Real-world Interactions Research Articles

Gendered AI in fully autonomous vehicles: the role of social presence and competence in building trust

Purpose This study aims to examine users’ perceptions of gendered artificial intelligence (AI) interfaces in the context of autonomous vehicles (AVs). It focuses on the gendered effects of social presence, warmth and competence on trust and introduces the moderating role of perceived autonomy as a key factor. Design/methodology/approach A between-subjects experimental design was used (n = 309), using a 360-degree virtual tour simulation with gendered voice assistants. Findings As AVs are perceived as highly autonomous, the impact of gender on social presence intensifies, affecting trust. Female voices enhance social presence, conveying warmth but also perceived competence traits. Notably, competence impacts trust more significantly than warmth. Research limitations/implications The study’s experimental approach might not fully capture real-world interactions with AVs. Future research could benefit from field and longitudinal studies. Practical implications These findings are crucial for AV designers and interface developers. They highlight the importance of considering human-like characteristics such as gender and enhance perceptions of competence in developing highly autonomous AI interfaces. Social implications Addressing gender stereotypes in AV design is vital to ensure inclusivity, to cater for a diverse user base and to give all users a trustworthy experience. Originality/value This study is pioneering in its examination of how gender stereotypes impact trust toward AVs, an area previously unexplored, despite the significant influence gender might have due to the high level of autonomy and traditional car-related stereotypes. It uniquely identifies feminine traits as denoting competence in highly autonomous technologies like AVs, especially where safety is critical. This challenges traditional gender stereotypes and emphasizes the need to rethink attributes associated with competence and trust in AI.

When Your Words Fail: The (Hidden) Impact of Making Communication Errors in Suspect Interviews and Suicide Negotiations

Making communication errors affects law enforcement officers’ thoughts and feelings during suspect interviews and suicide negotiations. Specifically, these errors can lead to more negative emotions such as self-oriented anger, shame, and guilt, as well as stress in suspect interviews, and distraction in suicide negotiations. Law enforcement officers may use a variety of responses to resolve communication errors, ranging from taking responsibility by apologizing, exploring what is going on, deflecting blame, or providing unrelated comments. Taking part in the role-playing exercise that was designed for this study, and reviewing the study results, allowed law enforcement officers to safely experience and reflect on the consequences of communication errors. Learning more about how to handle communication errors helps officers to mentally prepare for real-world interactions and enhance their emotional awareness, as well as improve their ability to support colleagues during and after incidents.

Examining the relationship between language development, executive function, and screen time: A systematic review.

This systematic review study examines the relationship between language development, executive function, and screen time in early childhood. The early childhood period is a crucial phase for the development of the brain, during which fundamental language and executive function skills undergo rapid evolution. This review synthesizes findings from 14 peer-reviewed studies that focused on language development, executive function, and screen time together to provide a comprehensive understanding of their relationship. The findings of current study were categorized under four themes: screen content and adherence to guidelines, parent-child interaction and the family context, passive and active screen time, and attention issues. The findings suggest that interactive and educational screen content may positively influence language development and executive functions when aligned with recommended screen time guidelines. In contrast, excessive passive screen time, such as watching television, has been associated with negative impacts on cognitive and social skills, particularly affecting attention, memory, and emotional regulation. The prevalence of attention problems is found to be higher in individuals who engage in high levels of screen time. This highlights the necessity for balanced consumption of screen media. The review emphasizes the pivotal role of parent-child interaction, where high-quality engagement and verbal scaffolding during screen time can mitigate adverse effects. Furthermore, socioeconomic and cultural factors also play a significant role. Higher socioeconomic status (SES) families are better able to manage screen time and leverage educational content to support development. These findings have the potential to inform the actions of parents, educators, and policymakers. Adherence to recommended screen time guidelines can mitigate the potential negative impact on executive functions and language skills. Furthermore, the importance of limiting passive screen time and ensuring a balance between screen use and real-world interactions and play opportunities is also highlighted.

Integration of augmented reality technology to enhance English speaking skills for tourism industry employees in Sumbawa regency

The integration of Augmented Reality (AR) technology into language learning has gained attention due to its potential to enhance speaking skills in practical contexts like the tourism industry. This study examines the long-term effectiveness of AR in improving English speaking skills fluency, accuracy, pronunciation, and grammar among tourism employees in Sumbawa Regency. The researchers developed VirtuoBot, an AR-based learning platform designed to simulate real-world interactions through interactive dialogues. A mixed-methods approach was used to assess the impact of AR-based language training on speaking proficiency. 90 participants from four major tourist sites took part in pre-and post-tests, supplemented by qualitative interviews. Results showed significant improvements in fluency, accuracy, pronunciation, and grammar after the AR intervention. Challenges like limited access to AR technology and workforce readiness were also noted. This study suggests that AR, through VirtuoBot, is a promising tool for enhancing English speaking skills in tourism, contributing to better service quality. Further research is recommended to address accessibility challenges and optimize AR training methods, ensuring sustainable implementation and broader adoption in tourism sectors of developing regions like Sumbawa.

Temporal contact patterns and the implications for predicting superspreaders and planning of targeted outbreak control.

Directly transmitted infectious diseases spread through social contacts that change over time, but outbreak models typically make simplifying assumptions about network structure and dynamics. To assess how common assumptions relate to real-world interactions, we analysed 11 networks from five settings and developed metrics, capturing crucial epidemiological features of these networks. We developed a novel metric, the 'retention index', to characterize the distribution of retained contacts over consecutive time steps relative to fully static and dynamic networks. In workplaces and schools, contacts in the same department formed most of the retained contacts. In contrast, no clear contact type dominated the retained contacts in hospitals, thus reducing overall risk of disease introduction would be more effective than control targeted at departments. We estimated the contacts repetition over multiple days and showed that simple resource planning models overestimate the number of unique contacts by 20%-70%. We distinguished the difference between 'superspreader' and infectious individuals driving 'superspreading events' by measuring how often the individual represents the top 80% of contacts in the time steps over the study duration. We showed an inherent difficulty in identifying 'superspreaders' reliably: less than 20% of the individuals in most settings were highly connected for multiple time steps.

The effect of instruction on the development of L2 interactional competence: Listener responses in Chinese as a second language

Listener responses (LRs) are utterances produced by non-primary speakers (e.g. oh, okay). It is an important component of L2 learners’ interactional competence (IC). Previous studies have indicated that learners develop their capacity of deploying LRs naturally through study abroad programs. However, it remains unknown whether LRs can be effectively learned through classroom instruction, yet it is a critical issue facing language educators and learners because classroom is the primary locus for L2 learning for those without study abroad opportunities. Taking up this issue, this study compares two groups of beginning-level L2 Chinese learners, a control group with regular instruction and an experimental group with extra IC instruction, where authentic materials from L1 speakers' talk were employed to teach specific interactional practices such as LR and repair. Each learner was asked to record two spontaneous conversations with an L1 speaker, one at midterm and one at the end of the semester. Using a mixed-methods approach, this study first identifies specific types of LRs used by beginning-level learners. Statistical analyses then reveal that the two groups’ development from midterm to semester end differs significantly only in reactive expression and follow-up action. Complementing the quantitative analyses, the subsequent conversation analysis discovers qualitative developments of learners in the experimental group; their LRs show not only increased linguistic complexity and greater variety over time but also heightened sensitivity to context and capacity to advance the interaction. These findings suggest that systematic classroom instruction on IC can enhance L2 learners’ ability to engage in real-world interactions, thus supporting that IC is teachable and learnable in the classroom even for beginners with limited linguistic proficiency.

Interactional competence and performances of compliments and consolations by learners of Japanese

Abstract In recent years there has been an increased recognition of the need for learners to understand language as social action. However, language learners are often left on their own to develop the ability to adapt to the unanticipated demands of real-world interaction. To encourage this development, attention is turning toward the notion of interactional competence (IC). The present study contributes to efforts to better understand Japanese learners’ IC. Specifically, this study aims to better understand how learners naturally accomplish compliment and consolation speech acts and for what social purposes, including how learners adapt linguistic forms to do acts that go beyond the act most commonly associated with those forms. Conversation analysis of foreign language housing mealtime conversations revealed two core observations. First, based on preempirical descriptions, compliments occur with more frequency than consolations. Compliments also tend to be deployed as the first part of a compliment-response pair, while consolations tend to be deployed as a response to the public exposure of a challenge or a troubling emotional state. Second, when looking instead at how participants use turns that are construed as compliments regardless of linguistic form, we find that compliments are also used to do activities other than what their surface formulation suggests. A discussion of what these results mean for developing deeper understandings of IC in speech acts performed by language learners is provided.

Offline Reinforcement Learning for Adaptive Control in Manufacturing Processes: A Press Hardening Case Study

Abstract This paper explores the application of offline reinforcement learning in batch manufacturing, with a specific focus on press hardening processes. Offline reinforcement learning presents a viable alternative to traditional control and reinforcement learning methods, which often rely on impractical real-world interactions or complex simulations and iterative adjustments to bridge the gap between simulated and real-world environments. We demonstrate how offline reinforcement learning can improve control policies by leveraging existing data, thereby streamlining the training pipeline and reducing reliance on high-fidelity simulators. Our study evaluates the impact of varying data exploration rates by creating five datasets with exploration rates ranging from ε=0 to ε=0.8. Using the conservative Q-learning algorithm, we train and assess policies against both a dynamic baseline and a static industry-standard policy. The results indicate that while offline reinforcement learning effectively refines behavior policies and enhances supervised learning methods, its effectiveness is heavily dependent on the quality and exploratory nature of the initial behavior policy.

Empowering Dementia Communication via Virtual Reality AI-Driven Simulations

The use of effective communication skills is essential for direct care workers to provide quality care for persons living with dementia (PLWD) in long-term service and support settings (LTSS). However, direct care workers often report feeling unprepared, lacking competence in care, and fear interacting with PLWD. This proof-of-concept study proposes Virtual reality Communication Training Optimizing Real-world Interactions (VICTORI) intervention, aiming to develop and evaluate the usability, acceptability, and preliminary effectiveness of virtual reality (VR) simulated communication training for direct care workers using artificial intelligence (AI)-generated patients with dementia. Care-related and social communication scenarios in dementia care will be developed, based on current literature, our validated communication instrument, and AI technology. Six modules consisting of pre-briefing, simulation facilitation, and de-briefing will be developed, refined, and evaluated based on an advisory board of experts and direct care workers’ quantitative and qualitative feedback. To test the usability, acceptability, and effectiveness of the VICTORI intervention, we will use a single group pre- and post-test intervention design, integrating quantitative measures and qualitative semi-structured interviews in this mixed-methods study. Thirty direct care workers will be recruited at two LTSS in North Texas. Direct care workers will participate in six VR-AI simulation commu-nication training sessions that include communication with AI-generated patients, real-time feedback, and detailed evaluation of communication behaviors. Each simulation will be recorded to further evaluate the partici-pant’s communication skills using the Dyadic Communi-cation Observational coding scheme in Dementia care (DCODE). Additionally, participants' potential side effects during VR simulation will be assessed using continuous physiological data and self-reported questionnaires at each simulation session. Communication knowledge and com-petence in dementia care will be assessed before and after training using self-reported questionnaires. Direct care workers will also be interviewed to assess acceptability and satisfaction with the VICTORI training. This proposal will introduce innovative training methods that leverage AI technology into communication training in dementia care. Objective assessments, including AI-based feedback, physiological data, and observational assessments, ensure the reliability and validity of the data to evaluate plausibility of VICTORI intervention.

Host nest defence does not act as selective agent against plumage polymorphism in brood parasites.

Batesian mimicry in brood parasites is often viewed as an evolutionary strategy to mitigate host aggression. Female common cuckoos (Cuculus canorus) exhibit two morphs: the hawk-like grey and the rufous one, potentially maintained by apostatic selection. It was hypothesized that the grey morph's predator-like appearance deters host defences, while the rufous morph benefits from its rarity by evading host attention. Previous research predominantly utilized static cuckoo dummies, lacking insights into real-world interactions. We investigated the effectiveness of the cuckoo morphs in accessing great reed warbler (Acrocephalus arundinaceus) nests under natural conditions. Analysing video-recorded cuckoo attempts, we found no significant difference in nest-access success between the morphs. Both experienced a similar probability of physical attacks when hosts were present, and the rufous morph did not evade host detection more often compared with the grey morph. These results fail to support the assumptions of (a) Batesian mimicry, that hawk-like mimicry enhances nest access or reduces host aggression, and (b) apostatic selection, that the rarity of the rufous morph confers an advantage in successfully accessing the host nest. Future research should aim to identify stages in the cuckoo's life cycle or host interactions where colour polymorphism provides an evolutionary benefit.

Learning accurate neighborhood- and self-information for higher-order relation prediction in Heterogeneous Information Networks

Heterogeneous Information Networks (HINs) are commonly employed to model complex real-world scenarios with diverse node and edge types. However, due to constraints in data collection and processing, constructed networks often lack certain relations. Consequently, various methods have emerged, particularly recently, leveraging heterogeneous graph neural networks (HGNNs) to predict missing relations. Nevertheless, these methods primarily focus on pairwise relations between two nodes. Real-world interactions, however, often involve multiple nodes and diverse types, extending beyond simple pairwise relations. For instance, academic collaboration networks may entail interactions among authors, papers, and conferences simultaneously. Despite their prevalence, higher-order relations are often overlooked. While HGNNs are effective at learning network structures, they may suffer from over-smoothing, resulting in similar representations for nodes and their neighbors. The learned inaccurate proximity among nodes impedes the discernment of higher-order relations. Furthermore, observed edges among a target group of nodes can provide valuable evidence for predicting higher-order relations. To address these challenges, we propose a novel model called Accurate Neighborhood- and Self-information Enhanced Heterogeneous Graph Neural Network (ANSE-HGN). Building upon HGNNs to encode network structure and attributes, we introduce a relation-based neighborhood encoder to capture information within multi-hop neighborhoods in heterogeneous higher-order relations. This enables the calculation of accurate proximity among target groups of nodes, thereby enhancing prediction accuracy. Additionally, we leverage self-information from observed higher-order relations as an auxiliary loss to reinforce the learning process. Extensive experiments on four real-world datasets demonstrate the superiority of our proposed method in higher-order relation prediction tasks.

Exploring the Link Between Early Technology Exposure and Developmental Milestones in Childhood.

Background This study examined the relationship between touchscreen exposure and developmental outcomes in early childhood using the Ages and Stages Questionnaire, Third Edition (ASQ-3), a widely-used developmental screening tool. Methods During a well-child visit, parents of 51 children aged 18-36 months completed the ASQ-3 and a survey on their child's touchscreen habits (timing of introduction, daily usage duration, independent use, engagement in creative activities). Medical records were reviewed for developmental diagnoses and health conditions. Analysis One-way analysis of variances (ANOVAs) examined differences in ASQ-3 domain scores (problem-solving, personal-social, communication, gross motor, fine motor) based on the age of touchscreen introduction and average daily usage time. Independent t-tests compared scores between groups defined by independent touchscreen use and creative activity engagement. Effect sizes were calculated for significant differences. Results In the problem-solving domain, children introduced to touchscreens before 12 months of age scored significantly higher than those introduced later, with a statistical significance of p<0.05 and an effect size of d=0.45. However, there were no significant differences in scores based on the duration of daily touchscreen usage. In the personal-social domain, children introduced to touchscreens at 12 months or later scored significantly higher than those introduced before 12 months, with a significance of p<0.05 and an effect size of d=0.51. Moderate daily usage of 30-90 minutes was associated with higher scores compared to lower usage of less than 30 minutes or higher usage of more than 90 minutes (p<0.05). Additionally, children who used touchscreens independently had significantly higher scores, with a significance of p<0.01 and an effect size of d=0.76. Those who engaged in activities such as coloring or crafting also had significantly higher scores, with a significance of p<0.01 and an effect size of d=0.82. In other domains, such as communication, gross motor, and fine motor, no significant effects of touchscreen exposure were observed. Discussion Earlier touchscreen introduction may facilitate problem-solving skills like experimentation and cognitive flexibility through interactive digital play. However, excessive touchscreen use (greater than 90 minutes per day) before 12 months could interfere with pivotal social-emotional learning from real-world interactions, impeding personal-social development. Moderate, supervised touchscreen exposure combined with creative hands-on activities appears optimal for fostering strong personal-social competencies like cooperation and emotional understanding during the toddler years. Conclusion Timing, duration, context, and content of touchscreen experiences in early childhood relate differently to specific developmental domains. Guidelines should consider these nuances to support overall healthy child development in our digital world. Larger longitudinal studies using multi-method assessments are needed.

Essential Skills in Teaching English to Children: A Comprehensive Overview and Research-Based Analysis

Teaching English to children presents unique challenges and requires a specific set of skills distinct from teaching adults. Children acquire language naturally through real-world interaction, play, and sensory experiences. This article examines the essential skills that teachers need to effectively teach English to young learners and optimize their language acquisition. It also incorporates findings from existing research to highlight the effectiveness of various teaching methodologies, stressing the role of motivation, interaction, classroom management, and technology in fostering successful English learning experiences for children.

Motion control of autonomous underwater vehicle based on physics-informed offline reinforcement learning

Online reinforcement learning (RL) methods for autonomous underwater vehicles (AUV) are time-consuming and unsafe due to the need for real-world interaction. Offline RL methods can improve efficiency and safety by training with dynamic models, but an accurate model for AUV is difficult to obtain due to its highly nonlinear dynamics. These limit the application of RL methods in AUV control. To solve this issue, we propose physics-informed model-based conservative offline policy optimization (PICOPO). It offers the advantages of small dataset, strong generalizability and high safety by combining the physics-informed dynamic modelling method and the offline RL technique. First, the PICOPO constructs a physics-informed model based on a small offline dataset to serve as the digital twins (DT) of the actual AUV. This DT can forecast the long-term motion states of AUV with high-precision. The RL-based controller is then trained offline within this DT, eliminating the need for real-world interaction and allowing direct deployment to the AUV without fine-tuning. In this paper, simulations and field tests are carried out to evaluate the proposed method. Our results demonstrate that PICOPO achieves accurate motion control with just 2000 samples and enables zero-shot sim-to-real transfer, showcasing strong generalizability across various motion control tasks.

Transformable Soft Gripper: Uniting Grasping and Suction for Amphibious Cross-Scale Objects Grasping.

Robotic grasping plays a pivotal role in real-world interactions for robots. Existing grippers often limit functionality to a single grasping mode-picking or suction. While picking handles smaller objects and suction adapts to larger ones, integrating these modes breaks scale boundaries, expanding the robot's potential in real applications. This article introduces grasping modes transformable soft gripper capable of achieving amphibious cross-scale objects grasping. Despite its compact and fully scalable design (20 mm in diameter prototype), it morphs into two configurations, gripping objects from 10% (2 mm) to over 1000% (200 mm) of its size, spanning a vast 100-fold range. To enhance its grasping efficacy, we derived theoretical analytical models for the two distinct grasping modes. Subsequently, we present a detailed illustration of the gripper's fabrication process. Experimental validation demonstrates the gripper's success in attaching or detaching everyday items and industrial products, achieving high success rates in both air and underwater scenarios. Amphibious grasping and card manipulation demonstrations underscore the practicality of this transformative soft robotics approach.

Evolution of cooperation on dynamical asymmetrical networks with costs

Social networks denote the structure of interactions among individuals. Adjusting social relationships is a typical reaction among humans attempting to interact with altruistic partners. Theoretical investigations proved that dynamical network structures promote the evolution of cooperation. However, individuals react differently to diverse social partners and interactions. Furthermore, a large variety of potential costs of partner switching occur in real-world interactions. Establishing and maintaining different interactions cost differently. Based on these motivations, this investigation studies the interplay between the dynamics on networks and the dynamics of networks, which entangles the evolution of strategies and topology of adaptive multilayer networks whose structure is divided into a gaming layer for reaping payoff and a learning layer for spreading strategy. Individuals react differently to overlapped and un-overlapped partners whose gaming interactions and learning interactions are identical or not. And the effects of overlap levels and rewiring costs on the evolution of cooperation are explored. Simulation results demonstrate that increasing the ratio of rewiring as well as the overlap levels can enhance the evolution of cooperation significantly. However, both too frequent rewiring and too high overlap levels can result in the increment of isolated nodes, and thus provide a sanctuary for defectors to survive and make cooperators hardly to occupy the whole population. Moreover, it is found that rewiring costs have dramatically different impacts on the evolution of cooperation for different overlap levels. For a low overlap level, increasing costs suppress the evolution of cooperation. Yet a high overlap level makes increasing costs favor the cooperator to spread. Our results reveal the condition for the domination of cooperation with asymmetrical interactions, which may provide a potential way to understand the evolutionary cooperation in human society.

Humans flexibly integrate social information despite interindividual differences in reward

There has been much progress in understanding human social learning, including recent studies integrating social information into the reinforcement learning framework. Yet previous studies often assume identical payoffs between observer and demonstrator, overlooking the diversity of social information in real-world interactions. We address this gap by introducing a socially correlated bandit task that accommodates payoff differences among participants, allowing for the study of social learning under more realistic conditions. Our Social Generalization (SG) model, tested through evolutionary simulations and two online experiments, outperforms existing models by incorporating social information into the generalization process, but treating it as noisier than individual observations. Our findings suggest that human social learning is more flexible than previously believed, with the SG model indicating a potential resource-rational trade-off where social learning partially replaces individual exploration. This research highlights the flexibility of humans' social learning, allowing us to integrate social information from others with different preferences, skills, or goals.

Driving engagement in metaverse-mediated tourism environments: exploring the role of perceived realism

PurposeDespite the promising development and marketing potential of the metaverse, our understanding of how realistic metaverse environments impact user engagement and behaviours remains limited. This study investigates the role of perceived realism in influencing user engagement, thereby affecting external search behaviour and visit intentions.Design/methodology/approachWe surveyed 270 active metaverse users to test the research model. The data were analysed using partial least squares structural equation modelling (PLS-SEM).FindingsThe results of our study show that three dimensions of realism – avatar involvement, perceptual pervasiveness and social realism – significantly enhance user engagement, which in turn influences external search behaviour and visit intention. In contrast, simulation realism and freedom of choice have minimal effects on absorption and dedication.Research limitations/implicationsThis study highlights the role of perceived realism in enhancing user engagement with the metaverse and its impact on physical world behaviours. It contributes to metaverse literature by demonstrating that engagement within the metaverse significantly influences physical world behaviours, including visit intentions and external search behaviours.Practical implicationsThis study offers practical guidance for developers to enhance user engagement in metaverse environments. Specifically, our findings advocate for visual and audio quality enhancements, greater persuasiveness of virtual spaces, improved avatar representativeness and a closer alignment of metaverse activities with real-life events.Originality/valueThis study advances the theoretical understanding of perceived realism by examining how its dimensions – such as visual and audio quality, avatar representativeness and alignment with real-life events – impact user engagement in the metaverse. It also explores how this engagement influences offline behaviours, thus bridging the gap between virtual and real-world interactions.

Imitation-inhibition training can reduce the observation-inflation effect in face-to-face scenarios.

Observing others performing an action can lead to false memories of self-performance-the observation-inflation effect. Previous research has indicated that this phenomenon might impact the memory of actions in real-world interactions. However, whether direct observation without interaction can lead to observation inflation remains unclear. In Experiment 1, participants passively observed the experimenter performing actions live. In subsequent memory tests, they indeed reported false memories regarding their performances. Building on this, Experiment 2 investigated the causes of the observation-inflation effect induced by "real" actions. Participants underwent imitation-inhibition training with the individuals they observed previously. The results revealed that participants who completed imitation-inhibition training reported fewer false memories in memory tests than those who completed imitation training. These findings suggest that even passive observation of "real" actions can lead to observation inflation, and the simulation of others' actions by individuals may be a potential underlying cause of their occurrence in real-life situations.

Decentralized human-like control strategy of mixed-flow multi-vehicle interactions at uncontrolled intersections: A game-theoretic approach

A critical challenge that future autonomous driving systems face is improving the ability to cope with complex real-world interaction scenarios such as uncontrolled intersections. In the near future, a mixed traffic flow of human-driven vehicles (HDVs) and connected autonomous vehicles (CAVs) will coexist in transport networks, which motivates us to explore the interaction between HDVs and CAVs to improve traffic efficiency and safety. To help CAVs better interact with HDVs and adapt to the mixed-flow environment, we propose a human-like decentralized control strategy for CAVs. First, a game-theoretic framework is proposed to model multi-vehicle interactions (including HDV-CAV, CAV-CAV interactions) in the mixed-flow environment. The existence of solutions is proven to ensure the feasibility of the proposed game-theoretic model. Next, a driving style recognition algorithm is embedded into the proposed model to help CAVs understand and predict human drivers’ actions. The proposed model is calibrated via a real-world dataset and used to simulate traffic in several testing scenarios. Real-world vehicle trajectories are used to verify the accuracy of generated vehicle trajectories in simulations. Experimental results indicate that 1) CAVs can take more reasonable actions to determine whether to yield while ensuring safety when competing for the right of way with HDVs using the proposed method compared with conservative driving strategies, 2) a higher penetration rate of CAVs can significantly enhance travel efficiency and lower collision risk at uncontrolled intersections.