Large language models (LLMs) offer emerging opportunities for psychological and behavioral research, but methodological guidance is lacking. This article provides a framework for using LLMs as psychological simulators across two primary applications: simulating roles and personas to explore diverse contexts, and serving as computational models to investigate cognitive processes. For simulation, we present methods for developing psychologically grounded personas that move beyond demographic categories, with strategies for validation against human data and use cases ranging from studying inaccessible populations to prototyping research instruments. For cognitive modeling, we synthesize emerging approaches for probing internal representations, methodological advances in causal interventions, and strategies for relating model behavior to human cognition. We address overarching challenges including prompt sensitivity, temporal limitations from training data cutoffs, and ethical considerations that extend beyond traditional human subjects review. Throughout, we emphasize the need for transparency about model capabilities and constraints. Together, this framework integrates emerging empirical evidence about LLM performance--including systematic biases, cultural limitations, and prompt brittleness--to help researchers wrangle these challenges and leverage the unique capabilities of LLMs in psychological research.
Gait encodes rich biometric and behavioural information, yet leveraging the manner of walking to infer psychological traits remains a challenging and underexplored problem. We introduce a hierarchical Multi-Stage Mixture of Movement Experts (MoME) architecture for multi-task prediction of psychological attributes from gait sequences represented as 2D poses. MoME processes the walking cycle in four stages of movement complexity, employing lightweight expert models to extract spatio-temporal features and task-specific gating modules to adaptively weight experts across traits and stages. Evaluated on the PsyMo benchmark covering 17 psychological traits, our method outperforms state-of-the-art gait analysis models, achieving a 37.47% weighted F1 score at the run level and 44.6% at the subject level. Our experiments show that integrating auxiliary tasks such as identity recognition, gender prediction, and BMI estimation further improves psychological trait estimation. Our findings demonstrate the viability of multi-task gait-based learning for psychological trait estimation and provide a foundation for future research on movement-informed psychological inference.
Traditional in-person psychological counseling remains primarily niche, often chosen by individuals with psychological issues, while online automated counseling offers a potential solution for those hesitant to seek help due to feelings of shame. Cognitive Behavioral Therapy (CBT) is an essential and widely used approach in psychological counseling. The advent of large language models (LLMs) and agent technology enables automatic CBT diagnosis and treatment. However, current LLM-based CBT systems use agents with a fixed structure, limiting their self-optimization capabilities, or providing hollow, unhelpful suggestions due to redundant response patterns. In this work, we utilize Quora-like and YiXinLi single-round consultation models to build a general agent framework that generates high-quality responses for single-turn psychological consultation scenarios. We use a bilingual dataset to evaluate the quality of single-response consultations generated by each framework. Then, we incorporate dynamic routing and supervisory mechanisms inspired by real psychological counseling to construct a CBT-oriented autonomous multi-agent framework, demonstrating its general applicability. Experimental results indicate that AutoCBT can provide higher-quality automated psychological counseling services.
There has been extensive prior work exploring how psychological factors such as anthropomorphism affect the adoption of Shared Autonomous Vehicles (SAVs). However, limited research has been conducted on how prompt strategies in large language models (LLM)-powered conversational SAV agents affect users' perceptions, experiences, and intentions to adopt such technology. In this work, we investigate how conversational SAV agents powered by LLMs drive these psychological factors, such as psychological ownership, the sense of possession a user may come to feel towards an entity or object they may not legally own. We designed four SAV agents with varying levels of anthropomorphic characteristics and psychological ownership triggers. Quantitative measures of psychological ownership, anthropomorphism, quality of service, disclosure tendency, sentiment of SAV responses, and overall acceptance were collected after participants interacted with each SAV. Qualitative feedback was also gathered regarding the experience of psychological ownership during the interactions. The results indicate that an SAV designed to be more anthropomorphic and to induce psychological ownership improved users' perceptions of the SAV's human-like qualities, and its responses were perceived as more positive but also more subjective compared to the control conditions. Qualitative findings support established routes to psychological ownership in the SAV context and suggest that the conversational agent's perceived performance may also influence psychological ownership. Both quantitative and qualitative outcomes highlight the importance of personalization in designing effective SAV interactions. These findings provide practical guidance for designing conversational SAV agents that enhance user experience and adoption.
Anton Vasiliuk, Irina Abdullaeva, Polina Druzhinina
et al.
Large language models (LLMs) hold the potential to absorb and reflect personality traits and attitudes specified by users. In our study, we investigated this potential using robust psychometric measures. We adapted the most studied test in psychological literature, namely Minnesota Multiphasic Personality Inventory (MMPI) and examined LLMs' behavior to identify traits. To asses the sensitivity of LLMs' prompts and psychological biases we created personality-oriented prompts, crafting a detailed set of personas that vary in trait intensity. This enables us to measure how well LLMs follow these roles. Our study introduces MindShift, a benchmark for evaluating LLMs' psychological adaptability. The results highlight a consistent improvement in LLMs' role perception, attributed to advancements in training datasets and alignment techniques. Additionally, we observe significant differences in responses to psychometric assessments across different model types and families, suggesting variability in their ability to emulate human-like personality traits. MindShift prompts and code for LLM evaluation will be publicly available.
Hiro Taiyo Hamada, Ippei Fujisawa, Genji Kawakita
et al.
Large Language Models (LLMs) such as ChatGPT have shown remarkable abilities in producing human-like text. However, it is unclear how accurately these models internalize concepts that shape human thought and behavior. Here, we developed a quantitative framework to assess concept alignment between LLMs and human psychological dimensions using 43 standardized psychological questionnaires, selected for their established validity in measuring distinct psychological constructs. Our method evaluates how accurately language models reconstruct and classify questionnaire items through pairwise similarity analysis. We compared resulting cluster structures with the original categorical labels using hierarchical clustering. A GPT-4 model achieved superior classification accuracy (66.2\%), significantly outperforming GPT-3.5 (55.9\%) and BERT (48.1\%), all exceeding random baseline performance (31.9\%). We also demonstrated that the estimated semantic similarity from GPT-4 is associated with Pearson's correlation coefficients of human responses in multiple psychological questionnaires. This framework provides a novel approach to evaluate the alignment of the human-LLM concept and identify potential representational biases. Our findings demonstrate that modern LLMs can approximate human psychological constructs with measurable accuracy, offering insights for developing more interpretable AI systems.
Leonie V. D. E. Vogelsmeier, Eduardo Oliveira, Kamila Misiejuk
et al.
Large language models (LLMs) offer the potential to simulate human-like responses and behaviors, creating new opportunities for psychological science. In the context of self-regulated learning (SRL), if LLMs can reliably simulate survey responses at scale and speed, they could be used to test intervention scenarios, refine theoretical models, augment sparse datasets, and represent hard-to-reach populations. However, the validity of LLM-generated survey responses remains uncertain, with limited research focused on SRL and existing studies beyond SRL yielding mixed results. Therefore, in this study, we examined LLM-generated responses to the 44-item Motivated Strategies for Learning Questionnaire (MSLQ; Pintrich \& De Groot, 1990), a widely used instrument assessing students' learning strategies and academic motivation. Particularly, we used the LLMs GPT-4o, Claude 3.7 Sonnet, Gemini 2 Flash, LLaMA 3.1-8B, and Mistral Large. We analyzed item distributions, the psychological network of the theoretical SRL dimensions, and psychometric validity based on the latent factor structure. Our results suggest that Gemini 2 Flash was the most promising LLM, showing considerable sampling variability and producing underlying dimensions and theoretical relationships that align with prior theory and empirical findings. At the same time, we observed discrepancies and limitations, underscoring both the potential and current constraints of using LLMs for simulating psychological survey data and applying it in educational contexts.
Elisabeth Valmyr Bania, Toril Sørheim Nilsen, Mikael Thastum
et al.
IntroductionSchool absenteeism represents a concern for students, educators, and parents alike. Teachers’ involvement is vital to students’ school life. Consequently, integrating schools and teachers effectively in absenteeism interventions is of great importance. However, few studies have investigated teachers’ perspectives on participating in manual-based, indicated interventions to promote school attendance. This study aimed to explore teachers’ experiences with the manual-based Back2School (B2S) intervention, which is based on cognitive behavioural therapy (CBT).MethodsSeven primary and lower secondary school teachers agreed to participate in individual interviews following their involvement in the intervention. These teachers engaged in various aspects of the intervention, including data collection, school sessions, and school meetings involving students, parents, and B2S group leaders.ResultsThe results indicate that some of the informants experienced increased competence and self-efficacy regarding school absenteeism following the intervention, while other informants did not have this experience.DiscussionThere is a need for more clarity and enhanced teacher involvement in future B2S interventions.
The study aims to explore the impact mechanism of brand ritual on online word-of-mouth communication, introducing the mediating variable—flow experience—and the moderating variable—consumer–brand relationship norms. The study uses the approach of the experimental research. In Experiment 1, with the watch as the experimental product and the advertisement as the online scene, 62 subjects in the pre-experiment and 132 subjects in the formal experiment are recruited to verify the main effect of brand ritual on online word-of-mouth communication. In Experiment 2, with the tea bag as the experimental product and the online press conference as the online scene, 73 subjects in the pre-experiment and 185 subjects in the formal experiment are recruited to verify the mediating role of flow experience in the impact of brand ritual on online word-of-mouth communication. In Experiment 3, with the scented candle as the experimental product and the promotional video of the e-commerce store as the online scene, 81 subjects in the pre-experiment and 269 subjects in the formal experiment are recruited to verify the moderating role of consumer–brand relationship norms in the impact of brand ritual on online word-of-mouth communication/flow experience. The results show that brand ritual is more effective in promoting online word-of-mouth communication than random action, flow experience plays a completely mediating role in the impact of brand ritual on online word-of-mouth communication, and consumer–brand relationship norms play a moderating role in the impact of brand ritual on online word-of-mouth communication/flow experience. The study not only reveals the impact mechanism of brand ritual on online word-of-mouth communication, but also provides strong guidance for companies to utilize brand ritual, flow experience, and consumer–brand relationship norms to promote online word-of-mouth communication.
The feeling of something belonging to someone is called "psychological ownership." A common assumption is that writing with generative AI lowers psychological ownership, but the extent to which this occurs and the role of prompt length are unclear. We report on two experiments to examine the relationship between psychological ownership and prompt length. Participants wrote short stories either completely by themselves or wrote prompts of varying lengths. Results show that when participants wrote longer prompts, they had higher levels of psychological ownership. Their comments suggest they thought more about their prompts, often adding more details about the plot. However, benefits plateaued when prompt length was 75-100% of the target story length. To encourage users to write longer prompts, we propose augmenting the prompt submission button so it must be held down a long time if the prompt is short. Results show that this technique is effective at increasing prompt length.
LLMs have revolutionized the landscape of information retrieval and knowledge dissemination. However, their application in specialized areas is often hindered by factual inaccuracies and hallucinations, especially in long-tail knowledge distributions. We explore the potential of retrieval-augmented generation (RAG) models for long-form question answering (LFQA) in a specialized knowledge domain. We present VedantaNY-10M, a dataset curated from extensive public discourses on the ancient Indian philosophy of Advaita Vedanta. We develop and benchmark a RAG model against a standard, non-RAG LLM, focusing on transcription, retrieval, and generation performance. Human evaluations by computational linguists and domain experts show that the RAG model significantly outperforms the standard model in producing factual and comprehensive responses having fewer hallucinations. In addition, a keyword-based hybrid retriever that emphasizes unique low-frequency terms further improves results. Our study provides insights into effectively integrating modern large language models with ancient knowledge systems. Project page with dataset and code: https://sites.google.com/view/vedantany-10m
Psychological research often involves understanding psychological constructs through conducting factor analysis on data collected by a questionnaire, which can comprise hundreds of questions. Without interactive systems for interpreting factor models, researchers are frequently exposed to subjectivity, potentially leading to misinterpretations or overlooked crucial information. This paper introduces FAVis, a novel interactive visualization tool designed to aid researchers in interpreting and evaluating factor analysis results. FAVis enhances the understanding of relationships between variables and factors by supporting multiple views for visualizing factor loadings and correlations, allowing users to analyze information from various perspectives. The primary feature of FAVis is to enable users to set optimal thresholds for factor loadings to balance clarity and information retention. FAVis also allows users to assign tags to variables, enhancing the understanding of factors by linking them to their associated psychological constructs. Our user study demonstrates the utility of FAVis in various tasks.
Although previous studies have investigated factors contributing to compulsive shopping, the specific role of alexithymia and its influence on emotional regulation in predicting this behavior remains underexplored. The study explores the link between alexithymia and compulsive shopping in young adults, focusing on whether emotional regulation difficulties predict problematic shopping behavior. A sample of 220 Italian young adults was assessed using the Toronto Alexithymia Scale (TAS-20) and the Shopping Behaviour Scale (SBS). Multiple regression analysis revealed that alexithymia's three dimensions explained 4.8% of the variance in compulsive shopping. Externally oriented thinking was the only significant predictor, while difficulty identifying and describing feelings were not. The findings suggest that individuals more focused on external realities are at higher risk for compulsive shopping. Improving emotional awareness and regulation may help reduce this behavior in young adults.
In psychoanalysis, generating interpretations to one's psychological state through visual creations is facing significant demands. The two main tasks of existing studies in the field of computer vision, sentiment/emotion classification and affective captioning, can hardly satisfy the requirement of psychological interpreting. To meet the demands for psychoanalysis, we introduce a challenging task, \textbf{V}isual \textbf{E}motion \textbf{I}nterpretation \textbf{T}ask (VEIT). VEIT requires AI to generate reasonable interpretations of creator's psychological state through visual creations. To support the task, we present a multimodal dataset termed SpyIn (\textbf{S}and\textbf{p}la\textbf{y} \textbf{In}terpretation Dataset), which is psychological theory supported and professional annotated. Dataset analysis illustrates that SpyIn is not only able to support VEIT, but also more challenging compared with other captioning datasets. Building on SpyIn, we conduct experiments of several image captioning method, and propose a visual-semantic combined model which obtains a SOTA result on SpyIn. The results indicate that VEIT is a more challenging task requiring scene graph information and psychological knowledge. Our work also show a promise for AI to analyze and explain inner world of humanity through visual creations.
Tujuan penelitian ini adalah untuk mengetahui pengaruh etika profesi auditor dan fee audit terhadap kualitas audit.
Desain / metodologi / pendekatan: dalam penelitian ini dilakukan analisis statistik deskriptif dengan pendekatan kuantitatif yang menggunakan teknik analisis regresi linear berganda dengan alat analisis SPSS 24.
Temuan Penelitian: Hasil dari penelitian ini menunjukkan bahwa etika profesi dan fee audit memiliki pengaruh terhadap kualitas audit.
Kontribusi Teoretis / Orisinalitas: Perbedaan penelitian ini dengan penelitian sebelumnya adalah pada teknik analisis yang digunakan, selain itu objek penelitian juga berbeda, pada penelitian ini yang menjadi objek penelitian adalah Kantor Akuntan Publik yang berada di Kota Pontianak dan Bandung dan struktur bisnis yang kompleks sehingga menjadikan penelitian layak untuk diteruskan. Berdasarkan permasalahan di atas, dan melihat pentingnya etika profesi serta sangat sensitifnya fee audit penulis tertarik untuk meneliti kembali dengan fokus KAP di Pontianak Bandung sebagai responden.
Keterbatasan dan implikasi penelitian: Peneliti menyadari keterbatasan dalam penelitian ini yang tentunya memerlukan perbaikan dan pengembangan untuk penelitian selanjutnya. Keterbatasan dalam penelitian ini adalah Variabel independen dalam penelitian belum memberikan kontribusi yang baik terhadap variabel dependen. Hal tersebut terlihat dari analisis koefisien determinasi dimana nilai R2 sebesar 66,6%. Sisanya sebesar 33.4% dipengaruhi oleh variabel lain diluar model ini sehingga disarankan bagi peneliti selanjutnya untuk menambahkan variabel-variabel independen yang secara teoritis dapat berpengaruh lebih besar terhadap kualitas audit. Selain itu data yang dikumpulkan untuk diteliti dan dianalisis berdasarkan pada persepsi masing-masing responden terhadap item-item instrumen penelitian sehingga dapat memungkinkan terjadinya bias atau miss perseption.
Economics as a science, Management. Industrial management
In early April the Department of Philosophy at Moscow State University for two days held the Second Mironov Readings, held in the memory of our colleague, corresponding member of the Russian Academy of Science Vladimir Vasilyevich Mironov, a long-time head of the Department. The speeches at the Conference touched upon a wide range of philosophical topics from exiting the Plato’s cage (by Professor Dagmar Mironova) to developing courseware for certain branches of philosophy. Professors of MGIMO University were represented at the Conference by Professor V. S. Glagolev and M. V. Silantyeva. The plenary report by B. I. Pruzhinin and T. G. Shchedrina from the editorial board of the Journal Voprosy Filosofii was followed by an energetic discussion of cultural problems and intellectual continuity. In the track devoted to the problems of culture, the reports by N. V. Kuznetsov (SPbU), A. M. Sokolov (SPbU), A. A. Krotov (MSU), and A. P. Zabiyako (Amur State University) stood out. M. O. Kedrova (MSU) deconstructed the text theory of Paul de Man, head of the Yale School of Criticism. M. A. Kobrinets (MSU) revealed the features of the interpretation of time in the philosophy of the theater of the second half of the 20th century. A brilliant analysis of the theories of historical memory was presented in the report by D. A. Anikina. These and other speeches presented the main directions of the development of domestic cultural knowledge. The influence of V. V. Mironov on this process, as well as on the whole process of the development of philosophy in our Fatherland, can hardly be overestimated.
Sören Dittmer, Michael Roberts, Julian Gilbey
et al.
In this perspective, we argue that despite the democratization of powerful tools for data science and machine learning over the last decade, developing the code for a trustworthy and effective data science system (DSS) is getting harder. Perverse incentives and a lack of widespread software engineering (SE) skills are among many root causes we identify that naturally give rise to the current systemic crisis in reproducibility of DSSs. We analyze why SE and building large complex systems is, in general, hard. Based on these insights, we identify how SE addresses those difficulties and how we can apply and generalize SE methods to construct DSSs that are fit for purpose. We advocate two key development philosophies, namely that one should incrementally grow -- not biphasically plan and build -- DSSs, and one should always employ two types of feedback loops during development: one which tests the code's correctness and another that evaluates the code's efficacy.
In 2017 a definition of spiral tilings was given, thereby answering a question posed by Grünbaum and Shephard in the late 1970s. The author had the pleasure to discuss the topic via e-mail with Branko Grünbaum in his 87th year. During this correspondence the question arose whether a spiral structure (given a certain definition of it) could be recognized automatically or whether "to some extent, at least, the spiral effect is psychological", as Grünbaum and Shephard had conjectured in 1987 (see exercise section of chapter 9.5 in "Tilings and Patterns"). In this paper, an algorithm for automatic detection of such a tiling's spiral structure and its first implementation results will be discussed. Finally, the definitions for several types of spiral tilings are refined based on this investigation.
Recently there have been numerous proposed solutions to the problem of logical omniscience in doxastic and epistemic logic. Though these solutions display an impressive breadth of subtlety and motivation, the crux of these approaches seems to have a common theme-minor revisions around the ubiquitous Kripke semantics-rooted approach. In addition, the psychological mechanisms at work in and around both belief and knowledge have been left largely untouched. In this paper, we cut straight to the core of the problem of logical omniscience, taking a psychologically-rooted approach, taking as bedrock the "quanta" of given percepts, qualia and cognitions, terming our approach "PQG logic", short for percept, qualia, cognition logic. Building atop these quanta, we reach a novel semantics of belief, knowledge, in addition to a semantics for psychological necessity and possibility. With these notions we are well-equipped to not only address the problem of logical omniscience but to more deeply investigate the psychical-logical nature of belief and knowledge.