This study investigates how different front-of-pack nutrition labels and time constraints influence consumers’ ability to select healthier food products. Participants performed a forced-choice task in which they selected the healthier option from pairs of food items displayed under five label conditions: a control condition without nudging (Control), a set of simple pictorial health icons (Pict), a traffic-light signal (Traffic Light), the five-category NutriScore label (NutriScore), and a combination of pictorial icons with NutriScore (Pict+Nutri). The task was conducted under two time conditions. In the Short condition participants had 5 s to respond, whereas in the Normal condition there was 40 s. Decision quality (proportion of correct selections) and response time were analyzed using mixed-effects models and post hoc comparisons. The combined Pict+Nutri label significantly improved decision quality relative to the control (p = 0.035), and the NutriScore alone showed a marginally significant improvement (p = 0.065). Time pressure did not diminish decision quality nor interact with label condition, indicating that the NutriScore nudge retained its effectiveness even when participants had only 5 s to respond. As expected, reaction times were shorter in the Short condition (p < 0.001). The label design itself did not significantly change overall reaction times. These results suggest that front-of-pack labels containing NutriScore information can support healthier decisions without increasing decision time, and that pairing NutriScore with simple pictograms yields the largest improvement in decision quality.

Despite being one of the three main themes of sustainability (global environment, workers, and animal welfare), animal welfare remains significantly underrepresented in research. This study applied the knowledge of ingredient branding to animal welfare and posed the following question: “Does promoting the raising environment for livestock cows increase the perceived value of milk among consumers?” In the food industry, ingredient branding can be categorized into raw materials, origin, and manufacturing/processing techniques. A common feature of these methods is their clear objective. In areas remote from consumers, such as animal welfare, the gap between ethical consumption attitudes and behaviors cannot be bridged unless the logic behind the product value is clear. Therefore, we hypothesized that an emphasis on comfortable facilities for cows would increase the perceived value of milk more than an emphasis on love. The results of a randomized controlled trial conducted in Japan support this hypothesis. This effect was more pronounced among middle-aged and older individuals and among men. This is because women and younger generations place more importance on emotional empathy, favoring “love by farmers.” These results suggested that animal welfare has the potential to become a source of customer value. Furthermore, this study strengthens this theory by adding animal welfare to the knowledge of food ingredient branding.

The nighttime economy is expected to play an important role in sustainable development for three main reasons: (1) elimination of overtourism by dispersing consumption; (2) shifting economic activity to avoid extreme heat caused by climate change; and (3) reducing travel volume by creating compact cities with work, living, and leisure activities close together. One key to the success of the nighttime economy is the comfortable design of urban lighting. This is because light has the effect of changing the perceived value of products and services. Therefore, this study focused on warm colors and lanterns (Toro) as lighting that creates warmth and emotion. A randomized controlled trial was conducted on 750 people aged 20-60 at Fugan Canal Kansui Park in Toyama Prefecture, Japan. The results revealed that warm-colored lanterns enhance the attractiveness of urban parks. The effect was significantly greater among younger women, making this measure particularly effective when targeting young women. Traditional and cultural facilities such as temples contribute to the atmosphere of a city. Lanterns, which were introduced along with Buddhism, are thought to similarly create elegance and further enhance the effects of light color. Urban development often involves the introduction of measures that incur enormous costs, but this is hardly sustainable development. As this study has shown, it is important to make efforts to increase the attractiveness of a city with as few resources as possible. Practitioners should recognize that the attractiveness of a city can easily be changed by the relatively low-cost resource of light.

The rapid expansion of e-commerce has reshaped the luxury market, making digital channels strategically unavoidable. While prior research highlights risks that online distribution poses to luxury values such as exclusivity and symbolic meaning, empirical evidence linking channel-specific assortment strategies to brand value remains limited. This study analyses the relationship between e-commerce assortment deployment and brand value, focusing on official (brand-owned) e-commerce (hereafter, official e-commerce) versus third-party e-commerce. The analysis covers seven leading luxury brands—Louis Vuitton, Hermès, Chanel, Gucci, Dior, Prada, and Fendi—selected from the BrandZ 2025 global top-ten luxury ranking. Data were collected from brands’ official Japanese e-commerce sites and six major global third-party platforms. Brand value was operationalised using BrandZ scores. A two-stage approach was employed: principal component analysis (PCA) to identify latent patterns of assortment deployment, followed by fuzzy-set qualitative comparative analysis (fsQCA) to examine configurations associated with high brand value. PCA reveals two dimensions: a “third-party e-commerce restraint–expansion” axis and an “entry expansion–selective assortment” axis within official e-commerce. fsQCA indicates that high brand value is achieved not by any single factor but by restraint in third-party e-commerce SKUs and discounting. These findings suggest that luxury brand value in digital environments depends less on quantitative expansion than on strategic control over assortments and channels, offering implications for managing official and third-party e-commerce in luxury contexts.

Japan’s Natural Disaster Memorial Monument Database (GSI) enables nationwide examination of how communities externalize disaster lessons in everyday places as cues that are perceptible, emotionally resonant, and repeatedly encountered. Using 2,098 monuments with both a construction year and a recorded target disaster year (of 2,377 entries; July 2025), we computed the year difference (construction year − recorded target disaster year). Most differences fall within 0-99 years (94.57%) and are concentrated within the first decade, indicating that many monuments are recorded as being constructed soon after disasters. Five monuments (0.24%) have negative year differences-constituting all negative-year cases in the analytic sample-an expectation-violating pattern in which the construction year predates the disaster year linked in the database and therefore requires interpretation. Document review of all five cases and fieldwork at two sites suggest two mechanisms: re-framing pre-existing monuments and representative-year selection within multi-event disaster histories. Negative year differences may thus reflect layered, updateable memory rather than simple errors. Treating such mismatches as analytic objects reframes monuments as dynamic material witnesses, highlights the role of visibility and repeated encounters in place-based risk-reduction education, and motivates metadata that can represent evolving narratives.

In recent years, as multicultural communities have continued to expand in Japan, the question of how to implement effective disaster education in communities where residents with diverse cultural backgrounds live together has become an increasingly important issue. This study examines and analyzes a disaster education practice centered on participatory disaster mapping, conducted collaboratively by Japanese residents and foreign residents in the Yoshizuka area of Hakata Ward, Fukuoka City. Drawing on findings obtained through fieldwork and interviews, the study finds that, through practices such as disaster map creation and a town-walk, dialogue around disaster-related issues not only enhanced participants’ understanding of disaster risks and preparedness measures, but also promoted multicultural understanding and the formation of interpersonal relationships among residents. Furthermore, the disaster maps co-created through this process stimulated residents’ sense of agency and responsibility toward disaster preparedness, thereby providing a foundation for sustained collaboration within the community.

In recent years, traditional crafts have seen reduced opportunities for daily use due to changes in social structures and lifestyles. To promote the inheritance of traditional culture, Hiwadabuki patterned panels were developed. These panels utilize Hiwadabuki (cypress bark roofing), one of Japan's traditional architectural techniques-to recreate its unique visual rhythm and texture. The purpose of this study is to consider the Hiwadabuki pattern as an example of a design symbolizing traditional skills and to clarify the characteristics of preferred frames for the development of art panels featuring this pattern. Using frames in 5 colors and 5 widths, we investigated preferences regarding color and width. Results showed no significant influence from color differences. Regarding frame width, a preference for 13mm was consistently observed across all participants. This preference is thought to be related to the harmony achieved with the periodic pattern inherent in the Hiwadabuki patterned panels.

This study examines how Japanese municipalities institutionally position direct evacuation to welfare shelters, focusing on human-centered decision-making under disaster conditions. A document survey of municipalities in Nankai Trough earthquake countermeasure areas identified three policy categories: explicitly permitted, conditionally permitted, and not permitted. A multinomial logit model revealed limited but statistically significant associations between municipal characteristics and policy positioning. Municipal area size was positively associated with explicit permission, while the number of residents requiring evacuation support was negatively associated with conditional permission. Overall, structural capacity alone does not fully explain institutional decisions. Interviews indicate that experiential factors, including disaster experience and administrative judgment, play critical roles.

Low-speed mobility devices (LMDs), such as electric wheelchairs, are important transportation options for elderly users. However, environmental factors during outdoor use can cause user stress. Evaluating stress and identifying stress-prone locations is essential for stress reduction. Heart rate variability (HRV) is widely used for stress evaluation, However, conventional short-term HRV analysis typically requires a time window of approximately five minutes, which limits its applicability for identifying stress within short-distance segments. Recently, ultra-short-term HRV (UST-HRV) analysis using windows shorter than five minutes has been proposed and has shown potential for stress evaluation. However, its effectiveness for evaluating stress during LMD rides has not been sufficiently investigated. This study examines whether stress during LMD rides can be evaluated using UST-HRV indices derived from 30 s and 60 s time windows. A field experiment using an electric wheelchair was conducted in an urban environment, collecting HRV, GPS, environmental data, video, and questionnaires. The experimental route was divided into 50 m and 100 m segments to ensure HRV measurement durations of 30 s and 60 s, respectively. Time-domain UST-HRV indices (SDNN, RMSSD, and pNN50) were calculated for each segment and compared between high- and low-stress-potential segments. The results showed that RMSSD was significantly lower in high-stress-potential segments for both segment lengths, whereas SDNN showed no significant differences and pNN50 differed only under limited conditions. These findings suggest that RMSSD derived from 30 s and 60 s time windows can be used as an indicator for evaluating stress during LMD rides over short-distance segments.

In design thinking education, fostering collaborative skills while sustaining learner engagement remains a persistent challenge, particularly in short-term educational activities. Collaborative performance in design tasks is commonly understood as consisting of core elements such as role allocation, decision-making, and time management, which together support effective teamwork. In addition to these cognitive aspects, affective factors, including engagement and psychological safety, play essential roles in creative learning processes and influence how deeply learners participate in collaborative activities. The objective of this study was to examine whether a short-term, edutainment-oriented design thinking exercise could enhance learner engagement and foster awareness of collaborative processes. A comparative study was conducted between MANGATHON-a short-term group activity in which teams collaboratively create four-panel comics by following structured guides and compete by sharing and evaluating their outcomes-and a conventional video-based instructional approach using both quantitative and qualitative evaluation methods. The results indicated that while teamwork comprehension scores were similar between the two approaches, engagement-as measured by the Net Promoter Score-was substantially higher for MANGATHON. Additionally, analysis of questionnaire responses revealed a marked increase in references to time management during teamwork. These findings suggest that short-term edutainment-oriented design thinking exercises using four-panel comics can substantially enhance affective engagement and promote experiential awareness of collaborative processes. This study provides practical guidance for applying lightweight, activity-based design thinking interventions in time-constrained educational and professional training contexts.

This paper reports on the outcomes of the engineering workshops and presents the current status and future prospects of regional community-building through community-linked teaching material development. In a society where the birthrate is declining and the population is aging, local industries face a serious labor shortage. Reports also show that many young people lose interest in science and engineering, take less part in local communities, and feel a low sense of belonging. These trends highlight a lack of educational approaches that build engineering skills and support regional community building by linking young people with regional industries. To address such challenges, we developed Exclamatone, an original electronic musical instrument teaching material, and implemented engineering workshops for young learners. The workshops provided hands-on, cross-disciplinary learning experiences in electronic circuits, programming, and design. Its goal was to cultivate engineering thinking skills, creativity, and sustain interest in engineering. Based on these outcomes, we are now expanding Exclamatone into a community-based engineering teaching material through industry-academia-government collaboration with local industries and municipalities. This initiative seeks to enhance young people’s place engagement, defined as interest in local industries and technical careers, and willingness to support the local community by creating meaningful opportunities to engage with engineering in a regional context.

Digital textbooks are expected to be officially implemented in Japan by 2030; however, their effects on learners’ concentration and comprehension remain unclear. Previous studies comparing paper-based and digital reading have reported mixed findings, and few have assessed concentration using physiological measures. This study examined how printed and digital textbooks influence learners’ behavior, subjective evaluation, concentration, and learning performance during reading and answering tasks. A within-subject experiment was conducted with 21 university students, each of whom used both printed and digital versions of an English textbook. Measures included in-text question accuracy, electroencephalography (EEG; frontal midline theta [Fmθ]), eye-tracking data, subjective questionnaires, and comprehension tests to evaluate concentration, usability, and learning effectiveness. Results indicated significantly higher accuracy on in-text questions in the printed condition, whereas delayed retention did not differ between formats. EEG analysis revealed higher Fmθ power during the answering phase in the printed condition, suggesting greater task engagement. Eye-tracking observations showed that the digital textbook required additional operational steps, such as zooming and activating writing tools, which frequently caused interaction errors and disrupted task flow. These findings suggest that printed textbooks promote more focused engagement during learning tasks, whereas digital textbooks may impose operational demands that interfere with concentration. Implications for improving the interaction design of digital learning materials are discussed.

This study examined the impact of overnight training sessions at evacuation shelters and subsequent evaluations by participants on discussions regarding issues within the shelter environment. Concretely, drills were conducted at a high school in the Nishikino area of Kuroshio Town, Kochi Prefecture, in both the peak summer and peak winter season. Most participants in the drills were high school students and teachers, so interviews and questionnaires were conducted with them. Additionally, their behavior during drills was analyzed through sequential evaluations using "Comfort level chart" and video recordings. As a consideration, it was implied that overnight drills conducted under unconventional conditions—specifically during the peak winter and peak summer seasons—may have contributed to identifying specific issues related to shelter environments.

Color Maps for astronomical visualization must balance scientific clarity and affective engagement, particularly in education and outreach. We conducted a two‑phase evaluation using ALMA data of a star‑forming region. Survey‑1 (A-L; N=23) used Kansei labels to identify preferred and problematic palettes. Survey‑2 (O-W; N=36) rated six scales and collected attributes. Scale‑wise means and k‑means (k=3) revealed three segments. Inferno (U) and Gold-Black (R) ranked highly; Hot-Cold (S, reversed) showed strong memorability with low negatives. Significant correlations (p < .05) linked gender with friendliness, design experience with beautiful/interest, and astronomy interest with misunderstanding. We interpret these findings in light of the perceptual‑uniform principles underlying U/W/P and propose audience‑aware Color Map strategies for education, exhibits/VR, and public communication.

This study investigates how sinusoidal surface patterns applied to cylindrical grips influence haptic impressions during comfort grasping. Thirteen samples were prepared by combining four wavelength levels and three amplitude levels of sinusoidal profiles on a 150-mm-long, 30-mm-diameter cylinder, along with a flat control grip. Twelve participants grasped each sample with one hand and rated seven questionnaire items-painfulness, hardness, smoothness, softness, ease of grip, force applicability, and comfort- using a nine-point scale. A repeated-measures analysis of variance showed significant main effects of sample type for all items, indicating that both grip profiles modulated impressions. To clarify the underlying perceptual structure, principal component analysis (PCA) was performed on z-standardized ratings averaged across participants. The PCA biplot revealed two nearly orthogonal bundles of loadings: one contrasting “painful” with “smoothness” and “ease of grip,” and another contrasting “hardness” with “softness,” “comfort,” and “force applicability.” These bundles represent two independent axes of perceptual variation induced by sinusoidal profiles-one related to discomfort versus smooth feel, and the other related to hardness versus soft and comfortable grip feel.

Mechanical vibration applied to the upper body effectively elicits physiological arousal and arousal-related emotions in emotionally charged contexts such as movies. Moreover, vibratory stimuli triggered by one’s own electrophysiological signals are suitable for interactive content such as video games. In previous work, this biofeedback framework was extended to dyadic settings in which vibratory feedback was mutually triggered by a partner’s physiological arousal; however, emotional effects were not confirmed. In this study, we redesigned the experimental protocol, including the gamified task and the questionnaire assessing empathy toward the confederate. Eight participants were paired and played a cooperative action game under a mutually vibratory biofeedback condition. Post-experiment questionnaires indicated that participants reported higher levels of tension, excitement, and arousal when vibratory biofeedback was present. Furthermore, they gave higher ratings for both the cognitive and emotional components of empathy. These findings suggest that the proposed biofeedback scheme may serve as an effective method for enhancing shared emotional experiences and interpersonal empathy in interactive content.

This study proposes a vibratory biofeedback system applied to the thoracoabdominal region and triggered by facial expression recognition to amplify users’ emotional experiences. Mechanical vibration to the upper body is known to enhance bodily arousal, which in turn influences arousal-related emotions. Based on this, the proposed system synchronizes vibratory feedback with users’ arousal responses inferred from facial expressions, enabling more immediate stimulus presentation than prior skin-conductance-based approaches and reducing temporal delays between emotional reactions and feedback. The framework was evaluated in a computer game context with six participants. Subjective evaluations showed significantly higher ratings for excitation and tense under the vibration conditions, both representative arousal-related emotions, whereas no significant effect was observed for frustration. These findings suggest the potential of the proposed method to enhance arousal-related emotional experiences.

In this study, a machine learning approach with Random Forest (RF) was performed to predict color emotion of two-color combinations by naturally dyed cotton fabrics with persimmon and indigo. A total of 9 stimuli consisting of two pieces of cotton woven dyed with persimmon and indigo respectively were evaluated by female college students for primary color emotion terms such as ‘Warm’, ‘Soft’, ‘Light’, and ‘Strong’ and composite terms including ‘Pleasant’, ‘Traditional’, ‘Elegant’, and ‘Dynamic’ using 7-point semantic differential scales. The composite color emotion terms were revealed as rarely correlated with single color emotion and sum of them while primary terms showed significant positive correlation coefficients, which means emotional additivity for color combinations could be limited to primary color emotion. RF algorithm was conducted by obtaining permutation importance of each color emotion surrogate models were constructed to obtain interpretable analytical expressions. The results revealed that ΣL and Δb as color combination variables were the most influential variables for ‘Traditional’ as well as physical lightness of persimmon dyed cotton. ‘Light’ and ‘Elegant’ had better performance to establish their prediction models as well by employing color combination variables and single color emotion scores. In contrast, ‘Pleasant’ and ‘Warm’ showed lower performance compared to the other emotions. The results could be helpful to understand complexity and contexts of color emotion for multi-colored textiles.

Retro fashion has gained renewed popularity among young consumers; however, the factors that generate a sense of novelty in contemporary retro fashion remain unclear. This study aims to clarify how design elements and styling methods influence perceived novelty using a Kansei engineering approach. Participants evaluated the perceived novelty of 10 fashion images using a seven-point semantic differential scale. Two surveys were conducted: the first involved participants aged from their 20s to 70s, and the second focused on female participants in their 20s and 50-60s to examine the relationship between perceived novelty and past experience. Statistical analyses revealed significant age-related differences in perceived novelty. In particular, participants in their 20s showed higher sensitivity to novelty than those in their 50s and 60s. Furthermore, silhouette, styling methods, and the combination of different fashion tastes were identified as key factors influencing perceived novelty. Notably, even when garment silhouettes were similar, differences in styling methods led to substantial changes in novelty evaluation. These results suggest that perceived novelty in retro fashion can be enhanced through styling and coordination without altering garment structure. This finding highlights the potential of styling-based approaches for sustainable fashion reuse and upcycling.

This study proposes a region-specific puffiness measure to analyze the silhouette features of Kawaii fashion. The proposed method quantifies region-specific puffiness by dividing the body into upper and lower regions at a body-proportion-based waistline. The datasets used in this study contain five Kawaii fashion styles and six luxury fashion brands. For the luxury fashion brands, outfits were manually categorized as either Typical-style or Kawaii-style. The results show that the five Kawaii fashion styles differ primarily in lower-body puffiness: Lolita exhibits the most voluminous silhouette, while Harajuku-type Kawaii presents a comparatively slimmer silhouette. Among luxury brands, Kawaii-style outfits generally show greater lower-body puffiness than Typical-style outfits, and some brands also show increased upper-body puffiness in their Kawaii-style outfits. These findings suggest that Kawaii fashion is not defined by simply enlarging the entire silhouette; rather, its expression varies by brand and by body region.

This research presents a comprehensive feature extraction framework for predicting perceived sustainability in fashion garments. Building upon our prior work that established color as a measurable signal of sustainability perception, this study significantly expands the analytical scope to include texture, shape, and deep learning features. A dataset of 1,080 product images across six garment categories (shirts, T-shirts, jackets, pants, shorts, skirts) was collected from Farfetch and Fashion Product Images Dataset, featuring isolated garments under controlled lighting conditions. A tournament-style selection experiment with 20 participants collected preference data using a composite scoring methodology combining popularity and retention metrics. The framework extracts 342 features per image organized into three categories: color features including colorfulness, brightness, saturation, dominant colors, and earth-tone detection; texture features using GLCM, LBP, Gabor filters, and Tamura descriptors; and shape features via Hu moments, contour analysis, Fourier descriptors, and curvature analysis. Correlation analysis reveals that color entropy (r = −0.297), pattern intensity (r = −0.251), and colorfulness (r = −0.258) are among the strongest predictors, indicating that simpler, less colorful garments are perceived as more sustainable.

Conventionally, impression quantification through review analysis relies on frequently appearing impression words, making it difficult to directly estimate impression scales that appear infrequently in review texts. This study proposes a method to predict impression evaluation scores for low-frequency scales in apparel products by leveraging existing scores for four frequent scales (“Dark- Bright,” “Tight-Loose,” “Formal-Casual,” and “Simple-Fancy”). Specifically, four high-frequency scales were vectorized using BERT with input sentence templates to construct semantic axes reflecting the fashion context. The position of the target scale on each semantic axis was calculated using cosine similarity, and the final score was estimated by applying weights based on the resulting direction and scores of the high-frequency scales. Using “Muted-Vivid” as an example, the estimated scores were shown to reflect semantically related scales, such as “Dark-Bright.” This method enables consistent score estimation across diverse impression scales, regardless of their frequency of occurrence in reviews.

Previous studies have reported that artificial intelligence (AI)-generated book endorsements may be more effective than human-generated endorsements in enhancing consumer purchase intentions. However, the factors underlying these effects have not been sufficiently clarified. Focusing on stylistic characteristics that differ across generations is a promising approach to investigate these factors. First, this study performed a comparative analysis of AI- and human-generated book endorsements, examining lexical, syntactic, and emotional features. The results indicate that AI-generated endorsements tend to employ more formulaic expressions, more concise sentence structures, and a higher frequency of emotional words, whereas human-created endorsements exhibit more diverse lexical choices, longer sentence structures, and a more restrained use of emotional words.

This study proposes a method to measure consumer lifestyle consciousness through from online reviews in the context of purchases by analyzing online reviews, focusing on wristwatches. Data on 100,248 watches and corresponding 929,240 reviews were extracted from the Amazon Reviews 2023 dataset. From multi-country questionnaires, we derived seed words for nine lifestyle consciousness factors using text preprocessing, synonym expansion, and SBERT-based relevance scoring. Next, a 200-dimensional Word2Vec model was trained on the watch-review corpus and a centroid vector was computed for each consciousness factor by aggregating its seed-word embeddings based on Distributed Dictionary Representation (DDR). We then calculated cosine similarities between centroids to examine contextual similarities among factors and found multiple positive similarity pairs, implying that these consciousness concepts may share review contexts rather than being fully independent. This approach enables marketers and designers to capture multidimensional consumer psychological tendencies from e-commerce data without relying on questionnaire surveys, providing actionable insights for developing marketing strategies.

Facial expressions are fundamental nonverbal cues for conveying human emotional states, and their automatic analysis has become an important research topic in affective engineering and human-computer interaction. The Facial Action Coding System (FACS) provides an objective and quantitative framework for describing facial expressions on the basis of facial muscle movements, known as action units (AUs). However, the extent to which emotional states estimated from AU information correspond to subjective human perceptions has not been sufficiently examined. In this study, we developed a regression model that estimates valence-arousal coordinates from AU intensities using Russell’s circumplex model of affect. To evaluate the validity of the estimated emotional states, a subjective evaluation experiment using the emotion circumplex was conducted, and the agreement between the model predictions and participants’ judgments was quantitatively analyzed. The results show that the proposed AU-based model achieves stable estimation performance, and the level of agreement varies across emotional directions. The method developed in this study presents a human- centered framework for examining the psychological validity of AU-based emotion estimation models.

Social cues, which function as extensions of emotions, play a critical role in facilitating interpersonal communication. However, existing tools for augmented emotion representation face limitations in terms of usability and computational complexity. We propose a visual tool design that represents five basic emotions and examines the minimum number of parameters necessary to model them. Informed by a previous workshop, we focus on modeling emotions by adjusting the irregularity and complexity of shapes. We found that the parameters could be reduced to noise and pace through logistic regression analysis, as supported by their distributions for selected parameter values. We found that pace captured arousal, while noise captured only certain aspects of valence. This study concludes that a third parameter is necessary to fully distinguish positive and negative emotions.

The Temporal Dominance of Sensations (TDS) method is a sensory evaluation technique that records time-varying changes in taste, aroma, and texture during food consumption, and is often used together with the Temporal Liking (TL) method, which records the temporal evolution of hedonic responses. Mathematical models that link sensory dynamics to liking dynamics are useful for identifying relationships between sensations and preference. Therefore, predictive models that accurately estimate TL curves from TDS curves are required. In this study, three models-reservoir computing, principal motion analysis, and the vector autoregressive model-were compared in terms of prediction accuracy. These models differ with respect to linearity versus nonlinearity and whether temporal causality is satisfied. A comparative evaluation based on the root mean square error (RMSE) of prediction revealed that the Reservoir Computing model, which is nonlinear and temporally causal, achieved the best performance, with an RMSE of 0.182 for TL curves on a 9-point scale.

Sensory experience during food consumption evolves over time, and preference can also change dynamically. Although Temporal Dominance of Sensations (TDS) and Temporal Liking (TL), when recorded in the same session, allow sensory transitions and preference changes to be aligned on a common time axis, it remains difficult to describe, in a direct and time-resolved manner, how individual sensory attributes relate to subsequent preference. This study proposes a dynamic sensitivity analysis that uses a reservoir computing model to link TDS and TL data. A perturbation-based analysis is conducted to examine how the predicted TL curve dynamically changes in response to a slight increase in a sensory attribute at an arbitrary time point. When applied to a strawberry TDS/TL dataset, the proposed method produced clear and interpretable sensitivity curves across sensory attributes, capturing both stable directional tendencies and time-dependent changes in influence. These results demonstrate that the proposed approach provides a practical means of characterizing when and in which direction individual sensory attributes are associated with shifts in temporal preference trajectories.

This research presents a human-centered approach for modeling shape similarity among procedurally generated implicit tree models rendered by ray marching. This procedural implicit representation supports the generation of large and diverse collections of tree shapes. However, organizing and comparing these shapes using conventional feature-based distances often leads to discrepancies with human perception. Such distances are computationally efficient, yet they do not necessarily reflect how humans intuitively judge similarity between complex natural forms. To address this issue, we adopt a relative evaluation paradigm based on triplet comparisons. Human participants are shown a reference tree and two candidate trees, and are asked to select which candidate is more similar to the reference. These relative judgments provide stable and interpretable information about perceptual similarity without requiring absolute similarity scores. From the collected triplet data, we estimate a human perceptual distance over a small set of representative tree shapes. We evaluate the proposed method using Precision@k, a retrieval-oriented metric that measures neighborhood agreement with human judgments. Experimental results show that the calibrated distance consistently improves retrieval consistency compared to the original feature distance. The proposed framework is interpretable, scalable, and applicable to a wide range of procedurally generated 3D shapes beyond tree models.

High-performance motorsport places immense physiological demands on the cardiovascular system, requiring robust physiological regulation. While our previous research successfully utilized Electroencephalography (EEG) to identify driver potential based on neural efficiency, relying solely on neural metrics is insufficient due to EEG's susceptibility to noise and artifacts in dynamic driving environments. To address this, we propose integrating Electrocardiography (ECG) to provide multimodal information that creates a more comprehensive assessment. This study validates the use of ECG alongside established EEG benchmarks to distinguish driver potential. Using experimental data from 15 participants on the Silverstone Circuit, ECG indices (SDRR, RMSSD, pNN50) were analyzed. The results revealed that ECG metrics successfully distinguished between “Potential” and “No Potential” groups, reinforcing the neural classifications with data. Specifically, “Potential” drivers exhibited superior physiological regulation (significantly higher SDRR and RMSSD), indicating a greater physiological reserve. Conversely, the “No Potential” group showed reduced Heart Rate Variability (HRV), reflecting higher stress responses. These findings confirm that integrating ECG data establishes a robust multimodal physiological approach, enhancing the reliability of driver potential identification in high-performance motorsport.

Virtual reality (VR) horror games provide immersive environments that can reliably elicit fear responses, making them suitable for physiological-based fear estimation. However, it remains unclear whether estimation models can reliably generalize to unseen users and across diverse VR games. This study proposed a transformer-based fear estimation model using heart rate variability and respiration signals to examine its generalization performance. Experiments were conducted on the VRMN-bD dataset collected from three VR horror games. All experiments in this study were conducted under a unified subject-independent evaluation protocol using leave-one-subject-out cross-validation, with time-series-based splits applied to prevent temporal information leakage. Performance was assessed using the macro F1-score. Under this subject-independent setting, model performance was evaluated in terms of within-game performance, cross-game generalization, and individual differences. Specifically, within-game evaluation achieved macro F1-scores of 0.630, 0.567, and 0.834 for the three VR horror games, indicating performance differences across fear stimuli. In addition, cross-game hold-out evaluations were performed by training the model on one game and testing it on another, achieving macro F1-scores above 0.60 across all game combinations, which suggests a certain degree of game-independent generalization. In addition, leave-one-subject-out evaluation was performed, revealing substantial variability among individuals and highlighting the impact of physiological differences on fear estimation accuracy.

For people to be convinced, the process is crucial. The reason is that what motivates human action is the mental state of being able to believe in the content. However, in the rapidly growing field of Generative AI (GenAI), the “importance of process” has not been sufficiently applied. Consequently, GenAI responses lack persuasiveness. While demand for Explainable AI (XAI) has grown in recent years, research often remains biased toward technical perspectives, neglecting studies on perceptual quality. Thus, we formulated the research question: “How does carefully communicating the process leading to a result impact the perceived reliability of GenAI responses?” We conducted a randomized controlled trial in an online environment, dividing 800 working men and women into two groups: one presented only with the answer, and the other presented with the answer along with a detailed explanation of the reasoning process. The results demonstrated that in conversational GenAI, carefully explaining the process behind an answer increase both the user's conviction and appeal toward that answer. This finding complements the current focus on technical development in the business-critical domain of GenAI by introducing the perspective of perceived quality.

In this study, estimation models for tourist attraction impressions are developed based on tourist attraction-specific linguistic features extracted from online review data. An impression evaluation experiment for tourist attractions is conducted to clarify their semantic space. Factor analysis of the experimental results yielded four factors. Adjective-noun pairs are created to represent attraction impressions by extracting nouns and adjectives modifying the nouns for specific locations’ tourist attraction online review data. Based on the indices explaining the importance of adjective-noun pairs for each attraction, latent semantic analysis is conducted to define the linguistic features of each tourist attraction. Multiple regression models and neural network models are constructed using linguistic features to estimate factor scores. The neural network models achieved sufficient accuracy in impression estimation.

As digital technologies become more prevalent in education, students increasingly rely on AI- and cloud-based tools to support note taking and review. These tools can automate information capture and improve information retrieval, but they may also introduce distractions, alter classroom interactions, and reduce critical thinking, creativity, and cognitive engagement. Prior research suggests that learning tools should supplement rather than replace active learning. At the same time, traditional note taking can impose high cognitive demands during lectures. These challenges highlight the need for a hybrid approach. This study explores students’ behaviors, cognitive load, and preferences related to note taking through semi-structured interviews. The findings will be analyzed to inform the design of a supplementary learning tool that balances automation with student engagement, supporting more effective learning experiences.

Research aimed at enhancing human motor performance in virtual reality (VR) has gained increasing attention in recent years. This study quantitatively measured perceptual thresholds for changes in the motion gain of virtual hands using the psychophysical method of constant stimuli. Across ten participants, the 84% detection thresholds were 1.24 for gain expansion and 0.77 for gain shrinkage, indicating nearly symmetric sensitivity around a gain of 1.0. These results suggest that humans exhibit comparable perceptual sensitivity to both expansion and reduction of movement gains, providing a fundamental reference for manipulating motor behavior implicitly and naturally in VR environments.

Drone Rider is a previously developed XR flight-experience system that integrates a footplate suspended by multiple bungee cords, vibration feedback, and wind feedback proportional to flight speed. In this study, we conducted a preliminary experiment to examine whether Drone Rider can function as an exercise-gaming system that combines enjoyment with physical exertion. Participants experienced four virtual fields (Free, Crane, Birds, Gate) and rated their enjoyment after each trial. Heart rate during each 2-minute experience, as well as resting heart rate before each trial, was recorded as an index of physical load. A multi-objective analysis based on enjoyment ratings and heart-rate change suggested that the Birds and Gate fields achieved relatively favorable balances between enjoyment and exertion. These results indicate that goal attainment and enhanced immersion may play important roles in realizing Drone Rider as an effective exercise-gaming system.

In human-machine collaboration, external assistance can improve task performance but may also diminish the agent’s sense of agency (SoA). SoA can be defined as the feeling of initiation and control over one’s actions and their consequences. This study examines how the SoA is affected by external assistance within different task goals, specifically when assistance substitutes for the agent’s own information‑gain process. In this regard, two variants of a multi‑armed bandit task that share a common generative process but differ in the nature of their goal (exploitative versus explorative) were implemented. The results suggest that under assistance, goal‑dependent divergence emerge. In the explorative condition, increasing information disclosure significantly decreased SoA whereas in the exploitative condition, SoA was not weakened across increasing information disclosure levels. These findings highlight that assistance which delivers knowledge exogenously undermines perceived epistemic progress when success depends on belief improvement. However, it is less detrimental when success depends on producing preferred outcomes through pragmatic action. These results motivate goal‑sensitive principles for the affective engineering of assistive systems that preserve the SoA.

In recent years, Japanese retailers have accelerated the adoption of self-checkout systems to address labor shortages, yet approximately 30% of users express anxiety about system errors. Particularly, the “instruction inconsistency state”—where errors occur despite following instructions—inhibits users' sense of self-efficacy and becomes a factor causing significant psychological burden. This study focuses on the “sense of control”—the feeling that users are leading the operation—and aims to clarify what information presentation designs suppress stress during error notification scenarios. Based on prior research, five design factors were extracted, and an experimental model was constructed. An evaluation experiment was conducted using 16 types of UI designs, and the influence of each factor was verified through multiple regression analysis. The analysis revealed that “Flow Visualization” is the most dominant factor in enhancing the sense of control. Furthermore, it was demonstrated that visualization of flow significantly reduces user stress levels, either through the mediating effect of the sense of control or directly.

As society becomes more globalized, foreign language education has become essential, and some users find it challenging to maintain motivation for learning. We developed a Chinese vocabulary learning application using a VR environment and investigated the relationship between the effectiveness of virtual rewards and learning motivation. The results revealed that differences in motivational factors between experienced and inexperienced language learners. Therefore, to further improve user motivation and usability, it is necessary to provide users with an appropriate rewards and gamification elements suited to their ability. To achieve this, predicting learners’ motivation levels and automatically adjusting rewards and gamification elements can be an effective approach. To provide a more comfortable learning experience for users, this study predicted user motivation levels using an LLM based on trial data from the developed Chinese vocabulary learning application. The results showed that past prediction data accumulated at Level 3, leading to improved prediction accuracy. However, the accuracy rate by level remains low, necessitating the exploration of methods to improve accuracy. For future development, the study aims to evaluate multiple prediction models and implement the most accurate one into the application based on the predicted learning motivation values.

This study explores a tangible feedback approach to enhance users’ awareness of spending during mobile payment interactions. We present an airbag-based device that dynamically changes its volume in response to simulated purchase actions, enabling users to physically perceive both the increase and decrease of virtual monetary balance. The improved prototype supports continuous inflation and deflation, allowing the full transition of the airbag’s form to be experienced before and after a transaction. Connected to a smartphone interface via a host PC, the system delivers real-time physical feedback synchronized with user actions. Through this design, the study highlights the potential of embodied, continuous tactile cues to influence spending perception in cashless payment contexts.

Sensory processing sensitivity (SPS) has been linked to heightened responsiveness to external input, yet its neurophysiological basis remains unclear. In this study, we examined whether local phase synchrony correlates with individual differences in SPS. We evaluated SPS in 188 healthy adults using the Japanese version of the Highly Sensitive Person Scale (HSPS- J19), originally proposed by Aron and Aron (1997). We also estimated the phase-locking value (PLV) of EEG during resting state. We observed a localized, frequency-band-specific association in a right frontotemporal patch (the triangular region defined by the three adjacent electrode lattice points F4-C4-T4): PLV was positively correlated with the ease of excitation subscale, specifically in the theta and beta bands (false discovery rate controlled). No effects survived in the left homolog, other frequency bands, or other SPS subscales. This pattern suggests that individuals with higher SPS exhibit increased right frontotemporal synchrony at rest, consistent with increased salience detection and control initiation and with maintained top-down prediction and sensory gating. These findings identify localized, band-specific synchrony as a potential electrophysiological signature of SPS.

This study investigates the predictive performance of models based on a data-driven approach in comparison to those grounded in human attention mechanisms. Participants were asked to identify aesthetically appealing areas within images of 3D models, and their selections were used to train a supervised learning model. The predictive accuracy of this data-driven model was evaluated against conventional human attention models using recall rates as the primary metric. Results indicate that the data-driven approach significantly outperforms traditional attention models. To address scenarios with limited training data, we further examined the effect of combining sparsely trained models with attention model. This hybrid approach yielded improved recall rates over attention model alone, even under data-scarce conditions. While attentional models offer interpretability and are grounded in theories of bottom-up visual attention, they fall short in modeling top-down attention and reflective aesthetic judgments. This limitation may stem from their basis in pre-attentive visual processing, which does not fully account for intentional or reflective visual selections. Our findings suggest that integrating even modest amounts of participant-generated training data with attention models can enhance predictive performance. Further research is needed to explore effective strategies for integrating and enhancing human attention models.

Visuo-motor integration is a cognitive process essential for executing smooth actions based on visual information. While previous affordance studies suggest that object perception automatically activates motor representations, the temporal dynamics distinguishing “pattern identification” from “motor judgment” remain unclear. Furthermore, conventional experiments using 2D monitors lack the depth cues necessary for realistic grasping tasks. To address these issues, this study developed a custom-built aerial display using polarizing plates to present floating visual stimuli, enabling participants to perform natural grasping motions without physical contact. We designed a novel dual-priming paradigm that independently manipulates the congruency of visual patterns (identification) and handle orientations (motor affordance). High-density EEG (64 channels) and behavioral response times were recorded, and data were analyzed for the right frontal (F4), midline (Fz, Pz), and sensorimotor (C3, C4) areas. Although behavioral analysis showed no statistical difference between conditions, EEG analysis (N=7) using Two-way repeated measures ANOVA revealed not significant but moderate interaction effects of condition on the N2b component in frontal and sensorimotor areas. These results suggest that the congruency checking process is more sensitively reflected in neural activity during motor preparation for priming tasks.

Recently, Invisible-body avatars have received growing attention in virtual reality (VR) research. Previous studies have shown that, under visual–motor synchrony with virtual hands and feet, illusory body ownership can extend from the limbs to an invisible whole body. While these studies have demonstrated robust subjective ownership effects, physiological evidence supporting invisible-body ownership remains limited. In this study, we replicated the invisible-body ownership paradigm and assessed physiological responses using galvanic skin response (GSR). Virtual threat stimuli were applied, and characteristic GSR fluctuations were observed under ownership-inducing conditions, consistent with previous findings.

This paper investigates the influence of environmental immersion on the visuo-haptic illusion (VHI) within low-fidelity Virtual Reality (VR) settings. While visual dominance often allows the brain to resolve sensory conflicts in favor of visual input, recent research suggests that material properties, such as texture, may be more resistant to this effect. We conducted a preliminary study using a "zero-tracking" setup with a Meta Quest 2, where a static virtual hand replaced real-time tracking to isolate the effects of environmental context. Ten participants were exposed to varying levels of scene immersion-ranging from a blank void to a detailed, auditory-rich sports stadium-while handling physical proxies with different textural properties. Our results indicate that high environmental immersion significantly strengthens visual dominance when haptic discrepancies are subtle, with acceptance rates approaching 100%. However, the illusion collapses when the material mismatch exceeds a certain "tolerance window," leading to haptic dominance regardless of the immersive context. These findings suggest that for low-cost VR simulations, prioritizing semantic and environmental fidelity can compensate for limited tracking hardware, provided that physical props maintain basic textural congruence.

Although creativity is widely considered to depend on attentional function as a core component of executive control, the role of attention to internal bodily states has not yet been examined in detail. Because interoception links bodily responses with affective processes, and anxiety is known to influence creativity via disruptions in attentional function, the relationship between interoception and creativity may vary depending on task anxiety. This study investigated whether the Attention Regulation factor of interoceptive sensibility is associated with creativity and whether this association is moderated by task anxiety. In addition, to detect interoception-specific effects, we controlled for general attentional control in the regression models. A total of 208 Japanese adults were recruited, and data from 198 participants were analyzed after exclusions. Creativity was assessed using a rice variety naming task, which yields indices of creative fluency and the proportion of divergent responses. Participants also completed the Japanese version of the Multidimensional Assessment of Interoceptive Awareness (MAIA-J) questionnaire as predictors, and the Effortful Control (EC) Scale as a control variable. Hierarchical models and regression analyses revealed a significant interaction between MAIA Attention Regulation and task anxiety. Specifically, among individuals with lower task anxiety, higher MAIA Attention Regulation was associated with higher fluency and a lower proportion of divergent responses (i.e., a higher proportion of convergent responses). These effects remained after controlling for general attentional control, suggesting interoception-specific effects. These findings suggest that bodily attention-related processes are an important component in understanding how anxiety modulates creative performance.

Convergent thinking involves the evaluation and selection of ideas and is essential to collaborative engineering design. Previous studies have identified neural correlation within individual convergent thinking, such as the dorsolateral prefrontal cortex (DLPFC) and collaborative processes involving the right temporoparietal junction (R-TPJ); however, the neural mechanisms underlying collaborative convergent thinking remain unclear. We examined whether cognitive task-based interventions modulate brain activity and subsequent idea-evaluation performance. Twelve right-handed male university students formed six dyads and performed an idea-evaluation task after rest, after a calculation task, and after a false-belief task. Electroencephalogram (EEG) signals were recorded using a 64-channel system, and beta-band (13-30 Hz) activity was analyzed. Performance was evaluated based on ratings provided by professional engineers. The results showed that idea-evaluation performance improved after the false-belief task but not after the calculation task. EEG analyses revealed increased DLPFC activation during the calculation task and increased DLPFC and R-TPJ activation during the false-belief task. These findings suggest that cognitive tasks can influence neural activity and subsequent convergent evaluation performance in collaborative engineering design.

Creative thinking in engineering design involves both divergent and convergent processes and is often performed collaboratively through face-to-face verbal communication. Although creative communication has been examined in previous neuroimaging studies, the neural mechanisms underlying divergent and convergent thinking during real-time collaborative engineering discussions remain unclear. In the present study, neural activity associated with creative thinking was investigated during a practical engineering design task involving alternating verbal discussions between two participants. Electroencephalography (EEG) data were collected from 24 adult male engineers during the idea generation and evaluation phases. Verbal utterances were classified as either divergent or convergent, and the preceding 25-s EEG segment was analyzed as the corresponding thinking period. EEG power during divergent thinking, convergent thinking, and resting states was compared across the theta, alpha, and beta frequency bands. Compared with the resting state, both divergent and convergent thinking segments exhibited increased EEG power across frequency bands, indicating a shared neural basis for creative thinking during verbal interaction. Direct comparisons further revealed localized differences, including higher theta activity in the right frontopolar region, higher alpha activity in the right middle temporal region, and higher beta activity in the right posterior temporal region during divergent thinking. These findings suggest that divergent and convergent thinking in collaborative engineering design emerge from a common neural foundation with distinct functional weightings, thereby advancing the understanding of creative thinking in realistic engineering contexts.

In this study, we developed an online version of the Voting Option Trade-off Evaluation (VOTE) task and applied it to university and graduate students in Japan. Participants first selected two policy domains they considered relatively important and two relatively unimportant, and then evaluated budget allocation proposals between paired policy domains under directionally biased constrained choice set. To capture the contextual influence, we computed a shift index reflecting deviations from neutral allocations toward left- or right-leaning options. A two-way repeated-measures ANOVA revealed a significant main effect of option-set orientation structure, whereas neither policy importance nor the interaction effect reached significance. These results indicate that voters’ allocation judgments are systematically shaped by the structure of the available options, independent of subjective policy importance.

The Vocal Temporal Dominance of Emotions (Vocal-TDE) method is proposed to capture dynamic emotional transitions through assessors’ vocal self-reports over time. This method expands the applicability of the conventional Temporal Dominance of Emotions (TDE) approach. Because conventional TDE method relies on real-time button-press tasks that require sustained visual at- tention and manual input, its use is limited in interactive situations or contexts involving physical engagement. The proposed method replaces button-based inputs with voice-based reporting, enabling hands-free measurement in experiences that require continuous phys- ical activity or where graphical user interfaces cannot be presented. We validated the Vocal-TDE method by comparing data obtained using the conventional TDE and Vocal-TDE methods with identical stimuli, namely a 90-second short film. The assessment frequency and the duration of each reported emotion showed no significant differences between the two methods. Furthermore, the temporal patterns of dynamic transitions across multiple emotional states exhibited similar tendencies. These results indicate that Vocal-TDE is a valid alternative to the conventional TDE method and broadens the range of experiences for which emotional dynamics can be measured.