Human Behavior and Emerging Technologies最新文献

The “social being” perspective has largely influenced the design and research of AI virtual agents. Do humans really treat these agents as social beings? To test this, we conducted a 2 between (Cyberball condition: exclusion vs. fair play) × 2 within (coplayer type: AGENT vs. HUMAN) online experiment employing the Cyberball paradigm; we investigated how participants (N = 244) responded when they observed an AI virtual agent being ostracised or treated fairly by another human in Cyberball, and we compared our results with those from human–human Cyberball research. We found that participants mindlessly applied the social norm of inclusion, compensating the ostracised agent by tossing the ball to them more frequently, just as people would to an ostracised human. This finding suggests that individuals tend to mindlessly treat AI virtual agents as social beings, supporting the media equation theory; however, age (no other user characteristics) influenced this tendency, with younger participants less likely to mindlessly apply the inclusion norm. We also found that participants showed increased sympathy towards the ostracised agent, but they did not devalue the human player for their ostracising behaviour; this indicates that participants did not mindfully perceive AI virtual agents as comparable to humans. Furthermore, we uncovered two other exploratory findings: the association between frequency of agent usage and sympathy, and the carryover effect of positive usage experience. Our study advances the theoretical understanding of the human side of human–agent interaction. Practically, it provides implications for the design of AI virtual agents, including the consideration of social norms, caution in human-like design, and age-specific targeting.

Smart cities are hubs of innovation and rapid development, where technology plays a significant role in shaping the urban environment. In such settings, the ability to adapt and think flexibly is crucial for individuals to thrive. The study’s goal is to understand which factors influence how well individuals are equipped with 21st-century skills and how they apply them in the dynamic context of modern urban living. Employing a quantitative research approach, data were initially collected from 600 Thai respondents through convenience sampling. A subsequent data-cleansing process refined the focus to 568 respondents, selected based on their high-score attainment. The study utilized percentages, means, and binary logistic regression in a comprehensive data analysis. Its key findings illuminate the influence of various factors, including score, age, career trajectory, and engagement with social media platforms like Instagram, LINE, and X, on skill development within the smart city context. This research offers a nuanced perspective on the myriad elements that foster individual and collective success in technologically advanced urban environments. It moves beyond mere acknowledgment of innate abilities or age-related wisdom, highlighting the critical roles of career advancement and digital engagement in skill enhancement. The implications of these findings are far reaching, especially for policymakers, educators, and industry leaders. They underscore the need for a comprehensive approach to shaping the future of urban living and working spaces, ensuring that individuals are not just equipped with the necessary skills but are also adept at applying them effectively in the smart cities of tomorrow. This study serves as a guide for these stakeholders, emphasizing the importance of fostering environments that support continuous learning, innovation, and adaptability in the face of technological advancements.

Background: Exclusive breastfeeding (EBF) of infants during the first 6 months of their life is strongly recommended. Video-based communication is an innovative method that could promote EBF. The present study is aimed at assessing the effectiveness of video-based intervention in promoting breastfeeding practices from birth to 6 months postpartum. This was evaluated through maternal reports, as well as EBF of infants at 3 and 5 months of age, using the deuterium oxide dose-to-mother (DTM) technique.

Methods: In a cluster randomized community trial, 16 communities were assigned to receive a video-based behavior change communication (Video-Health) or standard health care messages (Control). Pregnant women in their first trimester (12 ± 2 weeks) were enrolled and followed up, together with their infants, until 6 months postpartum. Data on breastfeeding practices were collected by questionnaire, monthly, in all the participants (n = 508). Human milk intake (HMI) and nonmilk oral intake (NMOI) were measured in a subsample of 60 mother–infant pairs at 3 and 5 months postpartum using DTM. Mixed models and logistic regression were used to examine the difference in continuous and discrete breastmilk practices between the intervention and the control arms, respectively.

Results: The majority of mothers initiated breastfeeding early (92.1%), 82.4% reported colostrum feeding, and 17.5% provided prelacteal feeding. No significant differences were found between Video-Health and Control arms (p > 0.05). The intervention significantly improved reported EBF rates at 4 and 5 months postpartum (p < 0.05). DTM results showed that the proportion of women practicing EBF was 50%–67% less than reported at 3 months for both arms and at 5 months in the intervention arm. The intervention did not significantly affect measured EBF at 3 and 5 months postpartum but did improve HMI at 3 months. Non significant differences in NMOI were observed at 3 months, but at 5 months, there was a significant difference between the study arms.

Conclusions: Video-based behavior change communication did not result in significant improvements in reported breastfeeding practices, except for a higher reported adherence to EBF beyond 3 months. Additionally, the intervention had no effect on EBF as measured by the DTM. Early introduction of non-breastmilk foods and liquids persisted, despite self-reported EBF extending up to 6 months postpartum.

Trial Registration: ClinicalTrials.gov identifier: NCT04414527

There remains a dearth of answers to certain inquiries regarding oral English presentations, despite the growing interest in this field in recent years. The current study examined Thai university students’ perspectives on their English presentation abilities, the difficulties they encounter when delivering English presentations in an online environment, and their preferred mode of presentation. They enrolled in a 12-week online course on English presentation and communication skills through the Zoom application. This study employed a qualitative methodology and utilized reflective writing from students in response to four open-ended questions that prompted respondents to consider their experiences. A thematic approach was applied to the reflective essays to identify recurring themes and patterns within the data. The findings of the analysis indicated that, overall, the students expressed the need for enhancement in their English presentation abilities. Furthermore, this research emphasized several challenges that students confronted when delivering online presentations, encompassing nonverbal communication, anxiety, the audience’s engagement, and addressing questions. Despite the apparent convenience of online presentations, they conveyed a preference for face-to-face presentations. Interaction, teacher and peer support, and a desire to become a professional presenter were the three factors that affected their preference. This study establishes a fundamental basis for subsequent investigations and proposes strategies that higher education instructors can implement to enhance their students’ English presentation skills.

In various domains, virtual reality (VR) has emerged as a valuable tool for simulating specific scenarios and facilitating training within controlled and secure conditions. As peripheral vision plays a crucial role in numerous contexts, it must be duly considered in VR simulations. However, while peripheral vision was examined in VR, comparisons between VR and the real world (RW) are rarely made. Therefore, this study is aimed at comparing RW and VR reaction times (RTs) to peripheral visual stimuli and the field of view (FOV). This comparison is achieved using the peripheral perception-R (PP-R) of the Vienna Test System and a programmed virtual replica. The virtual replica underwent additional testing, revealing good reliability for RTs but only moderate to poor for measurements of the participant’s FOV. The comparison between the two environments indicates slower RTs to peripheral visual stimuli in VR than in RW. This observed discrepancy is consistent with the results of previous studies investigating RTs to foveal stimuli and can be partially explained by the latencies in the hardware and software configurations used. Nevertheless, the observed correlations between the RTs in both conditions suggest comparable visual processing within the peripheral visual field and affirm the decent replication of the real PP-R in VR. Overall, the study’s results support the usage of VR as a tool for practicing and examining specific scenarios, including peripheral vision. The discrepancies revealed in the RTs between VR and RW emphasize the need to continue examining hardware and software components in VR research.

Social networks on the dark web are rich in data that provides valuable insight into the nature of the activities on the dark web and human behaviors related to these activities. It also encompasses a diversity of ideologies, interests, and thought patterns associated with illicit activities and businesses on the dark web. For this reason, social networks on the dark web constitute a powerful tool and a profuse data source for various investigative work. However, such investigations encounter considerable challenges related to the massive volumes of textual data, analyzing it effectively, and extracting knowledge from it. This knowledge can be used in various investigations and studies when representing it in ontologies as a unified and integrative data source. In this paper, we introduce a novel approach for extracting and representing knowledge hidden in dark web communities through topic modeling and ontology learning methods. We start from the conceptual design of the ontology and employ several stages of text processing and analysis to achieve the desired knowledge graph, DarkOnto. These stages include data cleaning and preprocessing, topic modeling using correlated topic model (CTM), class-topic similarity estimation, ontology construction, ontology population, and ontology evaluation, where the proposed approach achieved high results. Furthermore, we discuss the results, limitations, challenges, and future work. This paper presents a promising approach for extracting hidden valuable knowledge from dark web communities where investigating and conceptualizing criminal communities can be conducted efficiently.

Food recognition, a field under food computing, has significantly promoted people’s dietary decision-making and culinary customs. We present the design and evaluation of a sociocultural app for African food recognition using deep learning models such as transfer learning. Deep learning models have multiple processing layers that make them robust in image recognition. Based on this capability of deep learning models, we explored them in this study. A total of 3142 food image datasets were collected from three African countries: Nigeria, Ghana, and Cameroon. Using the datasets, we developed and trained a deep learning model for recognizing African foods. The model attained a test accuracy of 94.5%. The model was further deployed in a food recognition app. To evaluate the predictive ability of the app, we recruited 16 participants who were interviewed and subsequently used the app in the wild for 7 days. In a comparative evaluation between the app and human recognition capabilities, we found that the app recognized 71% of the instances of food images generated by the participants and tested with the app, while the human evaluators (participants) could only recognize 56% of the food datasets. Participants were mostly able to recognize some foods from their own country. Furthermore, participants suggested some design features for the app. In view of this, we offer design recommendations for researchers and designers of sociocultural food recognition systems.

Artificial intelligence (AI) is a rapidly expanding field. Given the dynamic nature of AI research, staying up-to-date with this evolving landscape is imperative to guide the global race in this field. Using the world’s largest bibliometric database (SciVal/Scopus), we analyzed the research output of the top 15 countries in AI research productivity between the years 1998 and 2022. Over the past 25 years, the global research output in AI has increased by 26 folds, with China leading the way in the overall number of AI publications (~281,650 articles). During this period, China and India had the highest fold growth (~240–250) and compound annual growth rate (CAGR) (~25). However, when normalized to population size and gross domestic product (GDP), Singapore (250 AI publications per million capita) and India (7.7 AI publications per billion GDP) were the best performers in 2022, respectively. When normalized to the number of researchers and gross expenditures on research and development (GERD), again, India was the best with 36.8 AI publications per thousand researchers and 0.77 AI publications per million GERD in 2020. When the overall quality of AI publications over the past decade was considered, Hong Kong and Singapore performed the best with respect to publications in higher percentile journals (the top 1%, 5%, and 10% journals), and Switzerland performed the best with respect to citations/publication and field-weighted citation impact (FWCI). Finally, when the quality measures of publications were normalized to country or research metrics, the three best performers were Hong Kong, Singapore, and Switzerland (although India stood out for citations/GERD). These three countries (along with the United States) also had the highest economic impact as measured by the patent citations per scholarly output, which was aligned with the high FWCI values for these countries. Long-term strategies to help leading countries gain an advantage in the AI frontier have been discussed at the end of the manuscript. These strategies include government initiatives and financial support, talent creation and networking, public awareness, and mindful considerations of rules, guidelines, and ethics.

Smartphone use during parenting is common, which may lead to distraction (also known as technoference). However, it is likely that some phone activities are less disruptive to parents and children. In this study, we explored smartphone use (via passive sensing across 8 days) within 264 parents of infants, measuring parents’ application use on their phone (e.g., messaging, social media, mobile gaming, video chat) and phone use across contexts (e.g., during feeding and at bedtime). We utilized latent profile analysis to identify profiles of users, revealing five user types: Moderate User Social Networkers (37%), followed by Moderate User Gamers (20%), Moderate User Video Chatters (17%), Low Users (15%), and Heavy Users (11%). Parents varied in their use, from Low Users, who used their phone approximately 2.4 h each day, spent only 13% of their child time on their phone, and used their phone for about 18 min at bedtime, to Heavy Users, who spent approximately 8 h a day, about 50% of their child time on their phone, and about 1 h at bedtime. Heavy Users showed higher depressive symptoms and poorer sleep (although not poorer sleep than Moderate User Gamers). Surprisingly, we found no differences between groups in perceptions of parenting stress, responsiveness to their infant, or problematic phone use and distraction. We also explored demographic differences across groups. We call for future work to examine parent phone use more comprehensively and holistically and to view specific phone use activities as simultaneously interconnected with other types of use activities.

The world of smart prisons is gaining momentum despite receiving limited attention due to the ability to maintain social order and reduce crime rates. Therefore, this research aims to synthesize the current knowledge of smart prisons, identify barriers to trend technologies, and provide valuable insights for practitioners, and policymakers. The systematic literature review (SLR) method by Kitchenham was used to collect data from Scopus from 2007 to 2023, which comprised 59 articles in the final phase. The high cost of smart technology, ineffective technology design, poor management, aging facilities, increased violence, ethical concerns, and lots more, were some of the barriers associated with the topic. The study highlights the need for targeted strategies, such as cost-benefit analyses, phased implementation, and comprehensive staff training, to overcome these barriers. The study also reveals varied approaches to smart prison adoption across countries, influenced by differences in economic resources and policy frameworks. The study emphasizes the importance of continuous evaluation, international collaboration, and adherence to ethical standards to sustain smart prison development. The result showed that smart prisons are important in the field of corrections.