OHOle Hätscher

Ole Hätscher

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ole Hätscher has a total of 86 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Large language model agent simulations of individual differences in social interactions

    by Ole Hätscher, Leon Pielage, Pascal Kockwelp, Simon Mats Breil, Anna Junga, Jennifer Dabel, Benjamin Risse, Bernhard Marschall, Mitja D Back

    A key challenge in the behavioral sciences is the lack of formal models that capture individual differences in social interactions. At the same time, recent advances in artificial intelligence offer new opportunities to simulate and assess interpersonal behavior at scale. In this study, we introduce an agent-based modeling framework that uses large language models (LLMs) to simulate short-term social interactions. Within a dyadic interaction cycle, an LLM-based agent continually adjusts its interpersonal behavior (agency and communion) to its prior behavior, the partner’s behavior, and the combined influence of behavioral defaults (differences in default behavioral tendencies) and interpersonal reactivities (differences in reactions to partner behaviors). We derive a mathematical model that formalizes how interpersonal behavior unfolds across the interaction, evaluate the model in simulations, illustrate how the framework expresses these interpersonal dynamics in natural language, and outline the framework’s flexibility for simulating individual differences in practical applications across the behavioral sciences.

  • Dynamic Personality Adaptation in Large Language Models via State Machines

    by Leon Pielage, Ole Hätscher, Mitja Back, Bernhard Marschall, Benjamin Risse

  • Open Access

    Artificial Intelligence and Individual Differences: A Perspective of the Differential Psychology, Personality Psychology, and Psychological Assessment Division of the German Psychological Society (DGPs)

    by Ramona Schoedel, Elisa Altgassen, Mitja D Back, Martje Buss, Markus Bühner, Julian M. Etzel, Benjamin Goecke, Sandra Grinschgl, Eric Grunenberg, Ole Hätscher, et al.

    In this paper, we examine artificial intelligence (AI) in the context of studying and measuring individual differences across three current topics: AI as a research tool, the design of human-AI interaction, and AI-assisted assessment. Drawing on selected examples, we illustrate the wide-ranging potentials of AI at the conceptual and methodological level, while also showing that its application raises new (research) questions - for instance regarding human competence in dealing with AI, the “personality” of AI systems, or psychometric standards in the use of AI-assisted assessment procedures. Against this backdrop, we discuss key prerequisites for a sustainable integration of AI into both research in personality psychology and individual differences, including critical validation, transparency, interdisciplinary collaboration, and the systematic consideration of individual differences in the responsible design of the human-AI interface.

  • Open Access

    Who stands out? Defining, measuring, and uncovering the elements of unusualness

    by Ole Hätscher, Yves Breitmoser, Marie I Kaiser, Johannes L Klinz, Lara Kroencke, Niclas Kuper, René Mõttus, Lisa M Niemeyer, Julian Scharbert, Uku Vainik, et al.

    Perhaps surprisingly, personality science has not yet adequately addressed a very basic aspect of individuality: The extent to which an individual differs from others, a phenomenon we refer to as unusualness. In this work, we provide a framework for defining, measuring, and uncovering the elements of unusualness. First, we define unusualness as the extent to which an individual’s personality deviates from that of others in the population, either through rare expression of individual variables (unusual values) or through atypical co-occurrence of multiple variables (unusual combinations), and distinguish it from related concepts in personality research. Second, we introduce and compare various methods to measure unusualness in a simulation study and demonstrate how reliability and validity of unusualness measures can be established. Third, we propose methods to uncover the elements of unusualness (i.e., unusual values, unusual combinations) and validate these methods through simulation. Fourth, we illustrate our framework using three empirical datasets (combined N = 1,834) and multiple assessment methods, showing that unusualness can be reliably measured and offering first insights into its generalizability, nomological network, incremental validity, and possible explanatory factors. Finally, we outline conceptual, methodological, and empirical directions for future research on unusualness.

  • Using machine learning to predict individual differences in psychological reactivities to social interactions.

    by Ole Hätscher, Johannes L. Klinz, Niclas Kuper, Lara Kroencke, Julian Scharbert, Eric Grunenberg, Mitja D. Back

    Individual differences in psychological reactivities (i.e., the degree to which individuals react differently to social interactions) are central to psychological research. Previous theory-based research has identified substantial individual differences in reactivities but few robust predictors of these differences. This work aimed to address two questions: First, can individual differences in reactivities to social interactions be accurately predicted at all? Second, what are the most important person-level variables for this prediction? A data-driven machine learning approach was applied to three large-scale experience sampling data sets (overall N = 5,047) to predict the extent to which individuals reacted with positive and negative affect to momentary social interaction characteristics (e.g., interaction depth). Individual differences in reactivities were extracted via multilevel modeling (i.e., random slopes) and then predicted with machine learning methods using a variety of person-level variables (i.e., sociodemographics, personality traits, and political and societal attitudes). The robustness of predictions was examined by built-in cross-validation and across independent samples. Feature importance and interactions were analyzed with SHapley Additive exPlanations values. Our results suggest that, whereas complex prediction models outperformed a baseline model in predicting individual differences in reactivities in most analyses, the overall predictive performance was limited. This finding underlines the importance of replicating machine learning results across outcomes and independent samples. We revealed several predictive patterns that can stimulate future research, elaborate on limitations of current machine learning approaches for intensive within-person data, and discuss the results against the background of dynamic conceptualizations of personality. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

  • Open Access

    Teaching Personalized Doctor-Patient Communication with AI: PerTRAIN – A Prototype for Interpersonally Responsive Virtual Patients in Medical Education

    by Anna Junga, Ole Hätscher, Jennifer Dabel, Gabriyel Mado, Alberta Ajani, Leon Pielage, Pascal Kockwelp, Simon Mats Breil, Helena Baur, Jan Siebenbrock, et al.

    Personalized medicine requires physicians to adapt clinical communication and decision-making to patients’ individual motives, emotions, and interpersonal behaviors. However, training these skills remains challenging, as established simulation-based formats—particularly actor-based simulations—are resource-intensive and difficult to scale. Consequently, there is a need for scalable and flexible training approaches that allow repeated practice across diverse patient personalities and clinical contexts.The PerTRAIN (Personalization Training in Medicine) project addresses this need by leveraging large language models (LLMs) to simulate virtual patients with dynamically adapting personality expression at scale. Grounded in Contemporary Integrative Interpersonal Theory (CIIT), patient behavior is modelled along the dimensions of agency and communion and updated in response to the medical trainee’s behavior, enabling systematic variation and dynamic adaptation of interpersonal behavior within medical scenarios. This allows trainees to learn how patient personality shapes communication, and to practice adaptive, patient-centered communication strategies and appropriate clinical decision-making.For the personalization trainings, we developed clinical cases in which personality expression substantially influences behavior in doctor–patient interactions, that represent common encounters in primary care, and that align with national guidelines for patient-centered care. An initial chat-based implementation enables structured interactions, dynamic personality adaptation, and iterative refinement. The system is designed as a complementary tool to existing simulation formats, offering a scalable, low-threshold environment for repeated practice and reflection. A first application in curricular teaching is scheduled for 2026. Future extensions of the framework are discussed, such as large-scale empirical validation, modelling long-term interpersonal trajectories, and the extension of interactions to multimodal formats.

  • Open Access

    Who Stands Out? Defining, Measuring, and Uncovering the Elements of Unusualness

    by Ole Hätscher, Yves Breitmoser, Marie I. Kaiser, Johannes Leonhard Klinz, Lara Kroencke, Niclas Kuper, René Mõttus, Lisa Maria Niemeyer, Julian Scharbert, Uku Vainik, et al.

    Perhaps surprisingly, personality science has not yet adequately addressed a very basic aspect of individuality: The extent to which an individual differs from others, a phenomenon we refer to as unusualness. In this work, we provide a framework for defining, measuring, and uncovering the elements of unusualness. First, we define unusualness as the extent to which an individual’s personality deviates from that of others in the population, either through rare expression of individual variables (unusual values) or through atypical co-occurrence of multiple variables (unusual combinations), and distinguish it from related concepts in personality research. Second, we introduce and compare various methods to measure unusualness in a simulation study and demonstrate how reliability and validity of unusualness measures can be established. Third, we propose methods to uncover the elements of unusualness (i.e., unusual values, unusual combinations) and validate these methods through simulation. Fourth, we illustrate our framework using three empirical datasets (combined N = 1,834) and multiple assessment methods, showing that unusualness can be reliably measured and offering first insights into its generalizability, nomological network, incremental validity, and possible explanatory factors. Finally, we outline conceptual, methodological, and empirical directions for future research on unusualness.

  • Open Access

    Teaching Personalized Doctor-Patient Communication with AI: PerTRAIN – a Prototype for Interpersonally Responsive Virtual Patients in Medical Education

    by Ole Hätscher, Jennifer Dabel, Gabriyel Mado, Alberta Ajani, Leon Pielage, Pascal Kockwelp, Simon M. Breil, Helena Baur, Jan Siebenbrock, Benjamin Risse, et al.

    Personalized medicine requires physicians to adapt clinical communication and decision-making to patients’ individual motives, emotions, and interpersonal behaviors. However, training these skills remains challenging, as established simulation-based formats—particularly actor-based simulations—are resource-intensive and difficult to scale. Consequently, there is a need for scalable and flexible training approaches that allow repeated practice across diverse patient personalities and clinical contexts. The PerTRAIN (Personalization Training in Medicine) project addresses this need by leveraging large language models (LLMs) to simulate virtual patients with dynamically adapting personality expression at scale. Grounded in Contemporary Integrative Interpersonal Theory (CIIT), patient behavior is modelled along the dimensions of agency and communion and updated in response to the medical trainee’s behavior, enabling systematic variation and dynamic adaptation of interpersonal behavior within medical scenarios. This allows trainees to learn how a patients personality shapes communication, and to practice adaptive, patient-centered communication strategies and appropriate clinical decision-making. For the personalization trainings we developed clinical cases in which personality expression substantially influences behavior in doctor–patient interactions, that represent common encounters in primary care, and that align with national guidelines for patient-centered care. An initial chat-based implementation enables structured interactions, dynamic personality adaptation, and iterative refinement. The system is designed as a complementary tool to existing simulation formats, offering a scalable, low-threshold environment for repeated practice and reflection. A first application in curricular teaching is scheduled for 2026. Future extensions of the framework are discussed, such as large-scale empirical validation, modelling long-term interpersonal trajectories, and the extension of interactions to multimodal formats.


  • Open Access

    LLM-Based Agent Simulations for Modeling Individual Differences in Social Interactions (LASSI)

    by Ole Hätscher, Leon Pielage, Pascal Kockwelp, Simon Mats Breil, Anna Junga, Jennifer Dabel, Benjamin Risse, Bernhard Marschall, Mitja D Back

    A key challenge in psychological science is the lack of formal models that capture individual differences in social interactions. At the same time, recent advances in artificial intelligence offer unprecedented opportunities to simulate and assess interpersonal behavior at scale. In this work, we bring these developments together by introducing LASSI, an agent-based modeling framework based on Contemporary Integrative Interpersonal Theory (CIIT) that uses large language models (LLMs) to simulate social interactions. Within a dyadic interaction cycle, an LLM-based agent continually adjusts its interpersonal behavior (i.e., agency and communion) in response to the perceived behavior of its interaction partner (either a human or another LLM-based agent). Technically, an agent’s interpersonal behavior is governed by a finite-state Markov model within a discretized agency-communion space. The agent’s state distribution is iteratively updated as a function of its prior state distribution, the interaction partner’s behavior, and the combined influence of complementarity (the tendency of communal behavior to evoke communal behavior and of agentic behavior to evoke less agentic behavior), behavioral defaults (individual differences in general behavioral tendencies), and interpersonal reactivities (individual differences in the strength of reactions to defined partner behaviors). We show how LASSI can be used to simulate, assess, and experimentally manipulate interpersonal dynamics, derive a mathematical model that formalizes how key interpersonal phenomena jointly shape interpersonal behavior, and demonstrate the framework’s flexibility for advancing theories of interpersonal behavior and supporting practical applications within and across psychological science.

  • Open Access

    Using Machine Learning to Predict Individual Differences in Psychological Reactivities to Social Interactions

    by Ole Hätscher, Johannes Leonhard Klinz, Niclas Kuper, Lara Kroencke, Julian Scharbert, Eric Grunenberg, Mitja D Back

    Individual differences in psychological reactivities (i.e., the degree to which individuals react differently to social interactions) are central to psychological research. Previous theory-based research has identified substantial individual differences in reactivities but few robust predictors of these differences. This work aimed to address two questions: First, can individual differences in reactivities to social interactions be accurately predicted at all? Second, what are the most important person-level variables for this prediction? A data-driven machine learning approach was applied to three large-scale experience-sampling data sets (overall N = 5,047) to predict the extent to which individuals reacted with positive and negative affect to momentary social interaction characteristics (e.g., interaction depth). Individual differences in reactivities were extracted via multilevel modeling (i.e., random slopes) and then predicted with machine learning methods using a variety of person-level variables (i.e., socio-demographics, personality traits, political and societal attitudes). The robustness of predictions was examined by built-in cross-validation and across independent samples. Feature importance and interactions were analyzed with SHapley Additive exPlanations values. Our results suggest that, whereas complex prediction models outperformed a baseline model in predicting individual differences in reactivities in most analyses, the overall predictive performance was limited. This finding underlines the importance of replicating machine learning results across outcomes and independent samples. We revealed several predictive patterns that can stimulate future research, elaborate on limitations of current machine learning approaches for intensive within-person data, and discuss the results against the background of dynamic conceptualizations of personality.