PsyWorld: Profile-Conditioned World Modeling for Human Behavior Prediction
Authors: Ziyuan Qin, Yang Qiao, Summer B. Frandsen, Michael T. Treadway, Liang Zhao
Organizations: Department of Computer Science, Emory University, Atlanta, Georgia, USA · Department of Psychiatry and Behavioral Sciences, Emory University, Atlanta, Georgia, USA
Understanding how stable psychometric traits and time-varying states jointly inform behavior is an important problem in computational behavioral science. Existing personalized agents commonly condition actions directly on a static profile, leaving realized state transitions unused as an explicit source of supervision. We introduce a profile-conditioned state world model that uses observed environment--behavior state transitions to learn transition-informed representations for behavioral prediction. During training, a posterior encoder quantizes realized transitions, while a prior predicts the resulting codes from the psychometric profile and pre-transition context. The prior code and predicted next state then condition macro-level behavior and simulator-grounded execution. Thus, the world model provides structured transition supervision instead of treating personality as a direct action descriptor. We evaluate the framework in VEET, a controlled behavioral simulator with human trajectories, and in AgentSociety as a structurally different synthetic environment. A participant-disjoint split of 240 training, 25 validation, and 48 test participants evaluates generalization to unseen people. On VEET, the model reaches 0.904 Top-1 accuracy (95% CI: [0.895, 0.913]) and 0.0033 Macro KL. Ablations associate continuous next-state modeling with most of the Top-1 gain, while discrete codes and prior alignment mainly improve Top-3 and distributional alignment. These results establish transition-informed world-model supervision as useful for subject-generalized behavior prediction, without implying causal psychological mechanisms or calibrated long-horizon state dynamics.
Figures & tables
Figure 1. A stable psychometric profile conditions latent transition codes together with the current environmental and behavioral state and macro action. The predicted next-state representation then informs subsequent behavior prediction. Dashed components use the realized future only during training; solid paths are available at rollout. A profile, current environment-behavior state, and current macro behavior enter a context encoder. A training-only posterior observes the realized state change and supplies a vector-quantized code, while a prior predicts a rollout code from pre-transition context. The code conditions a next-state decoder and a gated macro-behavior decoder.
Figure 2. Cross-environment qualitative diagnostics. Panel (a) compares human and agent trajectories for two held-out VEET participants. Panel (b) compares observed and gated VQ-WM macro-action distributions within AgentSociety-specific questionnaire-profile clusters; its compact labels C0–C3 correspond to A-C0–A-C3 in Table 4 . Here, n denotes the number of held-out macro-action events rather than unique participants. The panels visualize environment-specific outcomes and are not pooled. Panel (a) compares VEET human and agent trajectories. Panel (b) compares observed and predicted AgentSociety macro-action distributions across environment-specific questionnaire-profile clusters.
Figure 3. Exploratory associations between participant-level prior-code usage frequencies and seven psychometric scales. Cells report Spearman ρ ; ∗ denotes BH–FDR q<0.05 across 133 tests and † denotes uncorrected p<0.05 . No association survives FDR correction. The median pairwise sample size is 297, and codes are ordered by maximum absolute correlation. A diverging heatmap of Spearman correlations between 19 prior VQ-code usage frequencies and seven questionnaire scales. Correlations are small, and none survives false-discovery-rate correction.
Model
Top-1 ↑
Top-3 ↑
Macro KL ↓
Dist. Sim. ↑
VQ Codes ↑
VEET
Random
0.167
0.500
0.937
0.666
–
Majority
0.739
–
6.888
0.855
–
Direct SFT / BC
0.754
0.890
0.361
0.846
–
IRL policy
0.650
0.845
0.431
0.758
–
Ours
0.904 [0.895, 0.913]
0.989 [0.987, 0.990]
0.0033 [0.0023, 0.0046]
0.999 [0.998, 0.999]
19
Table 1. Main results under subject-level OOD evaluation on VEET and AgentSociety. The 240/25/48 train/validation/test participants are mutually exclusive, so the test set evaluates generalization to unseen people rather than new trajectories from people observed during training. Macro Top-1/Top-3 evaluate next macro-action prediction; Macro KL is DKL(Pdata(a)∥Ppred(a)) , and Dist. Sim. is the macro-action distribution similarity 1−JSD(Pdata,Ppred)/log2 . For Ours on VEET, the second line reports 95% participant-level bootstrap confidence intervals over all 48 test participants (1,000 resamples; seed 2026).
Variant
Top-1 ↑
Top-3 ↑
Macro KL ↓
Simple WM + decoder
0.901
0.952
0.012
VQ-WM
0.905
0.981
0.0078
VQ-WM + prior align.
0.904
0.989
0.0033
Table 2. VEET latent representation ablation.
Cluster
Profile tag
N
Top-1
Teleport R.
V-C0
Mid profile
7
0.910
0.18
V-C1
Elevated stress
18
0.913
0.13
V-C2
Higher stress / BDI
8
0.905
0.26
V-C3
High hedonic / low BDI
15
0.891
0.26
Table 3. Descriptive held-out VEET performance by environment-specific questionnaire-profile cluster.
Cluster
Profile tag
N
Top-1
Leisure R.
A-C0
Mid profile
13
0.558
0.16
A-C1
High hedonic / low BDI
14
0.606
0.27
A-C2
Elevated stress
10
0.600
0.25
A-C3
Higher stress / BDI
11
0.613
0.24
Table 4. Descriptive held-out AgentSociety performance by environment-specific questionnaire-profile cluster.
VEET
AgentSoc.
Metric
SFT/BC
DPO
SFT/BC
DPO
Completion ↑
0.168
0.332
0.664
0.990
Coin gain ↑
0.120
2.516
–
–
First-act. KL ↓
2.556
1.382
1.468
0.712
Move. gap ↓
0.873
0.520
0.453
0.267
Table 5. Micro-control results for SFT/BC and macro-conditioned DPO.
World models enable a predictive substrate for planning and action, yet existing formulations merely answer a physical question: what/where it is, and how will it evolve. Human behavior, however, is driven by hidden mental state (what a person believes, wants, intends, feels, and considers socially permissible), so a model that tracks the physical scene but not what each agent knows and believes about it predicts the wrong action for the right-looking scene. We formulate Mental World Modeling (MWM), a generic theoretical framework that makes mental variables core components of a world model rather than posthoc rationales: MWM aintains a coupled physical-mental world state, renders a target-specific partial observation, and simulates how candidate actions jointly update both components. We instantiate the framework in MENTIS, a training-free and fully inspectable baseline that decomposes the process into state parsing, target-observation generation, action decomposition, coupled physical and mental transition, and branch-level value evaluation. On a manually constructed, quality-controlled dataset of situated decision scenarios spanning text, image, and sounding-video stories, experiments with 8 modern LLM-based world models demonstrate that explicitly modeling the mental state is essential for predicting human decisions. Deeper analyses further expose the bottlenecks of current mental world modeling. We expect MWM as a next stage of world modeling, from simulating physical scenes to simulating the minds that act in them.
Hao Fei, Yiran Zhao
University of Oxford · National University of Singapore
World models enable large language model (LLM) agents to anticipate action effects and make informed decisions. For generalization, it is desirable for such world models to transfer efficiently to new environments with similar interaction structure and underlying dynamics. Recent approaches encode world dynamics as symbolic rules or executable programs, whose compositional structure in principle affords transfer, but still organize such knowledge around environment-specific objects. To address this limitation, we introduce RoleWorld, a symbolic world modeling approach which uses reusable prototypes and environment-specific parameter values as functional role abstractions to represent how objects behave in interactions. From interaction trajectories, a teacher LLM extracts role prototypes and executable transition rules defined over role instances, together with a skill that guides the actor LLM to use role information in decision making. In a target world, without the teacher, the actor uses this skill together with its current role assignments and parameter estimates, and actively tests a role instance hypothesis through interaction when its uncertainty affects the current decision. Feedback from each action refines role assignments and estimates parameter values online, with no LLM parameter updates and the transition rules held fixed. Experiments on MARS benchmark show that RoleWorld enables robust generalization across environments with significantly altered object behavior, suggesting that functional role abstractions can serve as an effective component for knowledge transfer in world modeling. Project Page: https://future-item.github.io/RoleWorld.
Boyuan Xiao, Pu Wang, Ji Feng +1
State Key Lab of CAD&CG Zhejiang University, China · Baiont Quant, China
Continuous multimodal sensing now allows human physiology to be observed throughout daily life rather than only during occasional clinical visits. However, most health artificial intelligence systems are designed to recognize current states, estimate risks or analyse individual biomarkers. They do not directly model how physiological states change in response to real-world events, behaviours, contexts and interventions. Here we propose the Physiological World Model (PWM), an event-conditioned framework for learning these changes at the level of the whole person. We introduce the HumanState Transition Token, a structured, quality-scored unit that connects the physiological state before an event with the event or action, relevant context and intervention information, the physiological trajectory after the event, observed outcomes and data quality. We describe four capability levels, from state representation to bounded intervention planning, together with four data acquisition and validation protocols. We also propose six benchmark tasks covering HumanState representation, forecasting across multiple timescales, individualized response prediction, simulation of alternative interventions, bounded planning and reliability under distribution shift. Together, this framework provides a practical path towards personalized health management, behavioural intervention design and clinician-supervised decision support, while clearly separating prediction from causal inference and making uncertainty, safety, governance and limits of use explicit.