psychological theory integration

Psychological theories have long suffered from a lack of formal modeling tools, hindering empirical validation, predictive accuracy, and interdisciplinary collaboration. To address this, this paper introduces automata theory—specifically finite-state automata (FSA)—as a systematic formal framework for psychological modeling, using the Lazarus-Folkman stress-coping theory as a canonical case study. The proposed approach yields a rigorously defined, executable semantic model that enables precise theoretical articulation, modular decomposition, and cross-theoretical comparability. The resulting FSA model supports automated consistency checking, dynamic behavioral simulation, and integrative analysis across multiple theories. This work bridges a critical gap in psychology by establishing the first systematic methodology for formal, computable modeling of psychological theories. It advances a new paradigm wherein theories become verifiable, reusable, and interoperable—thus promoting cumulative, reproducible, and computationally grounded scientific progress in psychological science.

Large language models (LLMs) are increasingly deployed in psychological research—as tools, targets of assessment, and cognitive models—yet recent evidence reveals severe measurement unreliability: factor structures of personality traits collapse, moral judgments reverse with minor punctuation changes, and theory-of-mind performance fluctuates dramatically under syntactic rephrasing. These “measurement ghosts” reflect statistical artifacts rather than substantive phenomena, threatening construct validity. Method: We propose the first validity-driven, six-stage workflow integrating psychometric principles and causal inference frameworks, dynamically calibrating validation rigor to research objectives and systematically governing the entire LLM psychology research lifecycle. Our approach includes construct validity verification, computational confound control, modeling of non-independent observations, and transparent experimental design. Contribution/Results: Applied to assessing “LLM selfhood,” our framework successfully disentangles genuine computational phenomena from measurement artifacts, establishing a reproducible empirical paradigm for AI psychology.

Current human behavior modeling in educational and training contexts suffers from insufficient psychological fidelity and limited interpretability. To address this, we propose a multi-agent psychological simulation system featuring a novel “Inner Council” mechanism—comprising heterogeneous agents representing core psychological constructs (e.g., self-efficacy, mental models, social construction) that collaboratively generate transparent, theory-grounded decisions. The system integrates established learning theories—including social learning theory and cognitive apprenticeship—to support high-fidelity, traceable behavioral generation. Empirical evaluation demonstrates significant improvements in simulating higher-order learning mechanisms—such as metacognition and deliberate practice—in teacher training and psychological research. Crucially, our approach overcomes key limitations of black-box neural network models by ensuring psychological coherence and behavioral interpretability, thereby advancing the rigor and applicability of computational models in educational psychology.

Current LLM agents lack quantifiable, psychometrically validated personality models that ensure human-comparable behavioral responses in social contexts. Method: We propose a four-stage modeling framework grounded in the Big Five Inventory (BFI), integrating semantic space alignment, scale embedding, empirical data fine-tuning, and simulated behavior generation—marking the first application of rigorous psychometric validation to LLM personality modeling. Contribution/Results: Our approach achieves high semantic consistency (0.92) in personality representation, strong correspondence with human scale responses (r = 0.87), and successfully replicates canonical associations between personality traits and risk/ethical decision-making. Critically, it establishes bidirectional interpretability between latent trait scores and observable behavioral outputs. This work advances LLMs as controllable, reproducible, and empirically verifiable tools for social science experimentation.

This study addresses the dual challenges of data collection bottlenecks and the lack of automation in theoretical discovery within psychological research by introducing the first end-to-end computational cognitive science framework. The proposed system employs a nested multi-agent architecture: an inner loop automatically constructs, fits, and critiques probabilistic cognitive models, while an outer loop autonomously designs and executes crowdsourced experiments to validate these models, thereby establishing a closed-loop “hypothesize–experiment–optimize” cycle. Integrating Bayesian model comparison with automated experimental design, the framework successfully recovers established theories on synthetic data and, through three rounds of human experimentation, discovers novel models that outperform existing ones in the literature. These results demonstrate a significant improvement in both the efficiency and accuracy of theoretical discovery, establishing the feasibility of automating scientific exploration in psychology.

Cognitive science has long relied on manually constructed theories, impeding automated scientific discovery. To address this limitation, this work proposes AutoCog—the first fully automated cognitive scientist system—leveraging large language model agents to close the loop from generating executable theories and designing discriminative experiments, to collecting human behavioral data and iteratively evaluating and refining theories. AutoCog represents the first framework in cognitive science to render theory construction explicit, executable, and cumulative. The system not only reproduces both established and unconventional decision-making strategies but also discovers and validates, through a preregistered experiment, a novel multi-cue decision theory that outperforms the initial model.

研究使用AutoCog系统通过模拟数据发现理论，并验证这些理论能否泛化到人类行为，结果显示模拟中发现的理论在人类数据上表现良好。

Current research on Machine Theory of Mind lacks a rigorous formal definition and a unified evaluation framework. This work addresses this gap by integrating empirical principles from cognitive psychology, neuroscience, and artificial intelligence to propose the first formal definition of Machine Theory of Mind. It introduces a metamodel that synthesizes cognitive modeling, formal logic, and systems analysis, accompanied by a corresponding empirical benchmarking framework. By establishing a clear theoretical foundation, a unified modeling paradigm, and an actionable evaluation methodology, this study advances the field toward systematicity and empirical verifiability.

本文通过实现目标导向情感理论的高保真计算模型，解决了情感计算中描述性模型与信号驱动架构之间的矛盾，展示了复杂情感特征如何从简单交互中自然产生。

This study addresses the lack of a rigorous evaluation framework for assessing whether foundation models can serve as explanatory models of human cognition and development, noting that behavioral alignment alone is insufficient to establish cognitive validity. The authors propose a four-stage inferential framework—task adaptation, construction of linking hypotheses, evaluation of behavioral correspondence, and systematic model comparison—that emphasizes the central role of linking hypotheses in mapping model outputs to human behavior. They argue that mere behavioral fit is inadequate; instead, cognitive plausibility requires integration of theoretical commitments, diagnostic tasks, and comparative experiments. By providing a methodological foundation for applying foundation models in cognitive and developmental science, this work substantially enhances their scientific validity as explanatory tools.