Science of Emotion and Sensibility

Scarcity and probabilistic access frequently characterize modern consumption contexts, yet little is known about how consumers’ perceived chance of encountering a product influences their product experiences. The present study examines how perceived chance affects product enjoyment and investigates the sequential mediating roles of self-product connection and serendipity as underlying psychological mechanisms. We demonstrate that when a single product is presented, a low (vs. high) perceived chance of acquisition strengthens consumer self-product connection, which in turn generates feelings of serendipity - the sense that the product encounter was a personally meaningful and fortuitous event - ultimately leading to greater product enjoyment. In contrast, when multiple products are presented simultaneously, this sequential indirect effect attenuates, as concurrent presentation divides attentional resources and activates comparative processing, thereby undermining the translation of serendipity into product enjoyment. These findings contribute to previous research on scarcity, consumer identity, and assortment structure by identifying perceived chance as a novel antecedent of experiential value and by demonstrating that serendipity serves as a key mechanism linking self-product connection to enjoyment. This study offers practical guidance to marketers designing promotional campaigns, lotteries, and limited-access product experiences and reveals interdisciplinary implications for emotional design and user experience design in digital environments.

keyword : Perceived Chance, Self-Product Connection, Serendipity, Product Enjoyment, Assortment Size, Uncertainty, 지각된 획득 가능성, 자기-제품 연결, 의외성, 제품 즐거움, 구색 규모, 불확실성

Crossmodal correspondences between sensory modalities are often attributed to shared affective qualities, yet the structure and variability of such affective representations remain poorly understood. Building on previous research demonstrating crossmodal affective mapping between taste and music, the present study investigated whether these mappings are supported by a low-dimensional structure and whether they vary across directions and individuals. Using affective rating data from 28 participants, crossmodal classification was performed between gustatory and auditory stimuli. The results showed that classification accuracy exceeded chance level in both directions, indicating that affective information generalized between the examined gustatory and auditory stimuli. However, performance was significantly higher in the taste-to-music direction than in the music-to-taste direction, revealing a clear directional asymmetry. To examine the structure of affective representations, classification was repeated using the composite valence and arousal dimensions. While this low-dimensional model exhibited a comparable performance in the music-to-taste direction, it was less accurate in the taste-to-music direction, suggesting that valence and arousal capture the core structure of affect but are not sufficient in all contexts. Consistent with this interpretation, permutation-based feature importance analysis revealed no dominant contribution of individual affective descriptors, indicating that crossmodal mapping relies on distributed representations. Finally, substantial inter-individual variability was observed. Crossmodal performance was not significantly correlated across the two mapping directions, providing no clear evidence for a common performance factor across directions. Importantly, affective consistency was positively associated with performance in the taste-to-music direction, suggesting that the stability of affective representations contributes to crossmodal mapping under specific conditions. Collectively, these findings indicate that affective mapping between taste and music is directionally asymmetric and that valence and arousal capture an important but not universally sufficient component of crossmodal affective information. Exploratory individual difference analyses further indicate that the consistency of affective responses may contribute to mapping performance, although these findings require replication in larger samples.

keyword : Crossmodal Correspondences, Valence, Arousal, Taste, Music, Individual Differences

This study developed and implemented a liberal arts course titled “Heartwarming Stories” tailored to the emotional and experiential characteristics of adult learners, and analyzed its educational effectiveness as well as student experiences. Grounded in narrative learning theory and the ADDIE instructional design model, the course targeted core competencies in sharing and communication. It was designed as a 100% online curriculum integrating diverse narrative media―including historical narratives, films, dramas, animations, and documentaries―with emotional music. The course was operated over 15 weeks for 143 adult freshmen at University D. The data were analyzed using quantitative course evaluations and qualitative content analysis of students' reflective assignments. The course evaluation yielded a high overall adjusted score of 4.45 out of 5.0. Qualitative content analysis of reflective essays revealed four key themes: (1) reinterpretation of life and self-reflection, (2) recognition of the meaning of relationships and love, (3) healing after loss and integration of grief, and (4) gratitude for current life and commitment to future growth. These results suggest that the life experiences of adult learners can be transformed into meaningful learning resources through a narrative-based emotional curriculum, allowing them to re-examine their self-understanding, emotional recovery, and relational values.

keyword : Adult Learners, Narrative Learning, Emotional Stories, Liberal Arts Education, ADDIE Model, 성인 학습자, 서사 학습, 감성 스토리, 교양 교육, ADDIE 모형

This study examined whether category congruency between a preceding emotion label and a subsequent facial expression facilitates facial expression processing and whether this effect is reflected in event-related potentials (ERPs). A non-emotion label condition (“shape”) was included to distinguish facilitation by congruent labels from interference by incongruent labels. Participants viewed the words “anger,” “happiness,” or “shape” and classified the subsequent facial expression using preassigned number keys; thus, emotion labels did not serve as direct response categories. Results showed that responses were faster in the congruent condition than in the incongruent and control conditions, whereas responses did not differ between the incongruent and control conditions. ERP analyses did not reveal an overall congruency effect; however, exploratory follow-up comparisons indicated a smaller P1 amplitude in the left posterior temporal region of interest in the congruent condition. This pattern was observed for happy faces in the emotion-specific comparisons. The behavioral findings support selective facilitation by congruent labels rather than interference by incongruent labels. The exploratory P1 finding further suggests that emotion labels may prepare the processing of subsequent facial expressions by providing a conceptual context, which could affect early visual encoding.

keyword : Emotion Labels, Facial Emotion Judgment, Congruency Effect, Event-Related Potentials, P1, 정서 명칭, 얼굴표정, 일치성 효과, 사건관련전위

Expression mismatch negativity (EMMN) describes an event-related potential difference waveform elicited by irregular, task-irrelevant changes in facial expression. This waveform can jointly reflect physical stimulus differences, repetition-related adaptation, structural face processing, attention allocation, emotional salience, and regularity violations. Thus, the presence of EMMN alone does not establish a specific processing function or underlying mechanism. This narrative methodological review examines traditional, identity-varying, reverse, and sequence oddball paradigms, alongside equiprobable and cascade controls, with a particular focus on emotional specificity, latency, and predictive coding. The accumulated evidence indicates that identical-stimulus comparisons and controls for repetition-related adaptation support interpretations of EMMN in terms of regularity violation. Nevertheless, early responses can overlap more strongly with N170 and repetition-related effects, whereas later responses must also be interpreted in consideration of the paradigm and comparison condition. Emotional deviants might elicit larger or earlier mismatch responses than neutral changes, suggesting that EMMN reflects the influence of emotional salience beyond general change detection. With appropriate control designs and time-resolved analyses, EMMN can provide a useful neurophysiological measure of the temporal dynamics of facial expression change detection and socioemotional predictive processing, as well as differences associated with clinical characteristics and individual variability.

keyword : EMMN, Facial Expression, Oddball Paradigm, Repetition Adaptation, Emotional Specificity, 얼굴표정, 오드볼 패러다임, 반복 순응, 정서 특이성

Autonomous vehicle (AV) accidents involve the system's automation level alongside perception, judgment, and control processes. Current traffic accident reports fail to capture this technical context because they focus on physical evidence like vehicle position and collision type. This study addresses this gap by developing an icon-based visualization system for AV accident reports grounded in a previously proposed Core Information Framework covering vehicle, environmental, and human factors. We designed 26 icons across 10 types by integrating ISO standards (3864 and 7010), Gestalt principles, established information design theories, and police reporting conventions. Each icon's form was optimized for a 1 cm print size to reflect practical medium constraints. Twenty experts across relevant fields evaluated the icons using ISO 9186-2 criteria (meaningfulness, legibility, and semantic appropriateness) alongside perceived trustworthiness for official reporting. All measures scored above the scale midpoint (overall means 3.69-3.76), indicating that the system functions as a readable visual language. In particular, the communication, traffic infrastructure, and AV icons received high ratings, and all four expert groups rated the system consistently above 3.6 on average. Ultimately, the proposed system offers a structured visual language for AV accident documentation and demonstrates how practical media conditions and perceptual demands shape public information visualization.

keyword : Autonomous Vehicle Accident, Accident Investigation Report, Information Visualization, Core Information Framework, Icon System, 자율주행차사고, 사고조사서, 정보시각화, 핵심정보프레임워크, 아이콘시스템

This study aimed to develop “smart” sportswear knitted with strain sensors to detect repetitive elbow and knee flexion and extension movements and examine the relationship between sensor voltage signals and the actual number of repetitions under actual-use conditions. To this end, design criteria were established through theoretical and case studies, as well as analyses of body dimensions and garment deformation associated with joint movements, in order to design and fabricate upper- and lower-body smart sportswear. Subsequently, one adult female participant performed dumbbell curls and lunges; signals from the left and right elbow and knee sensors were analyzed. The results were as follows. First, joint-sensor positional alignment, alignment between garment deformation and sensor strain direction, close-fitting garment structure, sensor dimensions, structural integration, flexible signal transmission, and detachability of the electronic module were identified as design requirements. Second, knitted strain sensors measuring 55 x 165 mm for the posterior elbow region and 60 x 175 mm for the anterior knee region were directly knitted into the garments with zigzag-stitched lead wires and snap-type MCU connections. Third, all four sensors exhibited periodic signal changes, with the voltage decreasing during flexion and increasing during extension. Fourth, in one representative set, ten signal cycles corresponding to ten actual repetitions were visually distinguished for the left and right elbow and knee sensors. These results indicate that the knitted strain sensors could generate periodic voltage signals corresponding to repetitive elbow and knee flexions and extension movements during actual wear. This study is significant for implementing a garment-integrated knit strain sensor system, while considering joint location and clothing deformation direction. The applicability to repetitive movements involving different major joints was also explored. However, this was an exploratory application test with only one participant; and the raw signal cycles were visually examined without a separate reference device for joint-angle measurement or an automated detection algorithm. Therefore, the findings should be considered an exploration of the relationship between joint movement and sensor signals under limited conditions rather than a validation of repetition-counting accuracy or joint-angle measurement performance.

keyword : Knitted Strain Sensor, Smart Sportswear, Garment-integrated Sensor, Joint Movement Monitoring, 니트 스트레인 센서, 스마트 스포츠웨어, 의복 통합형 센서, 관절 동작 모니터링

Robot vacuum cleaners have become autonomous devices common in domestic environments. Their navigation behavior can serve as a non-verbal cue through which users interpret the robot’s human-likeness, personality, and intentions. This study examined user interpretations at two levels: i) scale-based perceptions of anthropomorphism and animacy and ii) attributional expressions in open-ended responses. Differences in robot vacuum behavior type and prior usage experience were examined. Behavior type (design-intent consistent vs. design-intent deviant) was experimentally manipulated, while prior usage experience (yes/no) was treated as a pre-measured individual-level-difference factor in an inter-subject design. Forty adults viewed 20-second navigation videos created using generative AI and then completed a questionnaire. The manipulation check showed that the design-intent deviant condition was perceived as significantly stranger than the design-intent consistent condition. The main analysis revealed that anthropomorphism and animacy scores were markedly lower in the design-intent deviant condition, whereas neither the main effect of usage experience nor the interaction between behavior type and usage experience was prominent. In open-ended responses, the design-intent consistent condition elicited social-role expressions such as “helper” and “butler,” whereas the design-intent deviant condition elicited expressions such as “stubborn,” “headstrong,” and “uncontrollable,” attributing independent will and personality to the robot. These findings suggest that scale-based perceptions of a robot’s human-likeness, animacy, and attributional expressions derived from its behavior may not necessarily follow the same pattern. Furthermore, robot vacuum cleaner navigation behavior may perform beyond functional movement as a behavioral interface through which users interpret the robot’s intentions and personality, underscoring the need to consider varied layers of user interpretation during the design of domestic-use autonomous robots.

keyword : Robot Vacuum Cleaner, Generative AI, Navigation Behavior, Human-Robot Interaction, Anthropomorphism, 로봇청소기, 생성형 AI, 주행 행동, 인간-로봇 상호작용, 의인화

This exploratory study compared subjective evaluations of selected human-written disaster warning messages with versions restructured using generative Artificial Intelligence (AI). In total, 18 messages covering 6 hypothetical scenarios―earthquakes, tsunamis, wildfires, landslides, floods, and major industrial accidents―were evaluated. Each included a human-written original and versions restructured using ChatGPT-5 and Gemini Advanced. Data from 41/42 participants were analyzed. Participants rated perceived comprehension, accuracy, reading speed, reading ease, and actionability using a five-point Likert scale. Repeated measures ANOVA was followed by Bonferroni-adjusted comparisons and Pearson correlation analyses. AI-restructured messages generally received higher subjective ratings than selected human-written originals, although differences varied by disaster type and measure. Significant "main" effects of disaster and message type, as well as marked interactions, were observed. Variations between AI models were limited, except that Gemini received higher ratings for perceived accuracy and actionability regarding the major industrial accident scenario. All five measures were positively correlated. However, the absence of a comparable human-revision condition prevented the separation of the effects of AI use from those of message revision. The small sample, hypothetical scenarios, and self-reported measures also limit the generalization to objective performance or actual protective behavior. These findings provide exploratory evidence that generative AI may support disaster warning message restructuring and inform future controlled studies utilizing human-revised messages and objective measures.

keyword : Disaster Warning Messages, Generative Artificial Intelligence, Risk Communication, Communication Effectiveness, Actionability, 재난문자, 생성형 인공지능, 위험 커뮤니케이션, 정보 전달 효과, 대처가능성