured therapeutic activity, responsive support, provider augmentation, self-management scaffolding, and access or referral coordination. Two clusters remained at the research boundary of pattern discovery and model development. Focused reviews were then added to test four contested boundaries, expanding the final map to 69 anchors and 98 relations without requiring any additional clusters.

From this evidence map, six clinical functions were derived: assessment and case formulation; structured intervention and change support; relational responsiveness; holding unresolved experience; narrative and meaning formation and revision; and contextual and institutional connection. Each function was tested by removal and merger procedures. Removing any single candidate left distinctive objects or failures unexplained, and merging adjacent candidates erased clinically important differences, for example between understanding and action, or between intervention and institutional connection. Leave-one-review-out checks confirmed that every function remained supported by relations from multiple reviews.

The functions reveal tensions that individual performance metrics obscure. Assessment may produce a score or diagnostic candidate that supports communication and access, yet a self-diagnosis formed from social media or chatbot responses can fix a label into a person’s self-understanding before adequate evaluation. Structured interventions can deliver psychoeducation and exercises at scale, but a behavioral activation prompt may support one person and burden another, and cognitive reframing may be inappropriate when the central problem is violence, discrimination, or unsafe work. Relational responsiveness requires that someone remains answerable for the effects of a response; a chatbot may generate fluent, empathic-sounding replies without carrying responsibility for a missed crisis, growing dependency, or follow-up after a break in contact.

Across these functions, the study proposes a cross-cutting risk it calls premature stabilization. AI does not work on distress directly; it transforms experience into scores, risk categories, intervention targets, conversation summaries, or service destinations. Such provisional representations are useful scaffolds for judgment and action, but they can be stored, repeated, and transferred across settings, gradually acquiring clinical and technical authority. The danger is not rapid judgment itself but the closure of alternative formulations, narratives, relations, or pathways of support before closure is clinically warranted. A workplace wellness application that identifies elevated anxiety and suggests breathing exercises and self-checks offers genuine relief, yet it may also frame a situation shaped by exhausting workloads, intimidating supervision, and job insecurity as purely a problem of anxiety management.

As a normative response, the study advances reconfigurative care, a principle requiring two things simultaneously: that judgments and actions needed now, including diagnosis, risk assessment, intervention, and referral, remain possible, and that the representations used in those judgments stay revisable in response to new information, the person’s disagreement, changes in relationship, and the outcomes of support. This openness is not unlimited ambiguity. A risk score may justify immediate safety action, and a diagnosis may provide recognition and access. The question is whether these forms of stabilization become final too early, and whether the first form in which a person becomes institutionally legible can be questioned and replaced by another account and pathway when necessary.

Human responsibility, on this account, has interpretive, temporal, and institutional dimensions. Interpretive responsibility means treating automated classifications and summaries as materials for inquiry rather than self-explanatory facts, reading them alongside a person’s history, body, relationships, culture, and living conditions. Temporal responsibility means protecting time in which uncertain experience can remain open, without delaying action when safety demands it. Institutional responsibility means connecting distress to appropriate health, welfare, school, workplace, or community support while specifying who remains accountable after handoff. Revision is incomplete if a changed interpretation cannot change the support pathway, and referral is not accountable if it ignores the person’s meaning and timing.

Because this responsibility cannot rest on individual vigilance alone, the study argues that organizations must specify who reviews AI outputs, where a service user’s disagreement is recorded, when a case returns to human reassessment, and who assumes responsibility after referral or escalation. Training should include the practice of contextualizing outputs and reconsidering them with the person. In crisis pathways, automated detection must connect to accountable human escalation rather than end as an isolated response. The author notes the analysis is limited to English-language reviews and does not estimate the accuracy or safety of individual technologies, and that the six functions remain an interpretive framework rather than independently validated empirical categories. Still, the central conclusion stands: the value of AI in mental health care should be judged not only by the performance of individual outputs but by the clinical work and responsibilities preserved across the entire care pathway.

Subject of Research: A conceptual framework of clinical functions and responsibility in AI-mediated psychological care

Article Title: Reconfigurative care and clinical responsibility in AI-mediated psychological care ecosystems

Article References: Minami, U. (2026). Reconfigurative care and clinical responsibility in AI-mediated psychological care ecosystems. SSM – Mental Health, 10, Article 100703. https://doi.org/10.1016/j.ssmmh.2026.100703

Image Credits: AI Generated

DOI: 10.1016/j.ssmmh.2026.100703

Keywords: artificial intelligence, mental health care, psychological care, chatbots, clinical responsibility, care pathways, digital therapeutics, risk prediction, case formulation, narrative meaning, reconfigurative care, care ecosystem