PsyCLIENT: Client Simulation via Conversational Trajectory Modeling for Trainee Practice and Model Evaluation in Mental Health Counseling

Huachuan Qiu; Zhaoming Chen; Yuqian Chen; Yuan Xie; Yu Lu; Zhenzhong Lan

cs.CLJan 12, 2026

PsyCLIENT: Client Simulation via Conversational Trajectory Modeling for Trainee Practice and Model Evaluation in Mental Health Counseling

Authors: Huachuan Qiu, Zhaoming Chen, Yuqian Chen, Yuan Xie, Yu Lu, Zhenzhong Lan

Organizations: Westlake University · The University of Utah

Abstract

LLM-based client simulation provides a scalable approach to novice counselor training, counseling-dialogue synthesis, and interactive evaluation of automated counseling systems. However, existing approaches are limited by insufficient profile diversity, weak behavioral grounding, and the lack of open Chinese-language resources for simulated counseling clients. We propose PsyCLIENT, a framework that conditions simulated-client responses on client profiles, dialogue histories, and conversational trajectories specifying target behaviors and content constraints at each client turn. We also construct PsyCLIENT-CP, a dataset of 120 Chinese client profiles spanning 60 counseling topics. Evaluations involving 24 professional counselors show that PsyCLIENT receives higher ratings of perceived authenticity and training utility than the comparison methods. In a separate source-identification study, PsyCLIENT dialogues were more frequently misclassified as human-client interactions than those generated by the baselines. These results suggest that conversational trajectory modeling can transform static client profiles into behavior-guided, dynamically unfolding counseling interactions.

Figures & tables

Figure 1: Illustration of our automatic client simulation via conversational trajectory modeling. Step 1 is dataset construction. Step 2 is conversational trajectory extraction. There are 12 types of client behavior labels ( Li et al., 2023 ) , including "pi: Providing Information", "co: Confirming", "ec: Expressing Confusion", "de: Defending", etc. (see Figure 15 in Appendix C.3 ). For a detailed illustration, see Figure 4 in Appendix C . Step 3 is the mechanism of client simulation via conversational trajectory modeling. Each turn in a trajectory contains behavior labels and reference content spoken by the human client.

Figure 2: Distribution of topic annotations across client profiles in PsyCLIENT-CP. Values in parentheses indicate the number of profiles associated with each topic; a profile may contain multiple topics.

Client	Fluency ( $\uparrow$ )	Emotion ( $\uparrow$ )	Coherence ( $\uparrow$ )	Contextual Appropriateness ( $\uparrow$ )	Overall Authenticity ( $\uparrow$ )
AnnaAgent	6.00	6.25	4.75	5.25	4.75
ConsistentMI	4.50	5.00	3.50	3.75	3.50
Patient- $\Psi$	6.00	4.75	5.25	5.25	4.25
Roleplay-doh	4.75	5.00	4.50	5.00	4.75
PsyCLIENT	6.25	6.25	5.75	6.00	6.00

Table 1: Authenticity evaluations from four professional counselors.

Client	Listening ( $\uparrow$ )	Questioning ( $\uparrow$ )	Emotion-handling ( $\uparrow$ )	Technique-practice ( $\uparrow$ )	Recommendation ( $\uparrow$ )
AnnaAgent	5.25	5.00	5.75	5.00	5.25
ConsistentMI	4.25	4.00	3.75	3.75	3.50
Patient- $\Psi$	4.25	4.75	5.00	4.00	4.50
Roleplay-doh	5.50	5.00	5.25	4.50	4.50
PsyCLIENT	5.75	5.75	5.75	6.00	6.00

Table 2: Expert-rated training utility of different simulated-client methods.

Client	Fluency ( $\uparrow$ )	Emotion ( $\uparrow$ )	Coherence ( $\uparrow$ )	Contextual Appropriateness ( $\uparrow$ )	Overall Authenticity ( $\uparrow$ )
profile	4.37 1.07 **	4.57 1.25 **	4.61 1.13 **	4.59 1.17 **	4.50 1.39 **
+content	4.96 1.11 **	4.98 0.83 **	4.76 0.90 **	4.96 0.89 **	4.82 1.06 **
+behavior	5.32 0.87	5.07 1.12 **	5.09 1.22	5.16 1.09	5.11 1.11 *
PsyCLIENT	5.69 0.75	5.80 0.83	5.38 0.95	5.56 0.91	5.64 0.79

Table 3: Authenticity evaluations from professional counselors. Statistically significant differences between each variant and PsyCLIENT are marked with

{*}

(

p<0.05

) and

{**}

(

p<0.01

), as determined using the Mann-Whitney U test ( Mann and Whitney, 1947 ) . The best results are highlighted in bold. The second-best results are highlighted with an underline.

Client	Listening ( $\uparrow$ )	Questioning ( $\uparrow$ )	Emotion-handling ( $\uparrow$ )	Technique-practice ( $\uparrow$ )	Recommendation ( $\uparrow$ )
profile	4.89 1.09 *	4.95 1.79 **	4.65 1.09 **	4.63 1.22 **	4.61 1.11 **
+content	5.00 0.84 *	5.07 0.93 **	4.84 1.21 *	4.96 0.99 **	4.93 1.00 **
+behavior	5.18 0.98	5.30 1.06	5.16 1.06	5.20 1.16	5.11 1.23 *
PsyCLIENT	5.42 0.80	5.78 0.79	5.47 0.78	5.69 0.81	5.69 0.84

Table 4: Training utility evaluations from professional counselors. Statistically significant differences between each variant and PsyCLIENT are marked with

{*}

(

p<0.05

) and

{**}

(

p<0.01

). The best results are highlighted in bold.

Client	Number of Accurate Discriminations
profile	10
+content	10
+behavior	10
PsyCLIENT	8
AnnaAgent	10
ConsistentMI	10

Table 5: Number of accurate discriminations for each client type.

Judge	Human-Client		profile		+content		+behavior		PsyCLIENT
Judge	A	B	A	B	A	B	A	B	A	B
Qwen3.6-max-preview ( Team, 2025 )	43	44	45	45	42	43	44	41	40	37
kimi-k2-0905-preview ( Team et al., 2025 )	39	45	41	32	38	24	41	22	40	7
MiniMax-M2.7 ( Chen et al., 2026 )	32	45	32	5	32	5	27	8	22	3
deepseek-v4-flash ( Xu et al., 2026 )	34	38	45	19	42	25	41	22	38	16
deepseek-v4-pro ( Xu et al., 2026 )	42	41	37	41	36	32	32	36	32	31

Table 6: Source-discrimination results of LLM-based evaluation on 30-turn dialogue segments. Each value denotes the number of correctly discriminated cases out of 45. Lower values indicate that the dialogue source is harder for the LLM judge to identify. “A” and “B” denote the two prompt versions with reversed label orders, respectively. For details, see Figure 35 in Appendix E.1 .

Figure 3: Discrimination rate of human expert evaluation.

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 4: The detailed illustration of our automatic client simulation framework via conversational trajectory modeling.

Figure 16: Client responses conditioned with different client behavior labels. Client responses with resistance are highlighted with pink color.

Figure 35: The prompts for LLM evaluation.

Large Language Model-Based Role-Play Simulation

Large language model-based role-play simulation creates interactive agents that portray people, characters, or users for training, testing, and research.

See all 106 papers

PatientAct: Theory-Grounded Mental Health Client Simulation
Aug 13, 2026Sahand Sabour, TszYam NG, Yaqian Chen +3Mental Health Counseling User Simulation
The CoAI Group, DCST, Institute for Artificial Intelligence, Tsinghua University, Beijing, China · Department of Psychology, Beijing Normal University, Beijing, China · Counseling and Psychological Development Guidance Center, Tsinghua University, Beijing, China
Beyond Compliance: A Resistance-Informed Motivation Reasoning Framework for Challenging Psychological Client Simulation
Apr 12, 2026Danni Liu, Bo Liu, Yuxin Hu +5Large Language Model-Based Role-Play Simulation
Southeast University, Nanjing, China · Purple Mountain Laboratories, Nanjing, China · The Nanjing Derong Wisdom Information Technology Co., Ltd., Nanjing, China
MASC: A Multi-Agent Self-Calibration Framework with Latent Construct Alignment for Consistent Client Role-Playing in Psychological Counseling
Oct 6, 2026Shixin Peng, Kun Jiang, Jiaxing Zheng +2Large Language Model-Based Role-Play Simulation Multi-Agent LLM Systems
Central China Normal University, Wuhan, China · Shanghai Jiao Tong University, Shanghai, China

Aug 13, 2026J/K move · Enter open · S save

PatientAct: Theory-Grounded Mental Health Client Simulation

Sahand Sabour, TszYam NG, Yaqian Chen, Guanqun Bi+2

The CoAI Group, DCST, Institute for Artificial Intelligence, Tsinghua University, Beijing, China · Department of Psychology, Beijing Normal University, Beijing, China · Counseling and Psychological Development Guidance Center, Tsinghua University, Beijing, China

Mental Health Counseling User Simulation

Open paper AskPDF arXiv

LLM-based simulated clients are increasingly used to train novice counselors, evaluate LLM therapists, and generate synthetic data. However, current simulators produce overly cooperative clients that disclose too readily, accept therapeutic reframes without resistance, and resolve core issues within a single session. We trace these issues to profiles that lack causal depth and behavioral mechanisms that treat all content as equally accessible. We present PatientAct, a framework for client simulation grounded in established clinical theories. Our profiles integrate the 5Ps clinical case formulation, providing causal depth without tying the design to any single therapeutic modality. During simulation, profiles include a dynamic memory layer in which items carry trust thresholds (e.g., symptoms are available early, whereas formative memories require a sustained therapeutic alliance). At each turn, the client's emotional reaction and behavior are modeled before generating a response. If the therapist approaches gated content, PatientAct expresses resistance in terms of quantity, content, and style rather than defaulting to cooperation or a single resistance pattern. We evaluate our framework on 40 clinical situations and demonstrate that it generates diverse profiles with high clinical plausibility. Moreover, PatientAct significantly outperforms the baselines, yielding substantial gains in resistance quality and behavioral realism. Our code and data are publicly available via github.com/Sahandfer/PatientHub.

Abstract

Large Language Model-Based Role-Play Simulation

Explore similar work

PatientAct: Theory-Grounded Mental Health Client Simulation

Beyond Compliance: A Resistance-Informed Motivation Reasoning Framework for Challenging Psychological Client Simulation

MASC: A Multi-Agent Self-Calibration Framework with Latent Construct Alignment for Consistent Client Role-Playing in Psychological Counseling