Theory of Minds: Understanding Behavior in Groups through Inverse Planning

Sluggish cognitive tempo (SCT) is considered as a cognitive-emotional style. Theory of mind (ToM) skills form the basis of human cognition and social behavior. The aim of this study is to contribute to SCT literature by examining the relationship between SCT and cognitive and affective ToM in school-age children with attention deficit and hyperactivity disorder (ADHD). Fifty school-age children with ADHD (43 boys and seven girls) and 40 typically developing children (34 boys and six girls) were assessed using Parent-rated Barkley Child Attention Scale and Child Behavior Checklist/6-18, cognitive (first- and second-order ToM) and affective ToM (Reading Mind in the Eyes Test and Unexpected Outcomes Test (UOT)) tests. Multiple linear regression analysis was performed to determine the variables that may predict SCT levels in children with ADHD. Attention deficit and hyperactivity disorder-inattention (Exp(B) = .334, p = .027), internalization (Exp(B) = .305, p = .006), and UOT scores (Exp(B) = .933, p = .015) significantly predicted SCT severity in children with ADHD. SCT severity was significantly associated with impaired cognitive ToM skills as measured by second-order ToM (Exp(B) = 1.933, p = .045). Our findings may indicate that affective ToM developing with age, and impaired cognitive ToM skills are associated with increasing SCT severity.

This paper revisits the Simulation Theory (ST) as a framework for understanding human social cognition, challenging traditional ‘theory of mind’ or ‘folk psychology’ approaches. While these theory-based models posit that humans use an implicit body of knowledge to interpret and predict others’ behavior, ST emphasizes the use of mental simulation, leveraging the brain’s existing mechanisms for planning and prediction. By employing a predictive coding strategy, the brain minimizes cognitive load, interpreting others' actions through ‘inverse planning’ – a process that reuses one’s own action planning system to hypothesize the goals and intentions of others. The concept of agent-neutral coding is introduced, proposing that inputs for self and others are initially shared, reducing the need for explicit mental state attributions. This approach not only economizes cognitive resources but aligns with evolutionary perspectives on human social interaction in small, cohesive groups. In addition, the paper explores the role of perspective-taking and error correction in adapting shared mental representation. This reevaluation of ST underscores its efficiency and adaptability, offering a streamlined alternative to theory-based accounts of social cognition.

With the rapid development of digital technologies, intelligent generation of game narratives has become a crucial research direction in artificial intelligence and game design fields. However, existing algorithms lack deep understanding of group psychological mechanisms and struggle to generate authentic and credible multi-character interaction scenarios. Based on social proof effect theory, this study constructs an innovative multi-character interaction group behavior simulation algorithm aimed at enhancing the coherence, authenticity, and user experience of game narratives. The research employs a methodology combining theoretical modeling, algorithm design, and empirical validation. First, an "Environment-Cognition-Society-Behavior" quaternary interaction theoretical framework is constructed, providing in-depth analysis of environmental factors' influence mechanisms on group behavior, including the operational patterns of spatial layout, environmental complexity, and contextual cues. Second, the dynamic evolution mechanisms of social proof effects are systematically explored, revealing the inverted U-shaped relationship between group size and influence propagation, the S-shaped temporal curve characteristics of group behavior convergence, and the moderating role of individual differences in environmental adaptation. Building upon this foundation, a narrative generation algorithm based on Graph Neural Networks and Multi-Agent Reinforcement Learning is designed and implemented. Through the collaborative operation of a social proof intensity calculation engine, multi-character decision coordinator, and dynamic narrative generator, high-quality adaptive narrative creation is achieved. Through large-scale user experience testing involving 180 participants, the study validates the algorithm's effectiveness: compared to traditional methods, narrative logical consistency improved by over 40%, character behavior credibility scores reached 8.5 points, overall user immersion increased by 45%, average gameplay duration increased by 68%, and replay rate reached 73.2%. Algorithm performance testing demonstrates an average response time of only 127 milliseconds, memory usage reduced by 39.8%, CPU utilization decreased by 50.4%, exhibiting excellent scalability and system stability that fully meets industrial-grade application requirements. The research achievements not only provide crucial support for technological innovation in the gaming industry but also establish foundations for application expansion in education and training, social governance, mental health, and other fields, possessing significant theoretical value and broad practical application prospects. This study successfully validates the tremendous potential of psychological theories in artificial intelligence algorithm design, opening new pathways for interdisciplinary integration research.

This paper investigates a relatively new direction in Multiagent Reinforcement Learning. Most multiagent learning techniques focus on Nash equilibria as elements of both the learning algorithm and its evaluation criteria. In contrast, we propose a multiagent learning algorithm that is optimal in the sense of finding a best-response policy, rather than in reaching an equilibrium. We present the first learning algorithm that is provably optimal against restricted classes of non-stationary opponents. The algorithm infers an accurate model of the opponentýs non-stationary strategy, and simultaneously creates a best-response policy against that strategy. Our learning algorithm works within the very general framework of n-player, general-sum stochastic games, and learns both the game structure and its associated optimal policy.

A survey on multi-agent reinforcement learning and its application

The purpose of this article (review article doi: https://doi.org/10.48274/IBI54) is to provide a new perspective on Theory of Mind. We know that Theory of Mind (ToM), or mentalization, is the ability to attribute mental states, such as beliefs, desires, intentions, and emotions, to oneself and others, and to understand that these states may differ from those of others and from reality. This ability is fundamental to social interactions, communication, and understanding human behavior. In this context, ToM is enriched by Consciousness, becoming the Theory of Conscious Mind, providing more precise indications on how one\'s own and others\' mental states are attributed and how this new vision can be of concrete support in improving one\'s personal health and changing social relationships, promoting a new way of living life.

Matching low-cost camera and vector HD map is proven to be a practical and effective way of estimating the location and orientation of intelligent vehicles. However, map-based approach is viable only when the landmark observation is adequate and precise. In some areas with sparse and noisy observation, or even non-existent map matching features, the localization results may be unstable. In this paper, we introduce a novel algorithm by fusing visual odometry and vector HD map in a tightly-coupled optimization framework to tackle these problems. Our algorithm exploits the observation of visual feature points and vector HD map landmarks in the sliding window manner and optimize their residuals in a tightly-coupled approach. In this way, the system is more robust against the noisy HD map landmark observations. In addition, our method is able to accurately estimate vehicle pose even when landmarks are sparse. Experiments under two challenging scenarios with noisy and sparse landmark observations show that our method can achieve the Mean Absolute Error (MAE) at 0.1473m and 0.2496m respectively.

As one technique for autonomous driving, vehicular networks can achieve high efficiency with vehicle-and-infrastructure cooperation, bringing high safety and many value-added services. To achieve higher communication efficiency, much effort has been done to cope with the resource allocation issues for vehicular networks. Nevertheless, due to the strong nonconvexity and nonlinearity, the classical joint resource allocation problem in vehicular networks is typically NP-hard. The multiagent reinforcement learning (MARL) has emerged as a promising solution to tackle this challenge but its stability and scalability are not satisfactory when the amount of vehicles gets increased. In this article, we mainly investigate the issue of joint spectrum and power allocation in vehicular communication networks, and carefully consider the interactions between the vehicles and environment by incorporating the cooperative stochastic game theory with MARL, named complete-game MARL (CG-MARL), to achieve a better convergence and stability with the theoretical computational complexity <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {O}(n^{N})$ </tex-math></inline-formula> with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$n$ </tex-math></inline-formula> denoting the dimension of action space and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$N$ </tex-math></inline-formula> denoting the number of V2X Vehicular. Furthermore, the mean-field game (MFG) theory is employed to further enhance the MARL for decreasing the horrible computing resource consumption caused by the CG-MARL to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\mathcal {O}(n^{2})$ </tex-math></inline-formula> while maintaining an approximate performance. The simulation results demonstrate that the proposed mean-field-aided MARL (MF-MARL) for vehicular network resource allocation can achieve 95% near-optimal performance with much lower complexity, which indicates its significant potentials in the scenarios with massive and dense vehicles.

Metacognition facilitates theory of mind through optimal weighting of trait inferences.

Inverse modeling for computing a high-dimensional spatially-varying property field from indirect sparse and noisy observations is a challenging problem. This is due to the complex physical system of interest often expressed in the form of multiscale PDEs, the high-dimensionality of the spatial property of interest, and the incomplete and noisy nature of observations. To address these challenges, we develop a model that maps the given observations to the unknown input field in the form of a surrogate model. This inverse surrogate model will then allow us to estimate the unknown input field for any given sparse and noisy output observations. Here, the inverse mapping is limited to a broad prior distribution of the input field with which the surrogate model is trained. In this work, we construct a two- and three-dimensional inverse surrogate models consisting of an invertible and a conditional neural network trained in an end-to-end fashion with limited training data. The invertible network is developed using a flow-based generative model. The developed inverse surrogate model is then applied for an inversion task of a multiphase flow problem where given the pressure and saturation observations the aim is to recover a high-dimensional non-Gaussian log-permeability field where the two facies consist of heterogeneous log-permeability and varying length-scales. For both the two- and three-dimensional surrogate models, the predicted sample realizations of the non-Gaussian log-permeability field are diverse with the predictive mean being close to the ground truth even when the model is trained with limited data.

Human social behavior crucially depends on our ability to reason about others. This capacity for ‘theory of mind’ plays a vital role in social cognition because it allows us not only to form a detailed understanding of the hidden thoughts and beliefs of other individuals but to also understand that they may differ from our own1–3. Although a number of areas in the human brain have been linked to social reasoning4, 5 and its disruption across a variety of psychosocial disorders6–8, the basic cellular mechanisms that underlie human theory of mind remain undefined. Using a rare opportunity to acutely record from single cells in the human dorsomedial prefrontal cortex, we discover neurons that reliably encode information about others’ beliefs across richly varying scenarios and that distinguish self- from other-belief related representations. By further following their encoding dynamics, we show how these cells represent the contents of the other’s beliefs and accurately predict whether they are true or false. We also show how they track inferred beliefs from another’s specific perspective and how their activities relate to behavioral performance. Together, these findings reveal a detailed cellular process in the human dorsomedial prefrontal cortex for representing another’s beliefs and identify candidate neurons that could support theory of mind.

In the scenario of Industry 4.0, mobile smart devices (SDs) on production lines have to process massive amounts of data. These computing tasks sometimes far exceed the computing capability of SDs and require lots of energy and time to process. How to effectively reduce energy consumption and latency is necessary to be solved. To this end, we first propose an SDN (Software Defined Network)-based MEC (Mobile Edge Computing) system. In the MEC system, SDs can offload computation tasks to edge servers to decrease the processing latency and avoid the waste of energy. At the same time, taking advantage of SDNs programmability, scalability, and isolation of the control plane and the data plane, an SDN controller can manage edge devices within the MEC system. Secondly, based on a stochastic game, we study the computation offloading and resource allocation problems in the MEC system and establish a stochastic game-based computation offloading model. Furthermore, we prove that the multi-user stochastic game in this system can achieve Nash Equilibrium. We further consider each SD as an independent agent and design a stochastic game-based resource allocation algorithm with prioritized experience replays (SGRA-PER) to minimize energy consumption and processing latency with Multi-Agent Reinforcement Learning. Experiment results demonstrate that the proposed SGRA-PER is superior to MADDPG, Q-Mix, and MAPPO algorithms, which can significantly reduce the processing delay and energy consumption with dynamic resource allocation. Moreover, SGRA-PER can still keep a higher performance under the increase of SDs, which can be applied in a large-scale MEC system.

Reaching agreement on contribution behavior in different cultures-a public goods game with representatives in Japan and Germany

Modeling group behavior is very critical in the context of simulations designed for training personals or evacuation decision making strategies. The behavior of agents in such simulations plays a very important role. Agent behavior in complex environments such as emergency evacuation scenario involves collaboration with other agents. A simple example can be a group of people in a building trying to reach an exit by minimizing their travel time. A further complex situation would be involving several group of people trying to reach multiple exits. This paper describes how complex simulation environments involving group behavior can be constructed by fusing human behavior into the simulation. The developed prototype system is able to demonstrate some emergent human social behaviors, such as competitive, queuing, and herding behaviors.

The majority of multi-agent reinforcement learning (MARL) implementations aim to optimize systems with respect to a single objective, despite the fact that many real-world problems are inherently multi-objective in nature. Research into multi-objective MARL is still in its infancy, and few studies to date have dealt with the issue of credit assignment. Reward shaping has been proposed as a means to address the credit assignment problem in single-objective MARL, however it has been shown to alter the intended goals of a domain if misused, leading to unintended behaviour. Two popular shaping methods are potential-based reward shaping and difference rewards, and both have been repeatedly shown to improve learning speed and the quality of joint policies learned by agents in single-objective MARL domains. This work discusses the theoretical implications of applying these shaping approaches to cooperative multi-objective MARL problems, and evaluates their efficacy using two benchmark domains. Our results constitute the first empirical evidence that agents using these shaping methodologies can sample true Pareto optimal solutions in cooperative multi-objective stochastic games.