For social interaction on streaming platforms, chat feedback is a primary means for real-time engagement and expression of opinions. Given that social media is prone to stress by promoting social comparison, this study investigated how the valence and evaluation type of chat feedback influences streamers’ emotions, attention, and performance. In an online game-streaming context, participants engaged in a shooting game while receiving real-time chat feedback of different valence (negative, positive) and evaluation types (comparative, general). The results revealed that receiving negative feedback led to experiencing higher anxiety and lower self-efficacy and social support. Furthermore, comparative feedback negatively affected game performance and attracted more attention to the feedback. Interestingly, the tendency of comparative feedback to capture more attention was stronger when the valence was negative. These findings contribute to understanding the influence of social feedback on emotions and behavior and provide valuable insights for improving the user experience and performance.

In virtual reality, avatars represent ourselves and serve as a means to interact with others and the environment. Thus, understanding the self-recognition process in VR and designing self-avatars that have strong connections with the self is critical in enhancing immersion and efficiency of interaction in VR. In this paper, we investigated the characteristics of the self-recognition process in VR and the crucial factors that enhance the bond between the self and the avatar. In Study 1, we tested whether the advantage in cognitive processing of self-related information, often observed in reality, is also replicated for briefly embodied self-avatars in VR and how it is modulated by way of avatar selection and the degree of embodiment. The results showed that the self-avatar was processed more efficiently than other avatars, despite the constant changes in the appearance and a brief embodiment period. Moreover, the self-advantage effect was more pronounced for personally selected avatars, rather than those assigned by the experimenter. In Study 2, we compared the relative importance of motion and visual congruence in the process of identifying an entity as the self-avatar. The results indicated that motion synchrony between the user and the avatar is relatively more emphasized than the match in visual appearance when identifying oneself in VR. These findings highlight the underlying mechanisms and crucial factors for self-recognition in VR, and provide valuable insights for designing more immersive virtual experiences in various social VR applications.

Fashion stylists have historically bridged the gap between consumers’ desires and perfect outfits, which involve intricate combinations of colors, patterns, and materials. Although recent advancements in fashion recommendation systems have made strides in outfit compatibility prediction and complementary item retrieval, these systems rely heavily on pre-selected customer choices. Therefore, we introduce a groundbreaking approach to fashion recommendations: text-to-outfit retrieval task that generates a complete outfit set based solely on textual descriptions given by users. Our model is devised at three semantic levels-item, style, and outfit-where each level progressively aggregates data to form a coherent outfit recommendation based on textual input. Here, we leverage strategies similar to those in the contrastive language-image pretraining model to address the intricate-style matrix within the outfit sets. Using the Maryland Polyvore and Polyvore Outfit datasets, our approach significantly outperformed state-of-the-art models in text-video retrieval tasks, solidifying its effectiveness in the fashion recommendation domain. This research not only pioneers a new facet of fashion recommendation systems, but also introduces a method that captures the essence of individual style preferences through textual descriptions.

The monitoring of arterial blood pressure (ABP) in anesthetized patients is crucial for preventing hypotension, which can lead to adverse clinical outcomes. Several efforts have been devoted to develop artificial intelligence-based hypotension prediction indices. However, the use of such indices is limited because they may not provide a compelling interpretation of the association between the predictors and hypotension. Herein, an interpretable deep learning model is developed that forecasts hypotension occurrence 10 min before a given 90-s ABP record. Internal and external validations of the model performance show the area under the receiver operating characteristic curves of 0.9145 and 0.9035, respectively. Furthermore, the hypotension prediction mechanism can be physiologically interpreted using the predictors automatically generated from the proposed model for representing ABP trends. Finally, the applicability of a deep learning model with high accuracy is demonstrated, thus providing an interpretation of the association between ABP trends and hypotension in clinical practice.

Attention is known to be shifted reflexively by subliminal cues in static environments, but their effect when applied in dynamic environments remains unclear. This study examines the effect of subliminal cues in both static and dynamic environments and presents a novel technique of applying subliminal cues within videos. Experiment 1 confirmed the effect of subliminal cues in guiding covert spatial attention in a static environment. Experiment 2 investigated the effect of subliminal cues in guiding overall gaze distribution in video context by manipulating the frequency of subliminal cues to bias the viewer’s gaze towards a specific side. There was no main effect of cue frequency, but additional findings showed the possibility that the effect of subliminal cues occurred differently between gender, and other factors such as gaze orientation bias influenced the viewer’s gaze distribution. These results provide insights on application of subliminal cues in video contexts and render the directions for future studies.