Affective and Pleasurable Design

AHFE International

Accelerating Open Access Science in Human Factors Engineering and Human-Centered Computing

Affective and Pleasurable Design

book-cover

Editors: Shuichi Fukuda

Topics: Affective and Pleasurable Design

Publication Date: 2025

ISBN: 978-1-964867-40-3

DOI: 10.54941/ahfe1005975

Articles

Effects on player perception of jumping extensions with varying trajectories in VR

In conventional games, such as roleplaying and action games, players control characters using a remote control or control panel. This often greatly expands the physical capabilities of the characters, leading to improved game entertainment and a sense of exhilaration. However, in recent years, head-mounted displays (HMDs) have become increasingly popular in the entertainment field. Instead of remote controls, player movements are often synchronized with those of the characters in the game. This provides a superior sense of autonomy and immersion in the gameplay experience, compared with the experiences provided by existing games. Meanwhile, as the player’s physical abilities are strongly reflected in the characters, the range in which a character’s physical abilities can be expanded is limited. Such a drastic expansion of physical abilities as in existing games would cause a sense of discomfort. Therefore, we propose a method that can extend physical capabilities without causing discomfort. The proposed method will lead to the further expansion of the VR game experience. Methods such as redirected walking and jumping have been proposed to augment movement in a virtual space. These methods only extend physical capabilities such that they are unrecognizable, which is smaller than the range of extensions in conventional games. Therefore, this study focuses on the “jumping” action to establish an extension method that can significantly extend physical abilities without discomfort. The VIVE Tracker was attached to the waist and instep of both feet of each experimenter, and the corresponding position coordinates were obtained. The height of each leap was multiplied by the value to be extended. The effects of significant dilatation were investigated in a simple manner. The results suggest that the sense of motor subjectivity might decrease owing to a misalignment between the sensation of jumping and visual information. Next, we investigated the factors that caused the decrease in the sense of motor initiative and the timing of the perception of the sensory gap. We found that the factor that decreases the sense of motor subjectivity is the sense of speed and that subjects perceive the gap when they are ascending. Therefore, we evaluated the effects of changing the leap trajectory on the players.Five jumping trajectories were selected: a parabolic trajectory, warp trajectory, trajectory with the apex shifted forward, trajectory with the apex shifted backward, and gradual trajectory near the apex. The questionnaire asked about the sensation, impression, mood, subjectivity of the leap, and timing of discomfort. These results suggest that cognitive abilities differ between acceleration and deceleration. In addition, unrealistic jumps decrease the sense of subjectivity. In the future, we will conduct experiments to evaluate the differences in cognition owing to acceleration and deceleration and aim to propose an optimal extension method.

Takuto Adachi, Keiichi Watanuki, Kazunori Kaede

Conference Proceedings

Evaluation of Driver Overconfidence in Automotive Driving Using Physiological Data

This research aims to explore a method for the real-time evaluation of overconfidence during car driving. Overconfidence can lead to dangerous driving behaviors, making real-time detection crucial to reduce traffic accidents. In this study, a driving simulation environment was created using virtual reality (VR), and a right-turning scenario was used to encourage overconfidence during the experiment. The driving behavior data (accelerator, brake, and steering), physiological data (skin conductance and electrocardiogram (ECG)), and driving footage were recorded simultaneously. Overconfidence was measured by having participants watch the driving footage they created, obtaining both self-assessment from their own perspective and from an external perspective, as well as the difference between the two. The relationship between the measured overconfidence and the feature-extracted driving behavior and physiological data was analyzed. The results show that, when considering the period from the pedestrian crossing before the right turn until the increase in speed as the "pre-turn phase," and from the increase in speed until passing the pedestrian crossing on the turning side as the "turning phase," all subjects completed the turn and exhibited similar driving behavior during the turning phase. A feature-based analysis of the time-series data showed strong correlations between overconfidence and several features. In the driving behavior data, a significant negative correlation was observed between the minimum accelerator value during the turning phase (r = –0.718, p = 0.013). Furthermore, significant negative correlations were found between the average change in accelerator data in the turning phase (r = –0.676, p = 0.022), minimum slope of the accelerator in the turning phase (r = –0.644, p = 0.032), and minimum steering angle during the pre-turn phase (r = –0.622, p = 0.041). For physiological data, a significant negative correlation was found with the standard deviation of skin conductance (r = –0.662, p = 0.027). These results suggest the possibility of using driving behavior and physiological data to evaluate overconfidence in real time.

Kuno Takumi, Kazunori Kaede, Keiichi Watanuki

Conference Proceedings

Estimation of Intellectual Productivity Using Electrocardiograms during Computational Tasks with Cognitive Load

Long working hours have recently become a major problem in Japan. A means of eliminating long working hours is to improve factories. Thus far, understanding current work efficiency and establishing an estimation method to support it is essential. Understanding the work efficiency of individual workers allows us to propose efficiency improvement measures suited to each worker that contribute to eliminating long working hours and improving the working environment. Working long hours increases cognitive load and decreases work efficiency. With the development of information technology, tasks requiring cognitive load to capture and process information have increased, further affecting workers. Cognitive load refers to the amount of information processed by the working memory of the brain, and processing new information and completing tasks is believed to become more difficult when the cognitive load is increased. Therefore, in this study, we focused on tasks requiring cognitive load and examined a method for estimating work efficiency. In previous studies on cognitive load and work efficiency, biometric information was often used to evaluate cognitive load. Yamaguchi reported an increase in the LF and LF/HF components, indicators of sympathetic activity, and a decrease in the HF component, an indicator of parasympathetic activity, using a time-series frequency analysis of heart rate variability during continuous additive work. Mishima measured the cerebral blood flow in the prefrontal cortex during a verbal fluency task using fNIRS and reported increases in the relative values of oxygenated and total hemoglobin levels. These results suggest that biological information, such as heart rate variability and cerebral blood flow, is effective in assessing cognitive load. However, to date, there has been no research on the quantitative estimation of work efficiency under cognitive load using biometric information. In my research, I attempted to estimate work efficiency using cerebral blood flow in the prefrontal cortex. However, owing to the high cost of measurement devices and the burden of wearing them, issues remain in terms of practicality in the daily working environment. Therefore, in this study, we focused on electrocardiography as a biometric information that can be obtained more routinely and measured during a continuous addition task to establish a method to evaluate work efficiency. A machine learning method was used to estimate work efficiency, and a convolutional neural network (CNN) was used as the learning model. The continuous addition task in the experiment was performed on a VDT screen for 20 min, and the number of correct answers in 30 s was used as task efficiency. Additionally, electrocardiograms were obtained during the experiment. Learning models were created for each participant. The biometric data were time-series data, and CNN was used for feature extraction. PFI was used to evaluate the importance of each heart rate variability index.

Kosuke Sato, Keiichi Watanuki, Kazunori Kaede

Conference Proceedings

Development of a Fast and High-Precision Audio Noise Reduction System to Enhance the Accuracy of Emotion Estimation in Practical Applications

Speech-based emotion estimation has diverse applications, including mental health monitoring, human–computer interaction, and communication enhancement. The accurate estimation of emotions from speech is crucial in the detection of psychological stress, which is a growing concern in today’s high-stress societies. However, environmental noise significantly degrades estimation accuracy, and studies focusing on noise reduction specifically optimized for emotion estimation remain scarce. This study evaluated the impact of noise reduction on emotion estimation by comparing traditional signal processing (spectral subtraction, Wiener filtering) with deep learning-based methods (U-Net autoencoder, convolutional autoencoder). The effectiveness of each method is examined under continuous vehicle driving and transient clapping noise. The results indicate that traditional techniques effectively suppress continuous noise but struggle with transient noise, whereas AE-based methods, particularly U-Net autoencoder, significantly enhance the estimation accuracy in complex noise environments. This study underscores the importance of emotion-aware noise reduction and suggests that deep learning-based denoising techniques can significantly improve real-world applications. Future research will focus on further optimizing the AE architectures and integrating them into real-time systems.

Kanji Okazaki, Keiichi Watanuki, Yusuke Osawa

Conference Proceedings

Real-Time Adaptive Gripping Mechanism Using Object Classification and Feedback Control

Building on the foundation of our previous research, this study transitions from theoretical object classification to practical applications in adaptive gripping mechanisms. In the previous study, we introduced a TensorFlow-based machine learning model capable of classifying objects as hard or soft using visual data from image files or live camera input. The classification process addresses the initial challenge of identifying object types, and this research extends to the critical next step: enabling a robotic gripper to apply appropriate and dynamically adjusted pressure to objects based on their classification and stiffness characteristics.The core contribution of this study lies in integrating force-sensitive resistors (FSRs) into a robotic gripper design to facilitate real-time pressure sensing and feedback. FSRs are low-cost and reliable sensors capable of measuring the force applied to the contact. These sensors were strategically placed on the contact surfaces of the gripper to capture precise pressure data during gripping. This feedback, combined with the classification results, ensures that the gripper dynamically adjusts its force in real-time to handle a wide range of objects.The system architecture involves a microcontroller interfacing hardware with a TensorFlow-powered classification model running on a GPU-equipped computer. The workflow begins with an image input that is processed using the TensorFlow model to classify an object as hard or soft. This classification result is transmitted to a microcontroller that uses the data to determine the initial gripping force. As the gripper closes around an object, the FSR sensors provide continuous feedback on the applied pressure. If the sensed force deviates from the optimal value for the object type, the system dynamically recalibrates the motor torque to ensure secure and damage-free handling.To evaluate system performance, experiments are being conducted in which, to date, we have used 70 small objects categorized into hard and soft items. The TensorFlow model achieved a classification accuracy of 80.25% for both object classes combined. Pressure adjustment experiments demonstrated the ability of the system to maintain gripping forces within a ±12.75% to ±15.8% range of optimal values, minimizing damage to fragile objects while securely grasping more rigid ones. We believe that the ranges mentioned here will converge to lower values as more experiments are conducted and our system is refined. For example, glass cup required a maximum average gripping force of 13.41 N (corresponding FSR voltage reading of 3.97 V), whereas plush toys (flexible soft objects) required 4.16 N ((corresponding FSR voltage reading of 3.23 V), which is significantly lower than that required for glass cup (hard objects).This study bridges the gap between machine-learning models and real-world robotic applications by demonstrating how object classification can inform dynamic interactions with physical hardware. The integration of real-time feedback from FSR sensors introduces a level of adaptability that is crucial for applications in industrial automation, service robotics, and precision manufacturing.

Diptesh Kumar Mandal, Kazunori Kaede, Keiichi Watanuki

Conference Proceedings

AI-Driven Personalized Multisensory Design of Cultural Heritage: A Case Study of Kunqu Opera

With the evolution of global cultural consumption habits, the influence of traditional cultural heritage in contemporary society has gradually diminished. While Kunqu opera is highly regarded for its profound cultural value, its international influence remains limited due to changes in modern cultural consumption habits, language barriers, and the constraints of traditional dissemination methods. To enhance the appeal and impact of traditional culture in modern society, this study takes Kunqu opera, a Chinese intangible cultural heritage, as a case study and explores AI-driven personalized multisensory design strategies. By integrating artificial intelligence technologies, the study aims to create multisensory interactive experiences that engage users’ visual and auditory senses, enabling them to perceive the charm of Kunqu opera, stimulate emotional resonance, and enhance the effectiveness of personalized cultural transmission.This paper proposes a multi-layered design framework that encompasses user data analysis, personalized customization, and multisensory content generation. The study employs AI to deeply analyze and generate the artistic elements and cultural connotations of Kunqu opera. In an innovative approach, it presents Kunqu stories in the form of an animated audio comic, using popular music to express the opera's melodious and delicate vocal style and elegant poetry, while employing a fresh comic style to depict the intricate designs of Kunqu costumes, such as flowing sleeves, fans, and detailed makeup. This transformation of Kunqu's classic performances into dynamic audio-visual experiences allows users to engage interactively and appreciate the opera’s unique charm. On the visual front, generative adversarial networks (GANs) are used to reconstruct Kunqu costumes, stage settings, and performance actions; on the auditory front, deep learning algorithms are applied to perform style transfer and synthesis of traditional singing and music.This research aims to provide new insights for the protection and dissemination of cultural heritage through the AI-driven recreation of Kunqu opera, ultimately enhancing its international recognition and acceptance.

Tián Céng, Jie Zhou

Conference Proceedings

Exploring Cross-Sensory Perception in Dining Environments: The Role of Tactile Surface Properties on Users’ Visual and Gustatory Experiences

Sensory experiences within dining environments play a crucial role in shaping users behavior and emotional responses. This research explores the impact of tactile surface properties on users’ cross-sensory perception of vision and taste within the specific context of dining environments. A controlled experimental design was employed, categorizing material samples based on two tactile dimensions—surface roughness and hardness—resulting in four distinct groups. 24 participants engaged in blind tactile exploration of these materials while completing a cross-sensory perception questionnaire. Analysis of subjective evaluation data revealed that tactile surface properties significantly influenced both visual and taste perception, as well as emotional responses. For instance, materials with higher roughness were associated with heavier and warmer visual impression and specific flavor taste experiences such as saltiness or bitterness, evoking sensations of richness and depth. In contrast, smoother materials were associated with lighter and cleaner visual impressions, often linked to sensations of coolness and sweet-sour flavor profiles, thus fostering a more refreshing and pleasant emotional experience. Similarly, materials exhibiting greater hardness were perceived as more formal and reliable, intensifying the perception of stronger and richer taste. Electroencephalogram (EEG) data further demonstrated that variations in tactile properties activated distinct brain regions related to emotion and sensory integration, such as the prefrontal cortex and parietal lobe. By integrating both subjective and objective measures, this study reinforces the influence of tactile properties on cross-sensory perception. These findings offer valuable insights for designers in selecting materials for products in dining environments, such as tableware and kitchenware. By aligning tactile feedback with users’ visual and gustatory expectations, designers can create more harmonious, immersive, and emotionally fulfilling dining experiences.

Siyuan Guo, Bingqian Li, Yuanyuan Liu

Conference Proceedings

Consideration of Visibility in the Kuiper Belt Focusing on the Placement of Objects

The purpose of this study was to investigate the ability of the Kuiper Belt to identify object shapes. In the experiment, 10 participants performed a gaze input task simulating simple menu item selection and completed questionnaires. Task completion time and error rate were used as objective measures, and usability was used as a subjective measure. During the evaluation, a Friedman test was conducted for each indicator. As a post-test, a Steel–Dwass test was conducted on the indicators for which significant differences were found. These results suggest that usability may deteriorate rapidly when the number of objects placed is 10 or more. In addition, objects were identified by referring to their vertex positions, suggesting that objects with similar vertex positions tended to be misrecognized.

Ryo Ono, Keiichi Watanuki, Kazunori Kaede

Conference Proceedings

Evaluation of UX using Biometric Emotion and Intensity Estimation Machine Learning Models

In recent years, user experience (UX) has become an important evaluation criterion in the development of products and services, in addition to the traditional emphasis on functionality and price competitiveness. In addition to practical aspects such as usability and convenience, “emotional value,” which affects the user’s emotions and senses, is also considered an element of UX. Emotional value is the positive feeling that a product or service elicits in the user, resulting in increased satisfaction and purchasing behavior. Thus, emotion is an important component of UX, and based on the “emotional information function theory” of psychology, emotion has been shown to act as the basis for decision-making in situations where judgment cues are scarce. Therefore, accurately understanding emotions and reflecting them in the UX design process is critical for increasing the value of products and services.However, measuring user emotions during UX evaluation remains a significant challenge. Conventional emotion evaluation methods, such as questionnaires and interviews, are primarily subjective; however, they have several limitations. For example, they may not accurately reflect actual emotional states because they are based on respondents’ memories and biases. Furthermore, it is difficult to obtain real-time emotion data because work must be interrupted during the evaluation process. Therefore, using biometric information to evaluate emotions has attracted considerable attention. Biometric information (e.g., heart rate variability, skin electrical response, and EEG ) may be related to users’ emotional responses, reducing subjective bias and allowing real-time evaluation.Furthermore, recent advancements in machine learning technology have stimulated research into emotion estimation using biometric information. In these studies, models have been developed to estimate psychological characteristics such as pleasantness–unpleasantness and arousal–non-arousal of emotions using biometric information, with efforts focusing on Russell’s emotional circle model. Thus, machine learning has shown the ability to connect biometric information with a psychological framework and quantitatively assess the emotional components of UX.This study aimed to build a machine learning model using biometric information, develop a mechanism for estimating user emotions based on Russell’s emotion circle model , and examine whether the model can be applied to UX evaluation. Specifically, we developed a model that uses biological information such as heart rate variability, skin electrical response, and cerebral blood flow as explanatory variables to estimate the emotional position (pleasant–unpleasant, arousal–non-arousal). The accuracy and reliability of the model were evaluated , and the effectiveness of objective UX evaluation using biometric information was determined by comparison with conventional subjective evaluation methods. Through this study , we aim to develop a new evaluation method that can identify emotional factors in UX evaluation in real time and objectively, thereby improving the product and service design process.

Yuki Fujii, Keiichi Watanuki, Kazunori Kaede, Yusuke Osawa

Conference Proceedings

Dynamic Balance Ability Estimation Method Using Plantar Pressure Measurement for Developing Shoes to Assess Daily Living Walking Ability

In Japan's super-aged society, the increasing proportion of people requiring nursing care is partly attributed to falls caused by a decline in walking ability. Early detection of walking ability deterioration can potentially reduce the risk of falls. Although individuals tend to perform better in laboratory settings than in daily life, this finding highlights the importance of monitoring walking in daily living, which has not been sufficiently explored. This study aimed to develop shoes to evaluate the walking ability in daily life by focusing on balance, which is a key indicator of walking deterioration. Balance ability is divided into static balance, which maintains stability, and dynamic balance, which adjusts posture under unstable conditions; dynamic balance is crucial for walking. Common dynamic balance evaluation methods, such as the one-leg stand test, timed up-and-go test, and Berg balance scale (BBS), often suffer from ceiling effects, which make it challenging to accurately assess differences in balance ability among individuals with moderate or high abilities. To address this, we employed the index of postural stability (IPS) and the modified IPS, which correlate with the BBS and are unaffected by ceiling effects. These indices require specialized equipment such as stabilometers; thus, simpler and more efficient evaluation methods are desirable. We propose estimating the dynamic balance ability by analyzing walking motion using shoes equipped with force sensors to measure the plantar pressure. We developed and validated a compact, lightweight plantar pressure measurement device suitable for shoe integration. We used the device to measure the plantar pressure while walking and to estimate the dynamic balance ability.Finally, a 20-m walking experiment was conducted using the device. Machine learning methods were employed to estimate the dynamic balance ability by processing the plantar pressure data obtained during walking. The model using 16 plantar pressure data points showed the highest estimation accuracy, suggesting the potential of this system for evaluating walking ability in daily life.

Hideyuki Nagashio, Yusuke Osawa, Keiichi Watanuki

Conference Proceedings

Gamified Emotional Evaluation of Virtual Architectural Spaces:The G-SOR Framework and “Lost In Reverie”

With the rapid development of metaverse technology, virtual architectural spaces are playing an increasingly important role in digital experiences. However, existing emotional testing methods for virtual spaces face challenges such as insufficient immersion, lack of participant motivation, and limitations of single-variable research. This study proposes the G-SOR (Gamified Stimulus-Organism-Response) framework, which integrates environmental psychology's SOR model with game design theory, and develops the “Lost In Reverie”game testing platform based on this framework. The research first defined an emotional parameter library for four categories of spatial elements—geometry, lighting, material, and color—through a preliminary experiment (N=31). The game platform designed two core systems based on the G-SOR framework: a spatial parameter system (integrating parameterized definitions and construction methods for single elements) and a spatial immersion system (including task-driven exploration, visual illusion puzzle mechanisms, and emotional data collection). Comparative experiments (N=63) showed that, compared to traditional methods, the gamified approach significantly improved spatial immersion (28.5%, p<0.001) and testing motivation (114.3%, p<0.001). This study provides a new paradigm for virtual architectural space emotional research that combines entertainment with scientific rigor, while offering systematic methodology and parametric guidance for emotionally oriented design practice.

Han Jiang, Yan Yixing, Kai Zhang, Xiaomei Nie, Yuhan Dong, Xing Sun

Conference Proceedings

Construction of a PointNet-based Autoencoder Using a 3D Scene Dataset for Feature Extraction from Indoor Space Point Clouds Excluding Interior Details

In this study, to automatically construct a virtual space with a high degree of freedom of expression that reflects the spatial shape of the real space and the arrangement of objects, we focused on the global shape of the indoor space without interior details as the first step and constructed a PointNet-based autoencoder to extract the features of the shape. To train the machine learning model, we used ScanNet++, which is a 3D indoor space dataset converted into point cloud data. Feature extraction was performed using two types of point cloud data: (1) point cloud data not used in the training of ScanNet++, and (2) point cloud data of an indoor space obtained through 3D scanning of a real environment. Feature extraction was evaluated by comparing the shapes of the input point cloud, restored output point cloud and distance error. As a result, both the ScanNet++ data and the indoor space data were output as rectangular shapes, and the general shapes of the walls and floors of the indoor space were generally consistent, indicating that spatial features were extracted. However, the interior furniture and other objects were removed. To investigate the applicability of the model, feature extraction was performed using 3D objects with elliptical shapes in an interior space. In future work, we will investigate the development of an autoencoder that performs feature extraction by focusing on the local shape around each point using a point-cloud convolution method, along with feature extraction following region classification within the interior space.

Takahiro Miki, Yusuke Osawa, Keiichi Watanuki

Conference Proceedings

Investigating the Influence of Takeover Request Warning Methods on Driver Tension in Level 3 Automated Driving

Recently, the widespread use of e-commerce and online services has significantly increased the demand for logistics. Simultaneously, a shortage of drivers, largely owing to a declining workforce, has emerged as a critical issue. In response, there has been growing interest in autonomous driving technologies as a means of alleviating the burden on drivers. Level 3 autonomous driving, also referred to as conditional automation, allows vehicles to operate autonomously under certain conditions. However, the driver must control the vehicle outside these conditions. When transitioning from autonomous to manual driving, a takeover request (TOR) warning is issued by the system to prompt the driver to regain control. Previous studies have indicated that adjusting the level of tension caused by a TOR warning can improve driving performance after the warning is issued. However, limited research has focused on the specific warnings used in level 3 autonomous driving, especially those issued when the driver is fully disengaged from the driving task. Additionally, repeated exposure to warnings may result in habituation and thereby reduce the warning effectiveness over time.

Taiga Fukuzawa, Kazunori Kaede, Keiichi Watanuki

Conference Proceedings

Construction of a VR music live performance system to support improved sense of presence

Entertainment such as live music performances (live performances) and events are often held in urban centers for reasons such as transportation accessibility and venue size, resulting in regional disparities in entertainment. On the other hand, recent years have seen the emergence of virtual reality (VR)-based live performances, which could solve regional disparities in entertainment. VR live performances can provide many opportunities for people who are unable to attend live performances due to location and time constraints.However, the quality of the experience is expected to deteriorate due to the sense of realism and other sensations that are difficult to reproduce in Reality live performances. However, few studies have compared real-life live performances with VR live performances and examined the effects of factors such as the sense of presence on the audience. In this study, an experiment was conducted with the aim of constructing a system to support the improvement of the sense of presence in VR live music performances. In the experiment, we experienced an on-demand VR live performance and investigated the degree of realism and the differences from a real live performance. From the results, we analyzed the factors that are important for improving the realism of VR live performances. The system is then constructed and evaluated.Ten male participants (23.6±1.8 years old) who had participated in a live concert in the past and remembered the sensation were asked to experience a VR music concert in a standing position for 5 minutes, and a post-experience questionnaire was administered to compare the realism of the actual concert and the VR concert.The post-experience questionnaire consisted of a subjective questionnaire and open-ended questions regarding the sense of presence and the five senses. The subjective questionnaire used a 5-point Likert scale. The items related to the sense of presence consisted of 13 items, which were added by the researcher based on previous studies that have analyzed the sense of presence. 3 was used as the standard for the 5-point scale, based on the comparison with previous real-live experiences. The higher the value, the higher the evaluation of each element in the VR experience.According to the results of the post-experience questionnaire, the vibration and communication elements were rated low. The VR live experience and its components in this experiment did not reproduce the vibrations from the large speakers and communication with other people, such as the performers and the surrounding audience, that are felt in a real live performance. Therefore, the experience tended to feel like a one-way live performance, and the quality of the experience was considered to be compromised. In addition, factor analysis of the questionnaire results showed that among the body perception factors with the highest contribution rate, the values of dynamism, vibration, and sense of self-presence were the most significant, and thus had a large impact on the sense of presence in the live performance.

Shogo Yanaze, Keiichi Watanuki, Yusuke Osawa

Conference Proceedings

Analog Intelligence Approach to Human Augmentation

Today's world is the Industrial Society, which the Industrial Revolution introduced. Thus, it is product-centric and we have been paying efforts to increase production efficiency and product performance. But as is well know, every world changes from one to another and the Industrial Society is coming to its end. Thus many issues emerged. One of the greatest problems is the excessive use of energy.We cannot sustain the current excessive consumption. Although AI is getting wide attention these days as its potential solution, but AI consumes a large amount of energy. So Ai is not a solution at all.But if we look at things from a different side, we can have a self- sustaining society and enjoy life. If we pay attention to process, instead of product, we can create a new process and we can enjoy creation.The product-centric Society is scientific and technology-based. It is quantitative and is based on interval scale. But if we turn our eyes to process, its creation is nothing other than engineering. The basic of engineering is strategy. We perceive the environments and situations and are motivated to determine what action to take to establish our world in the realworld. If one action satisfies our emotion, then that's fine. If not, we repeat the process until we satisfy our emotions.We must remember"emotion"and"motivation"come from the same Latin word "movere". What characterizes humans is its diversity. Our body builds and how we move our bodies vary widely from person to person.But if we can satisfy such individual motivations, then our life will be rich and full of dreams for the tuture. Although we are entering an aging society, if we can provide elderly people with the dreams or expectations for the future, their life will be rich and they will do their best to make their dreams come true.Wha tcharacterizes us,humans, is growth. We grow with time. If such society can be developed, elderly people can also enjoy growing. They can live for tomorrow.In this paper, MahalhanobisDistance-Pattern approach is developed and we discuss how it canvfulfill our growth needs and make our life rich and happy for everybody including elderly people. Mahalanobis, researcher of design of experiments, developed Mahalanobis Distance to remove outliers. Most study of human factors are carried out in Euclidean Space approach. But they are scientific and quantitative. In a word, this is tactics. But we should remember that as William Wordworth pointed out in his "My heart leaps up", our heart leaps up when we are truly impressed and motivated. We need to make a decision how to fully satisfy our emotion. This is an issue of heart so that analog intelligence approach is called for. In a word, we need strategic approach and we need toprioritize our decision. Mahalanobis Distance answers to this need.We used to study detecting emotion from face, but most approaches took too much time and did not produce good results. During these challenges, we realized that we can detect emotion of a character in a cartoon at once. So, we introduced cartoon face model and compared it with a human face model. This way, we succeeded in getting emotion from face at once.Mahalanobis Distance-Pattern Approach developed here put these two successes together to increase human augmentation.

Shuichi Fukuda

Conference Proceedings

Motivational Engineering for Medication Adherence

While Japan's universal health insurance system reduces the burden of medical expenses, it may lack direct incentives to improve patient adherence to medication. To solve the problem, penalties and financial incentives have been considered as possible measures. However, these methods are not suited to the cultural background of Japan and may have limited effectiveness. In this study, we empirically verified a new intervention model suited to the Japanese context. Drawing an analogy between reporting on the progress of a graduation thesis and medication adherence, we focused on “social needs” within Maslow’s Hierarchy of Needs as a motivational mechanism. We utilized notifications and information sharing through Slack to engage students. Furthermore, we propose that a “motivation engineering” perspective is essential for problem solving, in which individual motivations are analyzed scientifically and the most appropriate intervention measures are designed. The experimental results suggest that understanding personal values and fostering habit formation are important factors in influencing medication behavior. Additionally, the findings highlight the importance of designing individual intervention measures and motivation strategies that account for cultural characteristics.

Tomoki Koike, Shunsuke Hirayama, Hikaru Mito, Hana Sato, Yoshiyuki Furumi, Hironori Takeuchi, Daigo Misaki

Conference Proceedings

Design of a Pupil Response Robot that Listens with Empathy

Many proverbs using nonverbal behaviors in human communication are related to human eyes, such as ``the eyes speak as much as the mouth'' and ``the eyes are the mirror of the soul.'' This is because human eyes are an interface that easily reflects own emotions and are recognized as a social clue to convey emotions and feelings in human communication. It is considered that people infer emotions of others in communication by integrating not only global cues such as "gaze" but also local cues such as "pupils." Therefore, it will be possible to clarify human emotional behaviors and design human-like emotional behaviors by recreating the artificial eyes that integrates gaze and pupils. In particular, one of the human-like behaviors is listening attitude by the listener. Listening attitude is an essential behavior for empathizing with the speaker's emotions or cheering his/her up and an important factor in building a trusting relationship with the speaker. This listening attitude has not been realized even in current AI and is an important challenge in solving human-likeness. In this study, a pupil response robot that listens with empathy was designed. This robot imitates human-like eyeball by projecting CG images of the iris and pupil onto hemispherical displays. In addition, it generates nodding movements, eye contact, and pupil dilation as an active listening attitude. Through experience events, it was confirmed that the developed robot is effective for the listening with empathy.

Yoshihiro Sejima

Conference Proceedings

Brand Concepts and Trends

What kind of values should you have when planning or starting a business? In a socialist economy or a state-sponsored company, this may mean running a business according to the orders or policies of the country, but in a liberal economy, you are free to do so.You are free to follow the value that "brands should be beautiful" or not, and you are free to have different values. You are free to choose to "behave beautifully" or not. You are free to aim to "become a beautiful brand" or not, and you are free to aim to become a different kind of brand. In this article, I suggest that you choose to "behave beautifully" and aim to "become a beautiful brand" in line with the value that "brands should be beautiful".However, in order for a business or company to continue, it needs to be supported and purchased by customers. When talking about "customers," you need to consider quantity and quality. In other words, you are free to aim to acquire as many customers as possible who are supportive, even if they are not that enthusiastic, or to acquire a small number of customers who are enthusiastic. In this book, we suggest that you aim to acquire a small number of passionate customers.You are also free to aim for "high profit, low sales," that is, to increase overall profits by selling large quantities at low prices even if the profit per item is small. On the other hand, you can ensure overall profits by selling "high profit, low sales" (a term coined by the author), that is, even if you sell small quantities, if the profit per item is high you can ensure overall profits. You are free to do so. In this book, we suggest that you aim for "high profit, low sales."Note that "high profit, low sales" and "high profit, low sales" are both terms coined by the author. It is strange that the term "high profit, high sales" is commonly used, yet there is no term to express the opposite concept. If there is no term because there is no situation in which it can be used, then this in itself may be a serious problem.Some people may frown upon the phrase "high profit," as being a rip-off. However, high prices and high profits are the results, but still, buyers buy the products because they consider them to be "worth the price" or "worth more than the price," so they are a reward or reward for creating value that is recognized, and a seed of hope or sowing for the future. In other words, it includes a reward for the past (a reward for the artisans and brands that created value) and an investment in the future (hopes for the artisans and brands that will continue to create value), and it can also be seen as high prices and high profits to create the next value. In other words, the products that consumers choose will change society, and the brands that consumers choose will build a society that is rich in quality, not quantity.

Shinya Nagasawa

Conference Proceedings

Personal Space for Avatar Communication in Real and Virtual Environments

In recent years, the development of the metaverse and avatar robots have led to construct an avatar society, where people communicate with each other using their avatars. Just as psychology has been developed as academic studies for building a better society in the real world, the construction of avatar psychology is necessary to build a better avatar society. This paper addresses the issue of personal space as one of the topics of avatar psychology.Personal space is the space in which people feel uncomfortable when other people exist, and it is important for us to have a suitable personal space for comfortable communication. Personal space is known to change depending on the relationships between people such as family, friends, business associates, and others, personal attributes such as gender and age, and the social and cultural background in which the person grew up. Considering such knowledge about personal space is important to improve work efficiency and realize comfortable communication of users when designing spaces where people gather, such as offices, conference rooms, restaurants, event halls, etc. In contrast, there are few research cases on personal space for avatars in the metaverse, and the most of the current spaces in the metaverse are designed using the knowledge about personal space for humans in the real world. In this study, we aim to build a more comfortable and productive metaverse space or a coexistence workspace with communication robots by considering the knowledge of personal space for various avatars. Specifically, the personal space for a CG avatar in a virtual space and the personal space for an avatar robot in a real space were measured through the experiments and it was compared with the personal space for a human in the real space. In particular, when measuring the personal space in a virtual space, accurate measurement can be performed by observing a real-size avatar model from a first-person perspective using an HMD. In addition, the impressions felt by users when facing a human, an avatar robot, and a CG avatar were evaluated using the SD (Semantic Differential) method.The experimental results showed that the personal space for a CG avatar was slightly larger than the personal space for a physical opponent such as a real human and an avatar robot. This difference is thought to be due to the fact that the impressions of familiarity, security, and emotional richness were smaller for a CG avatar than a real human or an avatar robot. It was also shown that the personal space for a CG avatar changed depending on the character expression and the body size of the avatar. Therefore, it is expected that these results can be used as a guideline　for improving comfortable communication and work efficiency when designing communication spaces in the metaverse or designing workspaces where avatar robots exist in the real space.

Tetsuro Ogi, Xiangyu Sheng, Yuki Kida

Conference Proceedings

User-Centered Sustainable Product Lifecycle Management Platform Design

This study proposes Eco Track, a dual-end digital platform designed for sustainable product lifecycle management. The system integrates environmental data provided by enterprises with behavioral data recorded by users, enabling a visualized and interactive management experience across four key product lifecycle stages: production, transportation, usage, and recycling. Employing User-Centered Design (UCD), emotional design, and gamification strategies, the research evaluates the platform’s interaction effectiveness and user acceptance through prototype development and usability testing. Results indicate that data visualization, emotional feedback, and incentive mechanisms significantly enhance users’ awareness of and engagement in sustainable behaviors. This paper establishes a replicable interaction design framework, offering practical design paradigms and methodological references for the development of future green digital platforms.

Yuan Quanjingzi, Mingjiu Yu, Deng Kai Chen

Conference Proceedings

The influence of Non-park urban greenery on the pedestrian mobility of seniors: A methodological framework for the analysis of affective states

The project aims to investigate specific physiological and cognitive aspects of Non-Park Urban Greenery (NPUG) and its effects on the pedestrian mobility of seniors. The interdisciplinary empirical work combines physiological data collection, self-reported emotional evaluations and, age-simulation technology. From a theoretical standpoint, the measuring of affective states was developed by answering a neuro-physiological declarative assessment based on voluntary re-orientation of attention . As part of the proposed framework, a self-evaluating matrix was developed in a mobile application to guide individuals to identify and register insights on their interior states of emotion. The app is a 2-dimensional digital matrix, adapted from the Russell grid for Affective States. The ‘bingo grid’ was exchanged for a more intuitive color grid with 8 regions arguing that it will guide seniors to easily evaluate affect: considering at the same time both the intensity (arousal) and the pleasantness (valence). The affective responses were recorded at specific points of a predefined pedestrian route in the campus of the university. The original hypothesis considered that a point within the regions could describe more accurately the affective state of the experience, while the app codes the numerical values. Real-time responses are from participants navigating a predefined route. Consequently, the paper provides the results of the experiment while outlining the preliminary guidelines to apply the methodological framework in the analysis of other green spaces. This novel approach that integrates technology with human-centered research aims to enhance our understanding of NPUG and its correlation to wellbeing.

Marcelo Sagot Better, Agnieszka Ptak-wojciechowska, Agata Gawlak, Dariusz Brzeziński

Conference Proceedings

The Impact of Visual Design Elements in Tea Bag Packaging on Consumer Behaviour: A Prototype and Pilot Study

Previous research has emphasised the importance of product packaging on consumers' purchase intention; however, empirical studies exploring the specific mechanisms of how packaging design visual elements influence consumer purchase intention remain scarce. In particular, there is a lack of investigation into different levels (sub-dimensions) of design elements, which fails to provide specific guidance for packaging design strategies. This study aims to fill this gap by developing bagged tea packaging design prototypes as visual stimulus materials to examine the effects of packaging design visual elements on consumer responses. First, this study developed 14 bagged tea packaging design prototypes through orthogonal experimental design, incorporating different combinations of five key visual design elements (colour, graphics, logo, typography and layout). These prototypes, after expert evaluation, were used as visual stimulus materials. Second, this study conducted a pilot study with 168 young consumers aged 20 to 29 using scales adapted from previous well-established measures (assessing consumers' brand experience, purchase intention and satisfaction) after expert validation. Preliminary findings indicate that the packaging design prototypes based on orthogonal testing can serve as visual stimulus materials for respondents, and the scale developed in this study has demonstrated good reliability and validity through confirmatory factor analysis (CFA), making it suitable for subsequent research. Moreover, considering sample size limitations, this study employed Kruskal-Wallis tests for preliminary exploration. The visual elements of packaging design (such as Colour, Graphics and Layout) showed significant effects on consumers' brand experience and purchase intention, providing support for the research hypotheses; although the significance of Logo and Typography is limited, they still provide a methodological basis for subsequent exploration of their direct and indirect effects. The results of this pilot study encourage further methods of enhancing consumer purchase intention using orthogonal test-based packaging design prototypes as stimulus materials.

Chang Liu, Mat Redhuan Samsudin, Yuwen Zou

Conference Proceedings

Product Design of Parent-Child Interaction Based on Intergenerational Integration Theory

The widespread use of electronic entertainment products has led to a decline in communication and outdoor activities between grandparents and grandchildren, thereby deepening the generational gap. Existing design studies frequently rely on subjective judgment alone and lack a systematic, quantitative analysis of user needs. To enhance intergenerational integration and improve interactions between grandparents and grandchildren, this study focuses on outdoor natural environments as the primary application scenario. User requirements were collected through interviews, then categorized according to intergenerational integration theory, and analyzed using the Analytic Hierarchy Process (AHP) to assess the relative importance of each requirement. The Quality Function Deployment (QFD) method was subsequently applied to prioritize these needs, thereby reducing subjective bias in the design process.As a result, the study identifies five core user needs: the emotional attribute of positive empathy, the connection attribute of tolerance and stimulation, the functional attribute of natural interaction, the consensus attribute of flexible feedback, and the normative attribute of clear rules. Therefore, based on these needs, the study proposes an outdoor exploration product that integrates the elderly's rich natural experience with advanced interactive technology. This product aims to foster intergenerational knowledge transfer and communication, to stimulate shared interests and topics, and mitigate intergenerational communication barriers.The study demonstrates the feasibility of applying intergenerational integration theory and quantitative analysis methods in product design, offering practical and scientifically grounded insights for developing intergenerational interactive products.

Weijie Jiang, Xin Chen

Conference Proceedings

Context-Based Interactive Experience Design System for Children's Food Education

The issues of childhood dietary disorders and obesity are becoming increasingly severe. Helping children to correctly perceive food is crucial for their healthy development. Current food education tools fail to connect dimensions such as food knowledge, food manipulation, food interest, and food etiquette, making it difficult to establish a correct food cognition from an early age. Learning and development of knowledge are often enhanced and consolidated through situational interaction. Therefore, introducing a contextual learning model into children’s food education to build a comprehensive experience system that encompasses personal context, physical context, and sociocultural context is essential. Stimulating children’s positive emotional experiences with food through situational resources thereby enhances their intrinsic motivation to learn. Methodology: Firstly, 15 food education situational elements were identified through literature review. Secondly, data from the target group was collected using a questionnaire survey and analyzed using factor analysis, principal component analysis, and correlation analysis. Finally, an interactive experience model for food education was constructed based on the revised situational elements, and design strategies were proposed. Significance: This study provides new perspectives and methods for food education experience design. By enriching children’s multisensory interactive experiences, it deepens their understanding of food at cognitive, behavioral, and emotional dimensions, thereby compensating for the lack of emotional dimension in traditional food education.

Xinyu Zhao, Wenxuan Yang

Conference Proceedings

Construction and Empirical Research for User Demand-Oriented Food Product Design Pathway

With the globalization accelerates rapidly and consumer cultures become increasingly diverse, food design has shifted from a focus on functionality to prioritizing cultural value, emotional connection, and personalized experiences. In particular, within the realm of cultural and creative foods, the central challenge lies in integrating local culture with contemporary craftsmanship to meet the varied demands of consumers. Dali, a city rich in historical and cultural heritage, provides abundant inspiration for designing cultural and creative ice creams. However, existing designs largely rely on surface-level cultural symbols, failing to meet consumer expectations for interactivity, emotional engagement, and innovation, and neglecting the exploration of deeper cultural meanings. This shortcoming limits both the market appeal and the cultural dissemination potential of Dali's cultural ice cream.This study employs a mixed-methods approach, integrating both qualitative and quantitative techniques, and proposes a food design framework based on the F-KANO-AD-TOPSIS integrated model. The research first collects and analyzes the preferences of 26 tourists through surveys and in-depth interviews. Using the KJ method, the study identifies four major preference categories for Dali's cultural ice cream design: functionality, enjoyment, ease of use, and emotional connection. From these, 20 key user needs were distilled. The F-Kano model questionnaire was then employed to classify the demand attributes, distributing 387 questionnaires and receiving 371 valid responses. A Cronbach's alpha coefficient of 0.781 indicated the reliability of the data. Based on the prioritization of these demand attributes, 10 significant user needs were identified: a refreshing and pleasant experience, comfortable taste, color harmony, a sense of ritual and participation, brand identity, convenient and safe consumption, diverse flavors, aesthetic appeal, social engagement, and entertainment value.The study then applies AD theory to translate these user needs into functional requirements, mapping them onto 10 design parameters: improved craftsmanship, optimized formulations, color coordination, creation of immersive consumption experiences, establishment of a distinctive brand image, use of healthy ingredients, flavor enhancement, aesthetic design improvements, integration of local cultural elements, and increased enjoyment. These parameters provide a structured approach to designing cultural ice cream. Finally, the TOPSIS method was used for a multi-criteria evaluation of three design proposals. Proposal 1 emphasizes brand identity; Proposal 2 integrates brand identity with diversified aesthetics; and Proposal 3 focuses on experiential consumption methods to increase enjoyment. The results show that Proposal 2 (a diversified design centered on the cultural IP of local attractions) performs best in terms of functionality, emotional appeal, and enjoyment, with a relative closeness score of 0.931, significantly outperforming the other proposals and existing market products.The findings demonstrate that the food design framework presented in this study is both scientifically sound and practical. It not only enhances consumer satisfaction with cultural ice creams but also boosts cultural transmission and market competitiveness. This research extends the theoretical framework of cultural and creative food design and provides valuable guidance for companies in the development of cultural food products, offering a significant reference for future research and applications in this field.

Xiaobu Zhou

Conference Proceedings

Design of Multi sensory Keyboard for Enhancing Communication among Long distance Lovers

Due to distance limitations, some long-distance romantic groups can only communicate with each other online. It is common for young people to use computers to communicate online. However, through investigation, it was found that this communication method gradually becomes monotonous and boring in terms of content and form over time, reducing the desire to communicate. As a computer input device, the keyboard has strong technical functionality and neglects the possibility of emotional care, leaving room for optimization and improvement.This study focuses on the communication needs of long-distance couples and designs a multi sensory keyboard that integrates multiple sensory interaction methods of visual and tactile senses. After preliminary user testing, the keyboard has achieved certain results in enhancing emotional connections, meeting personalized needs, and improving interactive fun.

Dingchen Sun, Ze Bian

Conference Proceedings

Effects of Motion Speed on the Impressions of Monster Characters Appearing in the Video Game Dragon Quest

Dragon Quest is one of the most popular series of video games. In this series, a player travels the fantasy world, battling with various cute and unique monsters in the itinerary. Not only the appearance of the monsters is designed as cute, but also the motions of them are designated to be perceived as cute in recent titles. The present study focuses on the attacking motions of the monsters. The attacking motions included downward swinging of the sword, downward punching motion, filling the power in their body, and so on. The attacking motions of 21 monsters that appeared in Dragon Quest Monsters: The Dark Prince were selected, and their speed were varied in seven steps; the speed multiplied by 0.5 to the original speed (x0.5), x0.63, x0.8, 1.0(original speed), x1.26, x1.58 and x2.0, and in total 147 motion stimuli were prepared. Ten participants watched each stimulus displayed on a screen and were requested to rate the impressions using semantic differential method. The mean values for the 24 seven-step bipolar scales were analyzed by factor analysis. The results of the analysis showed that the impression space for the motions were spanned by quickness, powerfulness and pleasantness with the cumulative contribution rate of 0.78. The results showed that the quickness did not increased but the powerfulness and pleasantness largely declined when the speed was faster than the original for most of the monsters. The quickness decreased largely but powerfulness and pleasantness did not change largely when the speed was slower than the original. The powerfulness was largely determined by the size of the monsters and the pleasantness was determined by the appearance of the monsters, regardless of the motion speed. Moreover, multiple-regression analyses were performed using the degrees of cuteness, preferability and coolness as dependent variables and the factor scores of the stimuli in the quickness, powerfulness and pleasantness as independent valuables. The results of the analyses showed that the pleasant stimuli were perceived as cute and preferable. Both a large monster with a slow motion and a small monster with a fast motion were perceived as cool. It was concluded that the speed of the motions was set at the fastest level where the monsters were perceived as pleasant and powerful in the game.

Kaito Miyoshi, Masashi Yamada

Conference Proceedings

Impressions of the Face of Female Character Produced in Computer Graphics

In recent decades, Japanese video games and anime, called Cool Japan Content, became very popular in the world. In the content, cute female characters, called "moe" or "kawaii" characters, frequently appear as well as beautiful matured females. The cases where the characters are produced using Computer Graphics (CG) increased recently. Especially, the facial parts are important to determine the impressions of the characters. Using CG models, it is easy to change the parameters of the parts of the face and they allow production of various faces including the faces with extreme balances. Many researchers investigated the relations between the parts of the face and the impressions in terms of real faces, but the perception of the faces was rarely investigated in terms of CG models. A study determined the best proportions of width to the length of the face, the eye opening to the width of the eye and so on, for the female faces which was perceived as "moe" and "beautiful" respectively, using a CG model. The results showed that the face of the female characters with the best "moe" and "beautiful" proportions looked young girl and matured woman, respectively. Another study showed that the angle of the eyes and eyebrows is an important factor for determining the impressions of the female characters. In the present study, two CG models; VRoid and Phantasy Star Online 2 New Genesis (PSO2) were used. VRoid allows to designate female characters in anime-like touch, while PSO2 designate realistic characters. The angle of the eyes and eyebrows were set at three levels; slanted, intermediate and droopy eyes. Two types of hair styles (long hair and short hair) were prepared. The other features were also set at three levels; young girl (moe), intermediate and matured (beautiful) according the results of the previous study. In total 36 stimuli were synthesized and presented on a computer display. Ten participants watched each stimulus and were requested to rate their impressions using semantic differential method with 21 semantic scales. The results of the factor analysis showed that the impressions of the faces were spanned by there factors; activity, potency and evaluation. The activity was largely affected by the maturity: The faces with the parameters of the young girls were perceived as active and cheerful and the faces with the parameters of the matured beautiful females were perceived as inactive and gloomy. The potency was largely affected by the angles od the eyes and eye brows: the faces with slanted eyes and eyebrows were perceived as powerful and droopy eyes and eyebrows were perceived as powerless. The evaluation was almost determined by the CG model: the anime-like touch faces produced by VRoid were perceived as pleasant and beautiful, whereas the realistic faces by PSO2 were perceived as unpleasant and ugly. The parameters used in the present study may be valid for the anime-touch faces.

Kyoji Hamazaki, Masashi Yamada

Conference Proceedings

Hybrid Co-creative Design Process Combining In-person and Remote Situations

The COVID-19 pandemic has shifted towards remote co-creative design, but we are now returning to in-person collaboration. Some research suggests that in-person problem-solving situations may not always be as practical as remote situations. This study focused on in-person and remote scenarios using the Double Diamond Model in the co-creative design process. The Double Diamond model consists of four stages: "Discover" and "Define" to identify the right problem, followed by "Develop" and "Deliver" to determine the right solution. This research was conducted based on the hypothesis that combining in-person and remote co-creative design processes, guided by the Double Diamond model, could enhance overall effectiveness. To explore this, I analyzed the execution of in-person and remote co-creative design workshops and examined the differences between these two approaches. The results indicated that participants were most satisfied with the co-creative design process in the following order: in-person, hybrid, and then remote situations. Notably, during the "Discover" and "Define" stages — which involve gathering information and defining the problem — the hybrid situation proved more beneficial than the remote one. Conversely, in the "Deliver" stage, which focuses on determining the right solution, the in-person situation demonstrated more advantages than the hybrid one. Based on these findings, I plan to propose a hybrid co-creative design process and evaluate its effectiveness in an international workshop setting.

Namgyu Kang

Conference Proceedings

Beyond Chatbots: Athlete AI as an Emotional Support Agent for Adolescents

In this study, we explored incorporating athlete personality traits into AI-based emotional support agents. We developed an athlete AI through a personality training process. A user study was conducted with four adolescents to compare athlete AI with traditional AI interactions. Our findings revealed that athlete AI successfully demonstrated distinctive personality traits and increased user willingness to share personal concerns, transforming from an information tool to a personality-driven support agent. While this approach showed promise, balancing personality expression with natural conversation emerged as a key challenge. This late-breaking work offers insights into designing specialized AI personalities for adolescent emotional support.

Ichen Lo, Muting Rau

Conference Proceedings