Navigating human-populated environments without causing discomfort is a critical capability for socially-aware agents. While rule-based approaches offer interpretability through predefined psychological principles, they often lack generalizability and flexibility. Conversely, data-driven methods can learn complex behaviors from large-scale datasets, but are typically inefficient, opaque, and difficult to align with human intuitions. To bridge this gap, we propose RLSLM, a hybrid Reinforcement Learning framework that integrates a rule-based Social Locomotion Model, grounded in empirical behavioral experiments, into the reward function of a reinforcement learning framework. The social locomotion model generates an orientation-sensitive social comfort field that quantifies human comfort across space, enabling socially aligned navigation policies with minimal training. RLSLM then jointly optimizes mechanical energy and social comfort, allowing agents to avoid intrusions into personal or group space. A human-agent interaction experiment using an immersive VR-based setup demonstrates that RLSLM outperforms state-of-the-art rule-based models in user experience. Ablation and sensitivity analyses further show the model’s significantly improved interpretability over conventional data-driven methods. This work presents a scalable, human-centered methodology that effectively integrates cognitive science and machine learning for real-world social navigation.

In recent years, predicting Big Five personality traits from multimodal data has received significant attention in artificial intelligence (AI). However, existing computational models often fail to achieve satisfactory performance. Psychological research has shown a strong correlation between pose and personality traits, yet previous research has largely ignored pose data in computational models. To address this gap, we develop a novel multimodal dataset that incorporates full-body pose data. The dataset includes video recordings of 287 participants completing a virtual interview with 36 questions, along with self-reported Big Five personality scores as labels. To effectively utilize this multimodal data, we introduce the Psychology-Inspired Network (PINet), which consists of three key modules: Multimodal Feature Awareness (MFA), Multimodal Feature Interaction (MFI), and Psychology-Informed Modality Correlation Loss (PIMC Loss). The MFA module leverages the Vision Mamba Block to capture comprehensive visual features related to personality, while the MFI module efficiently fuses the multimodal features. The PIMC Loss, grounded in psychological theory, guides the model to emphasize different modalities for different personality dimensions. Experimental results show that the PINet outperforms several state-of-the-art baseline models. Furthermore, the three modules of PINet contribute almost equally to the model’s overall performance. Incorporating pose data significantly enhances the model’s performance, with the pose modality ranking mid-level in importance among the five modalities. These findings address the existing gap in personality-related datasets that lack full-body pose data and provide a new approach for improving the accuracy of personality prediction models, highlighting the importance of integrating psychological insights into AI frameworks.

Head-gaze interaction is an integral mode of interaction in virtual reality (VR) applications, demonstrating high precision in fine manipulation tasks but low efficiency in large-scale object movements. To enhance the efficiency of head-gaze interaction, this study adjusted the control-display gain to compensate for the weaknesses of head-gaze interaction in a long-distance object-positioning task. We investigated the effect of the control-display gain of head-gaze interaction on movement time (MT) using a cohort of participants (n = 24) to perform experiments. The results showed that the MT first decreased as the gain increased from 1 to 1.5 and then increased afterwards. Further analysis showed that a high gain improved the interaction efficiency in the ballistic phase, but reduced the interaction efficiency in the corrective phase. To be able to obtain higher efficiency of interaction, we designed a dual-gain mode which set different gains in the ballistic and corrective phases. Evaluated using an additional experimental cohort (n = 24), our results showed that the dual-gain mode was more efficient than the mono-gain mode. Moreover, the dual-gain mode with optimal gains did not induce a more serious perception of inconsistency, confusion, nonacceptance and motion sickness, while it had a tendency to reduce the total workload compared to the interaction with normal gain. Our findings provide potential valuable design insights and guidance contributing to improving the efficiency of head-gaze interaction in virtual spaces.