Recently, virtual reality (VR) has become a readily available technology owing to its compatibility with many devices and applications, such as head-mounted displays (HMDs). It is easy to experience an immersive virtual space by using an HMD. Generally, three-dimensional (3D) videos are considered to provide more realistic depictions compared to two-dimensional (2D) videos. Specifically, even for the same video, the influence on the viewer differs between the two viewing methods. However, depending on the type of emotion invoked in the person by the image, the realism is not necessarily important, so 3D is not always a more useful viewing method. We prepared similar 2D and 3D images and conducted experiments by measuring cerebral blood flow, sweating, heart rate variability, and conducting a questionnaire survey. Then, in the image providing soothing, it was found that 3D expression tended not to be favored.

Fragrance influences the psychophysiological state of a person as the volume of fingertip plethysmogram changes during one’s aspiration of aroma. Recent studies have used aroma in various settings, such as for medical care, relaxation, and productivity-enhancing environments. However, no study has elucidated the kind of behavior shown on the fingertip plethysmograph near the stimulation and discrimination threshold for the sense of smell when the odor intensity of aroma changes. Although our research suggests that emotions can be evaluated using the Lyapunov exponent of chaos analysis, no study has clarified the appropriate method for chaos analysis. Therefore, in this study, we performed a chaotic analysis of fingertip plethysmograms for evaluating emotional changes near the stimulation and discrimination thresholds during aroma inhalation. Results were evaluated using a correlation dimension and Lyapunov exponent. The optimal analysis method was examined thereafter.

At present, visually impaired people have difficulty using mobility support facilities due to their dependence on the surrounding environment, such as sound and snowfall. In addition, people who later became visually impaired may find it difficult to use these facilities. The current study aimed to develop an easy-to-use walking guidance system that uses three-dimensional (3D) sounds, regardless of the surrounding environment, and examined the characteristics of the guidance sounds produced by the 3D sounds. We implemented this system using a head-mounted display capable of generating three-dimensional sound. Participants were asked to walk in a visually restricted environment to either a real guidance sound or a 3D guidance sound with three types of ambient sounds, and their walking trajectory was evaluated. As a result, we found that the 3D guidance sound showed the opposite tendency to deflect from the ambient sound. However, the guidance accuracy of the 3D guidance sound was similar to that of the real guidance sound.

In this study, we aimed to examine the usefulness of gait classification and feature visualization based on multivariate data for the development of a gait feedback training system capable of considering the physical differences among the trainees. The multivariate data considered in this study were the joint angles and the ground reaction forces. In addition, all multivariate gait data were labeled as gait “rarely associated with stumbling” or “frequently associated with stumbling”. A convolutional neural network was used to learn the gait features. Furthermore, the feature parts of the multivariate gait data used for classification were visualized on a heat map created using Grad-CAM. As the results indicate, a heatmap is able to show the feature parts of a gait frequently associated with stumbling, through which the trainee can adjust their gait.

This study aimed to propose a display that incorporates the Fraser-Wilcox illusion with a higher attention effect. Fraser-Wilcox illusion is a type of visual illusion in which a part of a still image appears to be moving. We compared the attention effects of shapes that induce the Fraser-Wilcox illusion and those that do not. As an evaluation method of attention recall effect, the visual discrimination task, which was an experimental technique to measure the characteristics of visual attention, was used. The correlation between the strength of visual illusion and the target reaction time was examined. As a result, a strong negative correlation was observed. The larger the strength of visual illusion, the faster the detection of the target. These results suggest that the Fraser-Wilcox illusion has an alerting effect on visual illusion.

Fragrance influences the psychophysiological state of a person; the volume of fingertip plethysmography is consequently affected by the aspiration of aroma. Further, it may make an individual uncomfortable depending on the amount of odor that they inhale. It has been reported that fingertip plethysmogram has a chaotic nature and is sensitive to pleasant and unpleasant emotions. Chaos analyses have demonstrated various significant findings: however, multiple setting values used in these analyses are arbitrarily determined and are not standard. Therefore, in this study, we compared the biometric index obtained by analyzing the fingertip plethysmogram through a fingertip volume pulse wave investigation. This was done by quantitatively flowing the scent directly into the nose and conducting a real-time subjective evaluation with the collaborators. Then, a suitable development time was examined. The Dynamic time warping (DTW) was used to compare the subjective evaluation and results of the analysis for various fingertip volume pulse waves. A most suitable development time was obtained based on the DTW results. The results suggested that emotions can be measured when an individual is exposed to fragrances.

This study aims to examine the relationship between visual fatigue and brain function in the prefrontal cortex (PFC) during a luminescent sentence reading task. We conducted a 10 min luminescent sentence reading task in a dark room. Twelve healthy young volunteers participated in this study and performed six repeated experiments. The time-course of oxyhemoglobin (oxy-Hb) in the PFC was measured using a 22-channel wearable near-infrared spectroscope. Subjective visual fatigue and sleepiness were assessed using the ' Jikaku-sho shirabe' questionnaire. In addition, the critical flicker-fusion frequency (CFF) was measured as the objective index of fatigue. CFF could be used as the objective support for sleepiness, because Delta CFF and the feeling of drowsiness, Delta text{I}, were negatively correlated ({r}=-0.24, {p}<0.05). In our analysis, oxy-Hb in the medial PFC (mPFC) exhibited a significant negative correlation with the feeling of eyestrain, Delta text{V}({r}=-0.37, {p}<0.01), but not with other fatigue indexes. It was suggested that the development of subjective visual fatigue is related to the deactivation conditions in the mPFC. This study played an important role as the control for the evaluation of objective visual fatigue during various colored LED luminescent sentence reading tasks.

Product-service systems (PSS), which create value by integrating physical products and services, have received much attention as a promising option to increase manufacturers’ revenue and reduce environ-mental impact. The process of designing a PSS re-quires collaboration among various experts who use domain-specific knowledge. Therefore, several re-searches have been investigated for developing design tools tailored to their expertise. However, while the specialization of design tools can be useful for experts, it hinders companies from ensuring the integrity of design information in different design elements. This re-sults in the failure in achieving expected benefits. To address these issues, this study applies the concept of interoperability to PSS design to integrate design information from different domains. In particular, ontological descriptions is adopted to achieve semantic interoperability in different design elements. The application of the proposed ontology to a lecture on PSS design highlights that the proposed method is effective for integrating information on PSS design elements and those between value creation and capture.

In this study, we evaluated the mental stress due to corner entry speed while riding a mobility scooter by measuring skin conductance. A mobility scooter is a four-wheel electric wheelchair with a steering bar-handle. These scooters, known as 'senior cars' in Japan, are generally sold for use by elderly people. Some of these scooters have a system that automatically reduces their speed to approximately 70% of the maximum depending on the steering angle at corners. However, because of mismatch with the intention of the driver, the preset deceleration rate may be uncomfortable for the driver. With the objective of providing appropriate assistance without any discomfort in such circumstances, we evaluated the mental stress of the driver. Skin conductance (SC) is one index for measuring mental sweating related to sympathetic nerve activity. First, we modified a commercially available mobility scooter for use as an experimental scooter such that it can be controlled using an external personal computer. Then, we measured SC in a task that involved turning right into a corner at a constant speed using the experimental scooter, and evaluated the mental stress due to the corner entry speed. A positive correlation between the amplitude of SC and entry speed was obtained. Further, the results of a questionnaire used to evaluate the ride discomfort after each experimental task indicated that 'No Fear' and 'Safety' decreased as the speed increased. As these results are consistent with the measurement results of the SC amplitude, we concluded that fear due to corner entry speed can be evaluated using SC.

Currently, visually impaired people find it difficult to use mobility support facilities because they depend on properties of the ambient environment, such as ambient sound and snow cover. Furthermore, people who became visually impaired in later life also find it difficult to use these facilities. Therefore, in this study, we aimed to develop a walking guidance system, using three-dimensional (3D) sound, that is not influenced by the surrounding environment and is easy to use. We verified the characteristics of the guidance sound produced by 3D sound. This system was implemented with a head-mounted display that is capable of producing 3D sound. In a visual restriction environment, blindfolded participants were asked to indicate toward a real or 3D guiding sound with two conditions: With head fixed or free, and using a tweet-like or water sound. We also collected subjective evaluations of the sounds. As a result, we found that a stationary sound including frequency components in a wide band, like a water sound, was able to be localized with a higher accuracy than a tweet-like sound, and was suitable as a guidance sound, according to subjective evaluations.

Fragrance influences the psychophysiological state of a person; the volume of fingertip plethysmography is consequently affected by the aspiration of aroma. Further, it may be uncomfortable depending on the amount of odor that you inhales. In this study, to observe changes in emotions during aroma aspiration, the flow of scent was continuously directed into the nose of participants quantitatively and directly. Subsequently, the stimulation threshold during aroma inhalation and change in emotions were evaluated through a questionnaire and fingertip plethysmography. We developed a fragrance generator to ensure quantitative aroma flow and participants were asked to subjectively evaluate aroma intensity and the level of comfort in real time.The results confirmed the chaotic nature of the time series data of fingertip plethysmographs. Moreover, it was confirmed that the Largest Lyapunov Exponent (LLE), which provides an indication of the amount of fluctuation, corresponds to the subjective evaluation of comfort/discomfort. The value of the index decreases during pleasure and increases during discomfort. In other words, it was confirmed that the pulse wave fluctuated when the participants were experiencing discomfort.

In the recent years, research on intelligent educational systems has attracted significant interest in exploring data from educational settings to understand learners' behavior and mental states. Hence, the means to deduce the mental state of a learner has been obtained from a variety of physiological indices. The current study aims to understand how a learner answers a question with confidence. By measuring the brain activity and eye movement by using a set of multiple-choice questions, this study analyzes the relationships between the confidence of answering the question, brain activation and eye movement at the time of answering. When there was a high confidence of answering, oxyHb tended to increase near the prefrontal cortex frontal pole and sight line movement to reread question text tended to decrease. The results show that this technique may be able to capture the mental state of the learner by measuring a combination of the change of oxyHb and sight line movement.

There have been several recent attempts to aid car drivers by providing information on internal and external car environments. The optimal amount of information must be determined to avoid confusion. In this study, the “Stroop task” was used for information processing, and the cognitive load was gradually increased by adding information in stages. We designed and conducted two tasks that originate in the “Stroop task”; these two tasks feature significant differences in cognitive load. We also measured brain activity using near-infrared spectroscopy (NIRS) under the assumption that such activity can be used as an index of cognitive load. Both tasks were associated with increased oxy-hemoglobin levels in the prefrontal area, and the task with a higher cognitive load was associated with a more substantial increase in oxy-hemoglobin; this indicates that oxy-hemoglobin levels may be used as an objective index for the evaluation of information-associated cognitive load.

To examine brain areas related to the cognitive load condition during the Stroop task, we proposed a method using a Deep Neural Network (DNN). We acquired cerebral blood flow data in congruent and incongruent tasks by near-infrared spectroscopy (NIRS) equipped with 22 ch. The data were used to train a DNN, and the influence of each factor on the output was evaluated. Our DNN model consists of independent input layers for each channel of NIRS, as well as fully-connected hidden layers and output layers. Our results suggest that the medial prefrontal cortex (focusing on cognition) and the left inferior frontal gyrus (focusing on language processing) were involved in the cognitive load during the Stroop task. These results in the Stroop task were consistent. Therefore, the proposed method’s utility was confirmed.

Emotion is an internal and subjective experience that plays a significant role in human life. There are several methods of recognizing emotions in people, the most authentic of which is using physiological signals, as they are beyond one’s control and strongly correlated with human emotions. This study aims to develop an emotion recognition system based on three physiological signals, namely, brainwave, heartbeat, and facial muscular activity. It utilizes deep neural network (DNN) and the T method of Mahalanobis-Taguchi system (MTS) to process the multiple physiological signals and further recognize the states of human emotion. As such, nine emotions are effectively recognized on a two-dimensional model through the DNN, then compared against several other algorithms, such as MTS, SVM, Naive Bayes, and K-means, where its superior accuracy is validated. Moreover, although the T method only improves the classification accuracy on the valence state, it rather obtains the intensity of emotion in different states. Furthermore, in this study, the proposed DNN is implemented into a wide range of applications for an accurate understanding of the human emotional states, whereas the T method is utilized to respond to the emotional intensity in different states. Finally, a real-time emotion recognition system is developed with DNN as the classifier; this system can directly monitor the variation of the human emotion through reliable and objective emotion analysis results from the physiological signals. Thus, the method can provide useful treatment effect information for robots or assistive apparatus serving activities of daily living.

In this research, to develop a biofeedback training system where trainees can efficiently train inadequacies that do not satisfy ideal walking using a deep learning, we examine a method that discriminates between ideal walking and nonideal walking. In the experiment, to examine the walking components used for the input data, the ground reaction force and joint angle were measured when young people walked normally and when they walked with a brace, to simulate elderly motions. Further, these data were discriminated between conditions as input data using a Convolution Neural Network (CNN). The average accuracy was 79.5% when all walking components were used as input data. In addition, it is thought that it is most suitable to discriminate walking by using all walking components, in consideration of implementation in the system.

In communication, it is very important for a speaker to understand the comprehension state of the speaking partner. In this study, the “comprehension state” is defined as whether or not the speaker’s message is clearly understood, which is difficult to accurately evaluate. This study aims to evaluate the comprehension state from the pupil size using machine learning. We conduct a word association task using elements that are similar to those used in conversations and measure the pupil size; this pupil size data is used as input data for machine learning. The results show that high accuracy is achieved by learning the low frequency components of the pupil size.

The authors' goal is to develop brakes that are more satisfying to users in sensory aspects by explaining the relationship between people's state of mind and a vehicle's physical quantities. However, there are no scales to measure such feeling. In this study, a psychological scale was constructed to measure the affective state of general automobile drivers, for the purpose of clarifying the effects of brake characteristics on the affective state. To construct the affective scale, two experiments were conducted. In the first experiment, a comparison of four types of vehicles was performed and then an interview survey was conducted to collect a broad sample of terms used to evaluate the affective state. From the results, it was hypothesized that the evaluation terms were divided into eight categories. For each hypothesized category, ten evaluation terms were selected from the terms obtained in the interviews, the items of the existing psychological evaluation scale, and the synonym survey results of them. In the second experiment, we evaluated terms we selected to conduct an assessment of 15 types of driving scenarios that were selected from among brake evaluation patterns. Next, we conducted a factor analysis using the assessment results. As a result, the affective states were classified according to seven factors (Well-Being, Anxiety and Surprise, Liveliness, Positive Feelings, Disgust, Boredom, and Concentration), with five items attached to each factor, yielding a total of 35 items. The reliability and the validity of each scale were investigated. Results have shown that the subscales have highly internal consistency and factorial validity.

In recent times, light-emitting diode (LED) lighting has been widespread in offices and manufacturing buildings, but for assembly operation in manufacturing situations, it has not yet been explored. In this study, we measure color preference and visual fatigue to evaluate LED task lighting for assembly operation. We conducted an experiment to obtain subjective responses to color preference and to measure visual fatigue under LED lighting. The participants were asked to assemble a gearbox model as a task under three different lighting conditions. As a result, lighting with high color temperature was rated with high visibility, fatigability, and low preference. On the other hand, critical fusion frequency, which indicates visual fatigue, was higher in lighting with low color temperature than in lighting with other conditions. This indicates that color preference as a subjective response discords from visual fatigue evaluated by biological information.

Light-emitting diodes (LEDs) are widely used for backlighting text indicators in vehicles. It is desirable that such text indicators have high legibility and low fatigue-inducing qualities. Suitable spectral distributions for text presentation in vehicles have not yet been clarified. This study examined legibility and fatigue when participants read Japanese sentences presented through backlight emission for 10, min in a dark room. The backlight indicators employed four type of white LEDs. The participants were asked to answer questionnaires to evaluate legibility and fatigue. Moreover, the critical fusion frequency and brain activity were measured to evaluate fatigue objectively. The questionnaire results show that all colored backlighting displayed almost an equal legibility. The overall result suggests that participants were most and least fatigued when viewing text illuminated by high color-rending and greenish white backlighting, respectively. Thus, fatigue can be reduced using LEDs with different spectral distributions.

In this paper, we proposed a walking training system in which a trainee walks on a treadmill while viewing virtual reality (VR) images using head mounted display (HMD) for the purpose of developing a more continuous and effective walking training. Furthermore, the walking patterns when walking on the treadmill while viewing VR images of walking on a slope and level ground were determined and compared. Consequently, the braking force when viewing a VR image that walking on a downward slope was larger than when walking without viewing a VR image. Therefore, the same tendency as walking on an actual downward slope was obtained when subjects walked while viewing a downward slope image, and it was shown that the walking pattern can be changed by the proposed walking training system.

The development of an automatic driving system is accompanied by the increasing importance of driver monitoring. It is necessary to estimate the state of the driver including actions with less load on the driver. In this study, we used the seat pressure as an indicator to assess the status of a driver. In the experiment, we measured the seat pressure during automatic operation under the designated state (forward gaze, cell phone use and sleeping) of the driver. Characteristic changes in the center of gravity position of the driver were confirmed during cell phone use and sleeping. Subsequently, we evaluated seating surface pressure data by calculating the accuracy of state estimation using machine learning. The results show that the accuracy of estimation corresponded to 76.8% in the overall evaluation. This suggests that it is possible to estimate the state of the driver during automatic driving using seat pressure.

With the ongoing development of information and communications technology, it is apparent that computers will be indispensable partners in future human life. In terms of computer information processing, techniques for coordinating human sensitivity and emotional levels are attracting attention and stand in contrast to conventional knowledge information processing. In this chapter, we present an overview of content-oriented research considering knowledge itself and describe an approach to handling knowledge of human sensitivity and emotions using computers. First, we provide an overview of existing content-oriented studies in the field of artificial intelligence (AI). Second, we discuss problems regarding the handling of knowledge of human sensibilities and emotions. Then, we introduce an ontology pertaining to the concept of color emotion as an approach to understanding human sensibilities and emotions. Finally, we describe the applicability of such an ontology to human-computer interactions and for the understanding of human nature through the concept of analysis-by-synthesis.

Due to the growing elderly population in many developed countries like Japan, assistive devices are increasingly needed to overcome the difficulties of their activities of daily living (ADL). The motion of standing up places a particular burden on the ankle and knee joints and connects static motions (e.g., sitting and lying) and kinetic motions (e.g., walking and climbing up/down the stairs). In our previous research, two types of standing-up assistance apparatuses were introduced according to the suitable standing-up motions. However, these devices are bulky and cannot be widely used in practical settings such as bus terminals. Regarding the suitable standing-up motions of the elderly in the previous literature, a standing-up apparatus with a pantograph mechanism is introduced to diminish the size of the mechanism based on three design principles: (1) the trajectory of the user's COG is an arc; (2) the trajectory of the user's hip joint while standing up is a straight line at an angle of 45 [deg] from the horizon; and (3) the user is supported until the knee angle reaches 60 [deg]. Simulations showed that the load on knee and ankle joint is decreased, and user may stand up easily with our proposed apparatus. A prototype of the apparatus that may be installed on various chairs in public was fabricated. Considering that the apparatus may be used in places without electric power supply, the apparatus is made simply with links and a gas spring instead of electric actuators. Experimental measurements of the trajectories of the center of gravity (COG) and hip joints show that the proposed apparatus may help users to stand in a way that minimizes stress on their lower-body joints.

Wearable assistive devices have been receiving considerable attention in academic circles. To make these devices efficient, we need additional research on the service lives of the mechanical elements used in these devices. The wearers of these devices frequently encounter unexpected movements that lead to motor failure in the devices. The purpose of this study is to develop an overload protection mechanism using a torque limiter, which can eliminate the overload torque delivered in the reverse direction to effectively prevent the device from breaking and ensure the safety of the user. To improve the service life of assistive walking devices, we designed a sandwich mechanism for the final gear of the servo motor. We made the material from rubber and configured it between a pair of circular plates. The surface tractive force delivered the required torque. When the surface load exceeded the maximum friction force, the circular plates slipped and protected the device. In this paper, we implement a torque limiter and prove its durability by performing experiments using two circular plate designs, one with grooves type and another without grooves type. We also use various materials to assess the applicability of the assistive walking device. The findings indicate that the with grooves type gives better torque performance; it achieves the same rated torque as the servo motor. Thus, this study recommends that with grooves type is particularly suitable for the elderly who require high assistive power. On the other hand, without grooves type is suitable for users who employ the device for extended periods because this type has an excellent service life. Our experiment proves that the torque limiter that we developed can withstand the load torque over 300 times for situations involving the loss of balance such as stumbling and slipping. Finally, we experimentally validate the improvement of walking performance by using this torque limiter.

In this research, a computational model that simulates the mental function of multicolor aesthetic evaluation through backpropagation neural networks was constructed. Helmut Leder's psychological model served as the theoretical framework. We determined the macro-architecture of the computational model through two psychological experiments using the semantic differential (SD) method. The aesthetic score of a multicolor stimulus is defined as the inverse of its factor score on the factor “Pleasure” extracted in the first experiment, and each of the three factors extracted in the second experiment—i.e., “Stability,” “Heaviness,” and “Presence”—is regarded as a simple perceptual feature. The genetic algorithm was then employed to optimize the hidden layer node number, the learning rate, and the momentum constant of each neural network. In two simulation tests, this computational model exhibited some predictive power, implying that the model can be regarded as a relatively successful approximation of the psychological mechanism of multicolor aesthetic evaluation. In addition, the results of the second simulation also show that the perceptual feature “Heaviness” possesses the principal impact on the aesthetic evaluation of multicolor objects, whereas the other two perceptual features “Stability” and “Presence” have a minor influence. The heavier and/or more stable a multicolor object is perceived to be, the less aesthetically pleasing it is. Conversely, the stronger the sense of matter presence a multicolor object elicits, the more aesthetically appealing it is. © 2016 Wiley Periodicals, Inc. Col Res Appl, 42, 216–235, 2017.

Recently, the viewpoint of service has become widespread. In learning domains, service providers have been required to adopt the viewpoint and to provide values co-created by teachers and learners. However, production by a service provider means the generation of potential value, whereas usage by a service receiver implies the generation of real value. That is, a receiver creates value independent of the value generated by a provider. Indeed, value co-creation takes place in a joint sphere where providers and customers interact with each other. Thus, service providers must attempt to generate the potential value through service design and play a role of co-producer of the real value in the interactions between the providers and the receivers. In the current study, we developed an ontology to share the knowledge on value co-creation and the information related to the interactions between the providers and the receivers. In concrete terms, we proposed ontological descriptions of receiver states for sharing knowledge in learning service design, and conceptualized elements of consensus building and value co-creation, according to the conceptual model of value co-creation proposed in a previous study. In addition, concepts pertaining to academic emotions are also described to represent contents of consensus building. Then, we demonstrated instances of a consensus building, which is extracted from conversations during counseling by a teacher (mentor) and a learner. As a result, we obtained a metamodel for co-design processes in the learning service design, and a conceptual framework of the knowledge sharing on learning states including academic emotions by the ontological descriptions.

We proposed a method that promotes walking which uses a two-dimensional emotion map. In this method, a subject listened to a beat sound while walking. The beat was adjusted according to the condition of the subject, based on his/her comfortable walking frequency. Simultaneously, the walking velocity, the stride length, and the heart beat of the subject while walking was measured, and LF/HF was calculated from the result of measured heart beat. Furthermore, we asked them to give information on the condition of their feelings, from pleasant or not by 5 levels in each case of the beat sound. By using these data, a two-dimensional walking condition map and two-dimensional emotion map were made and we found the relation between these two maps. Therefore, according to the emotional condition of the targeted person while walking, this method can promote or restrain walking adequately, only adjusting the beat sound.

In the current study, we formulated ontological descriptions for emotional states of users using motion assistance apparatus. In our previous studies, we developed apparatus that assists the ankle joints of patients in neuro-rehabilitation. Although we have confirmed the apparatus can improve the gait of the patient while walking, we have to capture and manage patients’ emotions as feedback of the usage. The ontological descriptions provide a feasible framework for common understanding while sharing and using this knowledge. In future work, we will develop a subset of this framework for more practical usage, in addition to expanding the descriptions of emotional states. After that, we will assess the validity and utility for design and implementation of software and hardware applications using this subset.

In this study, we measured brain activity using near-infrared spectroscopy (NIRS) when a person was feeling discomfort caused by vibrations. We used the variance in oxygenated hemoglobin (oxy-Hb) levels as an evaluation index. Correlation coefficients were derived from the results of brain function measurements and sensibility evaluation of discomfort using a questionnaire. As a result, a high negative correlation was observed between discomfort and both vibration and brain activation around the medial prefrontal cortex, and a high positive correlation was observed between discomfort and both vibration and brain activation around the lateral prefrontal cortex. This suggests the possibility of evaluating discomfort on the basis of brain activation.

Many studies have been conducted during the last two decades examining learner reactions within e-learning environments. In an effort to assist learners in their scholastic activities, these studies have attempted to understand a learner’s mental states by analyzing participants’ facial images, eye movements, and other physiological indices and data. To add to this growing body of research, we have been developing the intelligent mentoring system (IMS), which performs automatic mentoring by using an intelligent tutoring system (ITS) to scaffold learning activities and an ontology to provide a specification of learner’s models. To identify learner’s mental states, the ontology operates on the basis of the theoretical and data-driven knowledge of emotions. In this study, we use statistical models to examine constructs of emotions evaluated in previous psychological studies and then produce a construct of academic boredom. In concrete terms, we develop ontological descriptions of academic boredom that are represented with statistical models. To evaluate the validity and utility of the descriptions, we conduct an experiment to obtain subjective responses regarding learners’ academic emotions in their university course and describe them as instances on the basis of the ontological descriptions.

We developed a device that promotes walking which uses a two-dimensional emotion map for the exercise of the elderly. This map was made with two axes, the horizontal axis was pleasant or unpleasant, with the vertical axis being arousal or not, same as the circumplex model of affect (J. A. Russell (1980)). This device has two means to promote walking. One is an ankle joint assistance system, which consists of a servo motor and a controller which can adjust the targeted walking cycle, the targeted angle of the ankle joint, and the sensitivity of the sensors which was attached under the sole. The other is a beat listening system. The beat listening system makes the beat sound, which is decided for the purpose of the user, and based on the walking cycle. The user listens to the beat sound using head phones. By combining and using these two means simultaneously, users can change their gait and emotion, therefore, they can continue to walk for a long time.

We developed a walking assistance apparatus which can be used for gait training of neuro-rehabilitation for patients and for the promotion of exercise for the elderly. This apparatus assists only the ankle joints of the equipped person according to the posture of him/her while not only walking, but also getting up, sitting, and standing. By using this apparatus, the dorsiflexion and plantarflexion of the ankle joint of the equipped person can increase while walking. This apparatus can improve of the equipped person's gait.

This paper discusses human action modeling and its application to a control system that uses the Mahalanobis-Taguchi System (MTS). In this study, we define embodied knowledge as being included in tacit knowledge. We also define a set of skills based on experiences and intuitive sense as seen in creating an art, sport, craft, or other skilled task. Embodied knowledge is difficult to express explicitly. As our goals, we analyze embodied knowledge acquisition for human action modeling and apply to a control system by using MTS. An analysis of embodied knowledge using devices and pattern-recognition techniques to recognize un-explicit knowledge are being developed owing to recent improvements in technology. Embodied knowledge acquisition element of recognition can be represented as a pattern recognition technique. In this paper, we confirm that MTS is an adaptive method for recognizing pattern in human action modeling. We set up a control model including a controller using the MTS, which is modeled after an internal model of the cerebellum. We apply the controller based on the Recognition Taguchi (RT) method to invert the control of the pendulum. The result indicate that the controller is capable of detecting disturbance.

Here, we investigate and discuss the effect of accuracy of imitation for improvement of skills on brain activity. In order to improve the skills, learners combine and accumulate information of the skills through practice. Thus, we used near-infrared spectroscopy (NIRS) to investigate brain activity during the process of improvement. Evaluation of the level of knowledge acquisition with monitoring of brain activity can be an indicator of the learner's degree of skill progression. Therefore, our final goal is constructing a new learning model based on brain activity monitoring and improving learning efficiency. We experimented on the assembly operation by imitation learning that assumed work in the manufacturing industries from a previous example. As a result, we showed the possibility of brain activity shift with improvement of the skill. In this article, we targeted task accuracy and investigated whether the brain activity shift is caused by a progress in the task accuracy, act of practice, or some other factor. As a result, we showed a possibility that the trend shift in the right and left dorsolateral prefrontal area and frontal pole was not caused by the simple task accuracy improvement but by the action of practice, which helped subjects store the information.

We developed a new method that can diagnose damage on a gear tooth surface using a laser beam without a rotary encoder. The method procedure is as follows: 1) The tooth bottom, the tooth tip and their two median values are detected using the differential values of the laser reflection data. 2) The gear rotation speed is calculated with these four positions, and interpolated according to the rotation fluctuation. 3) Using the calculated gear rotation speed, the measured data can be converted to the corresponding gear rotation angles. Thus we can diagnose the damage of the gear tooth surface precisely and can easily set up the experimental measurements without being influenced by rotational fluctuation. To confirm the validity of the method, we conducted the diagnosis experiment and we created contour maps to show the diagnosis accuracy variation according to the fluctuations of the amplitude and cycle. Based on these maps, we found that the diagnosis accuracy of the damage size is the same irrespective of the presence or absence of a rotary encoder. The diagnosis accuracy of the damage location without using a rotary encoder is lower than the result obtained using a rotary encoder because we assumed that the detection of the damage start point is delayed using this new method. Furthermore, we defined the limit using the conditions of this method from the sampling theorem; the validity of this definition could also be confirmed from the contour map.

We developed a walking assistance apparatus RE-Gait® for not only the elderly but also apoplexy patients, which assist the ankle joint of the equipped person. By only assisting the ankle joint, the equipped person’s leg can be raised. To keep up the user’s motivation for exercise using this apparatus, it is necessary to assist not only physically but also mentally. Therefore we suggested the new method of promotion walking, using the walking assistance apparatus while listening to the beat sound, which was adjusted according to a two-dimensional emotion map. In this paper, the relation between emotion map and walking condition map is shown.

Improving added value and productivity of services entails improving both value-in-exchange and value-in-use. Value-in-use is realized by value cocreation, where providers and receivers create value together. In higher education services, value-in-use comes from learners achieving learning outcomes (e.g., knowledge and skills) that are consistent with their learning goals. To enhance the learning outcomes of a learner, it is necessary to enhance and utilize the abilities of the teacher along with the abilities of the learner. To do this, however, the learner and the teacher need to build a consensus about their respective roles. Teachers need to provide effective learning content; learners need to choose the appropriate learning strategies by using the learning content through consensus building. This makes consensus building an important factor in value co-creation. However, methods to build a consensus about their respective roles may not be clearly established, making such consensus difficult. In this paper, we propose some strategies for consensus building between a teacher and a learner in value co-creation. We focus on a teacher and learner co-design and propose an analysis method to clarify a collaborative design process to realize value co-creation. We then analyze some counseling data obtained from a university class. This counseling aimed to build a consensus for value-in-use, learning outcomes, and learning strategies between the teacher and the learner.

In the research field of Human-agent interaction, the uncanny valley is the crucial issue to realize co-existence of human and artificial agents. It is referred to as the phenomenon that human can feel repulsive against the agents whose appearance is considerably humanlike. There has been just theoretical based but not verifiable model providing an explanation for how it occurs. We hypothesized that when human observes that humanlike agent, s/he perceives it as both human and non-human, and the contradiction between the perceptions causes him or her to elicit negative response against it. We conducted the experiment where the participants were asked to judge whether face of agents or a person was depicted as that of a real person, with their eye tracked and their gaze direction estimated. The results indicated that observers had two-step information processing to the agent. Above all, we proposed a model generating the human negative response against humanlike agents, taking into consideration of the function of the brain regions such as amygdala, prefrontal cortex, hippocampus, and striatum. To verify the model, the transition of the emotional value (namely, positive or negative) was simulated on the basis of a qualitative description for the model. It is suggested that the model be proposed which is verifiable with many findings in the field of neuroscience.

We developed a motion assistance apparatus which can be used for neuro-rehabilitation of patients and for the promotion of exercise for the elderly. This apparatus assists only the ankle joints of the equipped person according to the posture of him/her while walking, getting up, sitting, and standing. By using this apparatus, the dorsiflexion and plantarflexion of the ankle joint of the equipped person can increase while walking. Therefore, this apparatus can prevent stumbling and promote walking and various motions.

The color research has much interest in exploring emotional responses to color stimuli, which are referred to as color emotions. In general, color emotions are measured through psychological methods and represented as values on rating scales. For further understanding of color emotions and development of their effective application, conceptual structure of color emotions and rating scales need to be shared among various disciplinary fields. One typical approach in sharing such knowledge is to clarify concepts as an ontology. In this paper, we specified color emotions based on psychological attributes and built an ontology with an ontology development environment Hozo and a top-level ontology YAMATO. As a result, our ontology describes 16 psychological attributes which are measured by rating scales on the basis of a concept of human awareness. Approximately 74.7 percent of instances of the psychological attributes, which are frequently used in articles on measurement of color emotions, are accounted by the 16 psychological attributes. By adding descriptions of psychological attributes which are infrequently used and have not been fully elucidate in the color research, our ontology would increase utility for knowledge sharing and be feasible in prompting scientific and engineered approaches to understand human mind.

We developed a method which can diagnose damage on a gear tooth surface by using laser beam without a rotary encoder. This method is as follows: 1) The tooth bottom, the tooth tip and their two medians are detected by the differentials of the laser reflection data. 2) The gear rotation speed is calculated with these four positions, and interpolated according to the rotation fluctuation. 3) By using the calculated gear rotation speed, the measured data can be converted corresponding to the gear rotation angle. Thus we diagnose gear tooth surface damage without being influenced by rotational fluctuation. We did diagnosis experiments and we made contour maps show diagnosis accuracy. From these maps, we got the following conclusions: 1) The accuracy of damage diagnosis is the same level regardless of the presence or absence of a rotary encoder. 2) The cycle of rotational fluctuation hardly affects the accuracy. 3) Bigger fluctuation amplitude makes the range accuracy worse, however the position accuracy improves.

Emotional responses to color stimuli are referred to as color emotions, and relationships between color attributes and them have been discussed in past studies. In the current study, we reveal relationships between color rendering properties and color emotions to provide a light system controlling color emotions along with color rendering property.We developed such a system and conducted two evaluations of the system performance and color emotions under the light generated by the system. The system we developed in the current study has four color (red, green, blue, and yellow) LEDs to enrich color rendering property. In addition, we conducted a subjective assessment to explore relationships between color emotions and color rendering properties. In the assessment, we studied color emotions against facial skin irradiated by various illuminants with different color rendering properties. We prepared sixteen different light conditions and the participants of the study were asked to respond to the stimuli with fifteen Likert scales. As a result, we confirmed that the color rendering property certainly affects color emotions, and we obtained regression models to represent color emotions by color rendering properties. Furthermore, we found that male and female test subjects experience different color emotions based on skin color.

We developed a motion assistance apparatus which can be used for neuro-rehabilitation of patients and for the promotion of exercise for the elderly. This apparatus assists only the ankle joints of the equipped person according to the posture of him/her while walking, getting up, sitting, and standing. To confirm the effectiveness of the apparatus, we carried out the experiment of walking with wearing the apparatus, and the angle variation of the ankle was measured. From the results, by using this apparatus, the dorsiflexion and plantarflexion of the ankle joint of the equipped person can increase while walking. Therefore, this apparatus can prevent stumbling and promote walking and various motions.

Many studies have been conducted during the last two decades examining learner reactions within e-learning environments. In an effort to assist learners in their scholastic activities, these studies have attempted to understand learner mental states by analyzing participants' facial images, eye movements, and other physiological indices and data. To add to this growing body of research, we have been developing IMS (Intelligent Mentoring System) which performs automatic mentoring by using an ITS (Intelligent Tutoring System) to scaffold learning activities and an ontology to provide a specification of learner's models. To identify learner mental states, the ontology operates based on theoretical and data-driven knowledge of emotions. In this study, we use statistical models to examine constructs of emotions evaluated in previous psychological studies, and then produce a construct of academic boredom.

For designing the rounded communication between human and agent, humanlike appearance of agent can contribute to human understandability towards their intension. However, the excessive humanlike-ness can cause human to feel repulsive against the agent, which is well known as the uncanny valley. In this study, we propose a model providing an explanation for how the human negative response is fomred, based on the brain regions and its function, including the amygdala, hippocampus, cortex and striatum. This model is described with quantitative reasoning and simulated. The results indicate that as human observes a humanlike agent, the emotion goes negative and the brain regions were more activated in comparison with the case human observes a person. © 2014 Springer International Publishing.

Recently, the research in intelligent educational systems has much interest in exploring data from academic settings to understand learners behavior and mental states. We have been developing IMS (Intelligent Mentoring System) which performs automatic mentoring by using an ITS (Intelligent Tutoring System) to scaffold learning activities and an ontology to provide a specification of learner's models. To provide learner's models in mentoring, the ontology describes qualities and quality values on awareness which are conceptualization of subjective evaluation. To specify relationships among qualities on awareness in academic settings, this study organized academic emotions in the psychological research and proposed their ontological descriptions.

Recently, the research area of intelligent educational systems has increasingly approached Educational Data Mining, which addressed a variety of learner aspects such as mental states. Although data of eye movements is promising in realizing mental states of learners, we have to empirically understand features of eye movements in advance due to difficulty in handling them. Toward the final goal of automatic estimation of learner mental states based on eyes, this study experimentally described changes of eye movements by confidence of correct answers. In our experiment, participants were asked to answer multiple-choice problems and respond to questionnaires about the problems, with the results indicating that transitions of eyes in initial processes where all choices were scanned differed depending on the levels of confidence.

Recently, the research in intelligent educational systems has much interest in exploring data from educational settings to understand learners behavior and mental states. To further understand learners based on such the data and sophisticate supportive interventions by educational systems, knowledge of relationships between learners' behavior and mental states must be shared. To propose a framework to share the knowledge, this study attempted ontological descriptions for learners' eye-movement and mental states in solving computer-based multiple-choice problems. The current study forms a technical basis for development of an IMS (Intelligent Mentoring System) in which an automatic mentoring function is implemented with ITS (Intelligent Tutoring System).

Recently, impressions of web pages formed by visitors have become an important tool to increase the number of repeat visitors to the web page. Therefore, the management of knowledge on consumers' impressions obtained in several study fields is an essential task in current industrial design. However, no methods that realize such knowledge management have been established. Thus, our study proposes and implements a knowledge management method that can effectively provide knowledge of impressions to web designers to help them in building attractive websites. We introduce a framework for the description of impressions that depend on perceptual fluency, which can serve as an useful indicator of pleasure. We can extract the features of objects that affect impressions on the basis of perceptual fluency. We specify the relationship between objects and impressions by modeling the concepts of awareness, perception, and self-report on the basis of an ontology development environment Hozo and a top-level ontology YAMATO. We then instantiate a case where a person has a good impression of a Web page, and we describe the relationship between a perception and a stimulus in such a case. Our approach demonstrates that ontological modeling of impressions helps us to understand the correspondences between affections and physical irritations.

This study aims to examine the relationship between visual fatigue and prefrontal cortex (PFC) activity during a backlit text reading task. We conducted a 10 min luminescent sentence reading task in a dark room. Time-course of oxy-Hb in the PFC was measured using a 22-channel wearable near-infrared spectroscope (NIRS). Subjective fatigue was assessed using the "Jikaku-sho shirabe" questionnaire. The critical flicker-fusion frequency (CFF) and LF/HF were assessed as objective index of fatigue and autonomic nervous system. 192 trials consisted of twelve healthy young volunteers and twelve LED treatments were analyzed. In the analysis, a linear non-gaussian acyclic (LiNGAM) model for causal discovery was introduced in addition to the correlation analysis. Our results suggested that the development of visual fatigue and central fatigue is related to the deactivation conditions in the medial PFC and Left PFC.

<p>Mobility scooters are three or four-wheeled electric-motor-driven wheelchairs with a handlebar for steering. In the use environment of a mobility scooter, there may be many obstacles that need to be avoided, such as pedestrians on a narrow sidewalk or cars on the edge of the road. These obstacles not only interfere with smooth driving but also induce mental strain on the user such as fear and anxiety, which may cause a decrease in comfort and an accident. The purpose of this study is to estimate physical obstacles in the road environment when driving a mobility scooter from operation information and physiological information of the user obtained during driving. We measured skin conductance, manipulation amount of accelerator, steering torque, and acceleration value in the horizontal plane when driving on a few courses with physical obstacles. We tried to build a machine learning model that estimates the type of physical obstacles in the course based on these time-series data by using a random forest classifier.</p>

<p>This study aims to examine the benefit of visual fatigue evaluation based on prefrontal cortex (PFC) measurement by near infrared spectroscopy (NIRS), and to evaluate fatigue on various colored LED backlit text. In our method, eight adult participants read a LED backlit text 10 minutes in a dark room. Before and after task, subjective fatigue questionnaire "Jikaku-sho shirabe", subjective visual fatigue questionnaire (SVF), and the critical flicker-fusion frequency (CFF) were measured. The time-course of oxy-Hb in the PFC was measured using a 22-channel NIRS. Heart rate variability (HRV) was measured to access an autonomic nervous system balance. Six types of white LEDs were used for backlighting: general white as a control (W), high color-rendering white (WH), bluish white (WB), reddish white (WR), greenish white (WG), and yellowish white (WY). In our results, correlation analysis suggested that the development of central fatigue is related to deactivated condition in the mPFC, and the development of subjective visual fatigue is related to deactivated condition in the LPFC. In comparison between LED treatments, in WH condition, high subjective central fatigue, CFF decay and LF/HF were revealed, and the PFC inactivated. In WG condition, low subjective central fatigue and CFF decay revealed. Our findings suggested that participants get tired on high color-rendering white LED backlit text, and were less tired in greenish white LED backlit text. Also, NIRS provides benefits in the evaluation of visual fatigue.</p>

<p>Obesity has become a serious problem worldwide in the modern society with the change of food culture. Obesity would be a cause of many diseases such as hypertension, dyslipidemia and diabetes. Considering obesity, we examined satiety closely related to energy intake. In this study, we focused on the number of chews and chewing frequency, which are factors of satiety, and analyzed the influence on it. In the experiment, we compared satiety, hunger and masticatory muscle activity among the conditions when participants ate the same amount of food under the condition that the number of chews and the chewing frequency were changed. As a result, the datas of the increase rate of satiety and the masticatory muscle activity showed that participants of low weight and normal weight who consumed foods by chewing many times, and participants of obesity who consumed foods with large masticatory muscle activity for each chew could feel fullness earlier. Using this result, people can reduce the energy intake naturally by feeling fullness early in the diet, and it is expected to improve the eating habits of obese people.</p>

<p>In recent years, new indicators focusing on cognitive activities have been developed for the purpose of improving the accuracy of lie detection tests. It has been pointed out that eye movement is particularly effective. In addition, there have been many reports of lie detection based on gaze movements and blinks. But these indicators are greatly influenced by human intentional control. In this study, we focus on the pupil diameter, which is difficult for humans to control intentionally. we verify whether it can be used a new index of lie detection method. In this experiment, we analyzed changes in pupil diameter when presenting a critical question and a non-critical question in a CIT issue. As a result, when the pupil diameter was dilated at the time of question presentation, the pupil diameter at the time of presentation of the critical question was smaller than that at the time of non-critical question presentation. This is because the cognitive load is suppressed by making the participants unaware of the contents when the critical question is presented. This result suggests that the pupil diameter can be used as a new index in the lie detection method.</p>

<p>When Fragrance influences the psychophysiological state of a person, and consequently the volume of fingertip plethysmography is affected by the aspiration of the aroma. Recently, studies have attempted to employ aroma in various scenarios, such as in smells for medical care, relaxation, environment improvement for productivity enhancement, and others. However, the type of fingertip plethysmograph behavior produced near the stimulation threshold that allows feeling the smell remains unclear, especially near the discrimination threshold, where fragrance intensity changes. In this study, the stimulation threshold at the time of aroma suction and the emotional change around the discrimination threshold are evaluated by analyzing both a questionnaire and fingertip plethysmograph measurement. As human biological systems produce chaotic behaviors such as heart beat rhythms and the plethysmograph, chaotic time series analysis was performed for fingertip plethysmograph measurements. Then, we evaluated psychophysiological responses during aroma suction, using a feature of the chaotic behavior in fingertip plethysmograph. In addition, we analyzed and compared with the conventional method LF / HF. As a result, we confirmed the fluctuation in the fingertip plethysmograph even when the research participants responded that it did not smell in the subjective evaluation. Further, we confirmed the fluctuating tendency of the fingertip plethysmographs when the Largest Lyapunov exponent increased as the aroma hen the maximum Lyapunov exponent increased as the aroma the fluctuation of the fingertip plethysmographs becomes large and corresponds to the uncomfortable values in from the subjective evaluation. In addition, the relaxing effect by smell was confirmed using LF / HF.</p>

<p>This study aims to examine the time-series effect of cognitive load on the cerebral blood flow. We used machine learning to extract features related to cognitive load in any brain region and time domain of oxy-Hb measured by NIRS. We conducted a 5min auditory stroop task. Ten healthy young volunteers participated in this study and performed five repeated experiments. The brain function in the prefrontal cortex was measured using 22ch wearable near-infrared spectroscopy. Cognitive load was assessed using questionnaire surveys NASA-TLX and accuracy. The results of analysis using neural network (NN) and layer-wise relevance propagation (LRP) suggest that around the inferior frontal gyrus (IFG) contain important features for classifying cognitive load, and oxy-Hb in IFG may increase with cognitive load. Our LRP method can extract common features to oxy-Hb within the same group, visualize the effect of cognitive load on individual data.</p>

<p>In this research, we conducted to generate Convolutional Neural Network (CNN) models which can distinguish between cancer cells and normal cells in the large intestine under 11 input image size (number of pixels) conditions (50×50 px, 100×100 px, 200×200 px, 300×300 px, 400×400 px, 500×500 px, 600×600 px, 700×700 px, 800×800 px, 900×900 px, 1000×1000 px) for the purpose of clarifying the effect of the input image size on the cancer cells identification in colorectal cancer pathology image. When we conducted to train models, we calculated model accuracy by 5-fold cross validation in each condition. As a result, the accuracy rate increased as input image size increased in 3 conditions (50×50 px, 100×100 px, 200×200 px). However, the accuracy rate of the model trained by 4 conditions (700×700 px, 800×800 px, 900×900 px, 1000×1000 px) were lower than that trained by other conditions. These results are thought to be caused by setting the kernel size to the same value in all conditions. Therefore, we concluded that it is necessary to determine an appropriate kernel size and stride for each input image size which can extracted features included in the input image when we conduct to generate a model for the purpose of distinguish between cancer cells and normal cells in the large intestine.</p>

<p>In recent years, the number of traffic accidents has decreased due to the development of driving support systems. Driving assistance system supports to the driver, however when driving assistance is not appropriate, the driver may be bothered by the assistance and distrusted of the system. In order to provide appropriate support, a method that driver behavior prediction has been proposed. In this research, we measured seat pressure distribution and driving information using a driving simulator, and predicted driving behavior using a convolutional neural network. In this research, we predicted five driving behaviors: "Go straight ahead", "Turn right", "Turn left", "right lane change", and "left lane change". As a result, it was possible to predict with an accuracy of 65.4% in inter participant evaluation and 96.4% in overall evaluation.</p>

<p>In this research, to develop a gait feedback training system optimized for individuals where trainees can efficiently train features that do not satisfy ideal walking using a deep learning, we examine classification and features visualization of gait using Convolutional Neural Network (CNN) and Grad-CAM. In the experiment, the thumb-floor distance of right foot was measured when young people walked normally and when they walked with a brace, to limit their movement. Further, these data were clustered to 3 clusters using k-shape method. And these data were learned and classified as input data and using these cluster as the label. As the result, the accuracy was 86.07%. In addition, the part where the feature in thumb-floor distance appears were visualized as heat map using Grad-CAM and it is confirmed that usefulness for gait training.</p>

<p>Mobility scooters are three or four-wheeled electric-motor-driven wheelchairs with a handlebar for steering. In the use environment of a mobility scooter, there may be many obstacles that need to be avoided, such as pedestrians on a narrow sidewalk or cars on the edge of the road. These obstacles not only interfere with smooth driving but also induce mental strain on the user such as fear and anxiety, which may cause a decrease in comfort and an accident. The purpose of this study is to estimate physical obstacles in the road environment when driving a mobility scooter from operation information and physiological information of the user obtained during driving. We measured skin conductance, manipulation amount of accelerator, steering torque, and acceleration when driving on a course with physical obstacles. We tried to build a model that estimates the type of physical obstacles in the course based on these time-series data by using a random forest, which is one of the machine learning methods.</p>

<p>Smoke separation is accelerating towards the Tokyo Olympics and Paralympics. However, there is no method for monitoring harmful chemical substances in tobacco smoke in real-time, so there are scientific lacks against non-smokers health whether the smoke separated space have no impact or not to their health. The main chemical substance in tobacco smoke is nicotine. We decided to build up a model first for nicotine. In this study, we use a newly developing gas analyzer which measures nicotine in real-time. From that nicotine measurement result, analyzes the nicotine behavior with FEM in a simulated space. Then make it learn the parameters of the nicotine behavior model, and using a different modeling and learning algorithm than general multi-agent reinforcement learning, we visualize the nicotine behavior in tobacco smoke through the new achieving model. First, we confirmed the availability of nicotine analysis with the gas analyzer.</p>

<p>The purpose of this study is to clarify feature for discriminating emotion while expressing fake smile. First, we experimented to record face expression in video. The flow of the experiment was following; we first displayed one of emotion-eliciting films, and participant expressed (fake) smile just after watching the film. At the end of the experiment, we let them assess what emotion they felt with two self-report measures. We prepared six emotion-eliciting films in order to record various face expression with various emotion. Then, we obtained face landmark positions from the face on the recorded video using a method proposed by Kazemi et al⁽¹⁾. We focused that eyes and mouth corners move well while smiling, and defined 15 feature values related to that landmarks. Furthermore, we grouped the feature values in emotion scores by clustering. There are significant differences in eye area between happiness and disgust, and in some mouth corners' feature values between happiness and fear as a result of comparison of clusters. we considered these differences resulted from disgust/fear, but negative emotion.</p>

<p>The caution making is used in various daily situations as a preventive measure of tumble accident which is one of the labor accidents. The purpose of this study is to propose a display that incorporates the Fraser-Wilcox illusion and has a higher attention effect. Fraser-Wilcox illusion is a type of visual illusion in which a part of a still image appears to be moving. We compare the attention effects of shapes that induce the Fraser-Wilcox illusion and those that do not. As an evaluation method of attention recall effect, the pre-cueing task which was an experimental technique to measure the characteristics of visual attention was used. The correlation between the amount of visual illusion and the target reaction time was examined. As a result, a strong negative correlation was observed. The larger the amount of visual illusion, the faster the detection of the target. These results suggest that the Fraser-Wilcox illusion has an alerting effect on visual illusion.</p>

<p>The purpose of this study is to estimate the condition of second-hand goods from images using machine learning methods. Assuming that the condition evaluation of second-hand goods has a relationship with the damage that exists in them, we proposed and verified a method to estimate the condition of second-hand goods based on information of the damage. In this study, we chose men's dress shoes for condition evaluation and chose four types of damage: wrinkles, worn out of heel, dirt on outsole, dirt on inside as damage considered to be related to condition evaluation. (1)We create a CNN model that determines the presence or absence of damage, and (2) create a classification tree model that evaluates the state of the product from information on the presence or absence of damage and (3)We evaluated the condition of second-hand goods by combining two created models. As a result of verification, it was shown that the proposed method could evaluate the condition of used goods.</p>

<p>Warning at the time of switching from automatic to manual operation of vehicle is required to assist the driver appropriately. In this research, we focus on the driver's immersion in typing with a personal computer during automatic driving. We examined changes in driving operation after switching to manual operation when the warning sound is presented under four conditions including two conditions of beep sound and two conditions of presentation voice. The beeper was able to react quickly but it was found to be dangerous for driving operations. In voice presentation, it was suggested that a warning simply presented could not assist appropriately but presenting hazard avoidance instruction enabled quick reaction and assisted in calm driving behavior.</p>

Currently, the visually impaired find it difficult to use mobility support facilities because they depend on ambient environment such as ambient sound and snow cover. Further, people who acquired visual impairment in the later stages of life also find it difficult to use these facilities. Therefore, in this study, we aimed to develop a walking guidance system using 3D sounds that will not be influenced by the surrounding environment and will be easy to use. Further, we verified the induction effect in the blindfold state and the ambient sound. This system was implemented with a head mount display that is capable of producing 3D sound. Under a visual restriction environment, blindfolded participants were asked to walk toward a real or 3D guiding sound with three types of ambient sounds, and then, we measured the walking locus. We found that the 3D sound tends to provide veering tendency opposite of the ambient sound. However, the guiding accuracy of the 3D sound is similar to that of real sound.

<p>筆者らはこれまで，蝶の色彩美に焦点を当て，特に美麗種の多いアゲハチョウ科の配色傾向を調査してきた．このための主な手法として，蝶画像を色の類似度で分類し，各クラスターの色彩分布を示すグラフから配色傾向を分析してきた(梯ら,2018; Kakehashi et al. 2018)．本研究ではこれらのデータを統計的に処理するためにガウス混合モデル（Gaussian Mixture Model: GMM）による分析を試みた．GMMによって得られた結果を予め定義した配色型と比較したところ，主に次の三つの傾向が得られた．①明度コントラスト（低明度の面積の方が支配的），②類似彩度（低彩度の面積が支配的），③類似色相（橙～黄緑の面積が支配的）．これらの結果はOu and Luo(2006)の実験的色彩調和論をはじめとした先行研究の結果のうち，equal hue, equal chroma, unequal lightnessと一致していた．</p>

In this paper, we evaluated the mental stress recorded against corner entry speed while riding a mobility scooter by measuring skin conductance. A mobility scooter is a four-wheel electric wheelchair with a steering handle. These scooters, known as “Senior Car” are generally sold to the elderly. Some of these scooters have a system that automatically decelerates to approximately 70% of the maximum speed depending on the steering angle at corners. The preset deceleration rate can cause discomfort to the driver. To provide appropriate assistance without any discomfort, evaluating the mental stress of the driver is necessary. Skin conductance (SC) is one method for measuring sweating. It can measure mental sweating related to sympathetic nerve activity. We modified a commercially available mobility scooter to be used as an experimental wheelchair such that it can be controlled using a PC. We measured SC in a task that involved turning right into a corner at a constant speed using an experimental wheelchair and tried to evaluate the mental stress against the corner entry speed. Resultantly, a positive correlation between the amplitude of SC and entry speed was obtained. In addition, we performed a questionnaire to evaluate the ride discomfort after the experimental task. The questionnaire results demonstrate that the evaluation value of the items “No Fear” and “Safety” decreased as the speed increases. This result was consistent with the measurement results of the SC amplitude. From these results, we conclude that fear against corner entry speed could be evaluated using SC.

In recent times, robots such as, service, communication, and cooperative robots, have been developed to assist humans in various tasks. Smooth communication is an important parameter to ensure appropriate assistance by robots. In communication, understanding the comprehension status of the partner leads to an easy-to-understand explanation. However, it is difficult to accurately understand the comprehension status of the partner. The purpose of this study is to clarify the relationship between thinking status and pupil size and blink rate. Word association tasks were conducted, wherein subjects associated a theme from five auditory words. Pupil size and blink rate were measured during the task using gaze measurement. The result showed that pupil size tended to increase during thinking and decrease with association. Decrease in blink rate distribution was observed after presentation of a word, following which a temporary peak occurred before association. However, no change in blink rate was observed after the association. Based on the results of the present study, it was revealed that thinking may be associated with pupil size and blink rate, and it can be suggested that the eye could be an index to evaluate thinking status.

Determination of the optimal amount of information is essential to avoid confusion. In this study, the “Stroop task” was used for information processing, and the cognitive load was gradually increased by sequential adding of information. We designed and conducted two tasks that originate in the “Stroop task;” these two tasks feature significant differences in cognitive load. We also measured brain activity using near-infrared spectroscopy (NIRS) under the assumption that such activity could be used as an index of cognitive load. Both tasks were associated with increased oxy-hemoglobin levels in the prefrontal area, and the task with a higher cognitive load was associated with a more substantial increase in oxy-hemoglobin; this indicates that oxy-hemoglobin levels may be used as an objective index for the evaluation of information-associated cognitive load. In addition, this study included two groups (Group A and B) of patients who undertook the tasks. Based on the timing and the number of mistakes, it was observed that subjects in group A demonstrated lesser information processing ability than those in group B. The findings of the present study thus indicate that oxy-hemoglobin variation may depend on the information processing ability of the subject.

It is important to mesure energy metabolism for human health. Current mainstream activity meters show large inaccuracies in the estimation of energy metabolism for small mothion. In this study, we examined the estimation method of energy metabolism during exercise with small motion by using electromyography and heart rate, both of which are correlated with energy metabolism. We chose wrist curl a type muscular strength training to represent movement with small motion, and considered the flexor carpi radials and biceps brachii as target muscles. Multiple regression analysis was used to estimate energy metabolism; this allowed us to estimate that there is an error of about 25% when estimating total energy metabolism. Although comparison was carried out at ten-second intervals, estimation was difficult because of the large variation of experiment values. This study suggests that total energy metabolism during exercise with small motion can be estimated by using myoelectric potentials and heart rate.

At nursing care sites, patients undergo an exercise regimen as rehabilitation and training for walking disorders. However, it is presumed that their motivation decreases over time because in a nursing home, walking exercises are performed in a closed space and are repetitions of monotonous movements. Therefore, it is necessary to develop a more varied and effective rehabilitation or training method. The purpose of this study is to develop a varied and effective training method wherein the trainee walks while viewing virtual reality (VR) images. Specifically, in the proposed method, the trainee views VR images of walking on a slope, while actually walking on a treadmill, and he/she controls the physical load. In this study, we conducted an experiment using a composed training system with the aim of analyzing the influence of watching VR imagery of walking on a slope on the braking and driving forces while walking on a horizontal treadmill. We improved the VR system to automatically adjust the treadmill speed from the position information of the trainee. Then, comparative experiments of the braking and driving forces were performed for cases of walking while watching images of level and slope grounds. As a result, the walking speed, braking force, and driving force tended to decrease when watching downward slope images compared to watching level ground images. Additionally, the walking speed, braking force, and driving force tended to increase when viewing upward slope images.

Students in many universities, technical colleges, and technical high schools are practicing their cutting skills using general-purpose machining tools. However, there may be insufficient guidance due to the lack of the number of instructors and a limitation on the number of general-purpose machining tools that they can use. In this research, we propose a system to support the learning of cutting skills using a virtual reality head-mounted display. Preparing students to use machining tools require many training sessions, especially for cutting using a general-purpose lathe. A user can experience three types of content: drawing, tools, and the general-purpose lathe. This learning system was provided to seventeen students, and we conducted a questionnaire on its ease of understanding, appearance, enjoyability, intuitiveness, and operability. Resultantly, enjoyability and intuitiveness were highly evaluated. The results suggest that our proposed system can be used as a learning method. To improve the operability of the VR system, we can make use of specialized controllers or additional force presentation devices.

<p>It is generally agreed that the risk of falling increases with age. Falls in elderly individuals decrease their daily life activity level due to physical damage and increase fear of falling again. It is proposed that many elderly falls occur when walking, and in most, the cause is stumbling. Therefore, the aim of this study was to analyze the factors of stumbling during normal walking to prevent falls. Then we determined the thumb-floor distance, the angle of lower limb joints and the ground reaction forces. Furthermore, we calculated the moments of the lower limb joints from the obtained results. And compared these values during normal walking between elderly group and young group. Consequently, a significant positive correlation was found between the maximum value of the longitudinal component of ground reaction force and the maximum value of thumb-floor distance at terminal swing. Furthermore, when the longitudinal component of ground reaction force was the maximum, the hip flection moment of the elderly group was significantly lower than young group. These results showed that the cause of the fall of elderly is that the driving force at terminal stance is weak due to weak hip flexion muscles and that the knee joint is not sufficiently extended at terminal swing.</p>

<p>In recent years, it is required to design a notification sound which is easy to hear even in the daily life sound. It is necessary for the notification sound to use sound of frequency below 2000 Hz that the elderly can hear easily. Therefore, in this research, we aimed to evaluate how the influence on the recognition and impression of the notification sound is caused by the difference in the timbre. In this experiment, the participants heard three kinds of timbre of sine wave, rectangular wave and sawtooth wave. The threshold of listening sound pressure level was measured, and the questionnaire to evaluate impression of notification sound by semantic difference method was conducted. As a result, we found that the sine wave is familiar and does not give much tension or discomfort. In addition, we found that the rectangular wave and the sawtooth wave are less familiar and give a sense of urgency and discomfort. It was suggested that by changing the timbre of the notification sound, it is possible to design a notification sound that the intention is more easily heard and accurately transmitted.</p>

<p>Demand for vital sign measurement systems have been increasing. Non-invasive systems are required that it can monitor health status over a prolonged period. Conventional devices can easily measure pulses. However, as the noise due to perspiration is superimposed on the waveform of the pulse, the peak of the pulse becomes unclear, so it is unsuitable for measurement for a long time. In this study, we devised a pulse measurement method that can deal with the above problem. We obtained the acceleration waveform from the carotid artery skin surface, using the acceleration sensor. These waveforms are pulse waves caused by the expansion and contraction of the carotid artery. By the measurement procedure that they devised, six of ten people were able to be measured the pulse. In addition, it became possible to measure the pulse by other four people straightening the neck and attached the sensor to the appropriate position carefully.</p>

<p>In recent years, various researches on abnormality detection using machine learning have been conducted. However, when an abnormality is detected, it is necessary for engineers to find the cause of the abnormality and deal with it. If it is possible to estimate the cause of the abnormality at the time of detection, engineers can prepare appropriate repair tools and replacement parts. Machine downtime can be shorten by this estimation. In this research, we proposed new method and conducted experiment to verify its usefulness. In the proposal method, we use convolution neural network (CNN) to detect an abnormality and estimation a cause of it. Operating sound and current value were used for input. In the experiment, we chose geared motors as a model of a rotating machinery, and we used proposed method to them. As a result, we got an average accuracy rate of 82.41% on detection of abnormality, and also got average accuracy 67.84% on cause estimation of it. From these results, it was shown that the proposed method is effective for detection of abnormality and cause estimation of it for rotating machinery.</p>

<p>In recent years, it has been said that the number of mastication and mealtime tends to decrease with the spread of fast foods and softening of foods. Decrease in the number of mastication causes many problems such as obesity, shrinkage of the jaw, loss of natural teeth. In this study, we examined a method to increase the number of mastication by changing the subjective texture of food by presenting mastication sound with bone conduction headphones when chewing food, based on a cross-modal. Moreover, we investigated what kind of mastication sound is effective for increasing the number of mastication from the viewpoint of texture. The result showed that there was a possibility of increasing the number of mastication. On the other hand, a certain result was not obtained with regard to the texture, because the texture which has the correlation with the number of mastication was different depending on the food. However, as the result of a questionnaire of the texture, the correlation between the texture of heavy and the number of mastication was found. Therefore, we confirmed the possibility of increasing the number of mastication by controlling the subjective heaviness of food.</p>

<p>In recent years, cognitive support research has been conducted to reducing the number of traffic accidents. There is a possibility of confusing the driver if the cognitive support is not appropriate. Therefore, there is a need for a method to support driver's perception of dangerous objects without hindering normal driving behavior. In this study, we measured driving behavior and cerebral blood flow when moving sound was used as cognitive support. By using the moving sound as the cognitive support, the brake operation at the time of avoiding the danger became moderate. From the results of cerebral blood flow, it was found that information can be transmitted with the same cognitive load as in normal operation. As a result, the possibility that the moving sound can appropriately assist perception of the surroundings of the driver has been shown.</p>

<p>Light emitting diodes (LEDs) have been widely used for the luminescent text display in automobiles. The purpose of the current study is to evaluate fatigue reduction characteristics in LEDs luminescent text display. For this purpose, we construct a model to estimate the participant's fatigue from physiological index. In the experiment, participants read Japanese sentences that were represented by backlight emitting for ten minutes. Brain activity were measured for the physiological index, and fatigue evaluation questionnaire was answered. The luminescent text display was implemented with LEDs of six kinds of white color. We constructed the deep learning model that inputs brain activity, and outputs "fatigue", "no fatigue" or "no change". The output was calculated from fatigue evaluation questionnaire. The model was able to estimate the fatigue with a validation accuracy of 86.4 %. Thus, the model showed the possibility that estimate fatigue based on the brain activity.</p>

Typically, it is difficult to transmit visual images in words. The transmission of the visual image includes various information. Therefore, efficient communication is necessary. In this research, we focused on the relationship between onomatopoeia and modalism, and conducted a survey to measure the distance. We asked for general group and design skill holding group. Transmission of visual image in communication, we collected 200 words of onomatopoeia that are frequently used. Then we carried out relevance research on sensory modality, using the MD (Modal Differential) method. Then, we conducted a hierarchical cluster analysis and the semantic structure analysis which is called as SS analysis. As a result, we drew the SS graph for each cluster, which shows the relevance and order-related structure. The results were compared between the general group and the design skill preservation group, and discussions were made on the differences depending on the attributes of the subjects.

<p>The increase in the aging population is one of the common problems in the world. Many elderly people experience difficulty in standing up from their seats owing to muscular depression. To improve the QOL (Quality of Life) and ADL (Activities of Daily Living) of elderly people, improvement of the traffic environment is necessary. We developed an assistive seat to help elderly people rise from their seats in a bus with fewer loads. We verified the performance of the assistive seat through EMG analysis. In conclusion, the developed assistive bus seat was effective in helping the user to stand up without difficulty.</p>

<p>It is difficult for users of electric wheelchair who suffer from quadriplegia or have compromised motor skills to manipulation of the joystick finely. Also, there is a risk causing a loss of control upon impacts with steps. In this research, we design a system that performs to assist fine control of the speed depending on the height of the step. This system uses a laser range finder to detect the step from remotely, and uses the valuation of human exposure to whole-body vibration in order to estimate discomfort that arose from the step height and step climbing speed. The potential of this system is also verified and manifested to reduce the impact caused by collisions with steps.</p>

<p>Currently, Japan is entering a super aging society, and as the population of elderly people increases, the number of care recipients are also on the rise. The burden on nursing care homes is increasing; specifically, the burden on caregivers in excretion assistance is huge. Therefore, the need to develop nursing care equipment to support such tasks is increasing. In this study, a preliminary investigation was conducted to determine the possibility of excretion time prediction using data from the diet of a care recipient. We requested and obtained meal data and excretion data from experiment collaborators. We attempted to design a system that predicts the excretion time by a process similar to machine learning using a neural network that corresponds to the two sets of data. We can roughly predict the excretion time. The result showed that it is possible to predict the excretion time from the meal record. In the future, we will include details of the food we eat and biological information to improve prediction accuracy.</p>

<p>Sign language is used for communicating with deaf persons. To facilitate the communication between the deaf and the hearing communities, a sign language translator is required. However, sign language varies from country to country. Furthermore, the number of words is increasing as society develops, and as a result, artificial translation has become challenging. Therefore, substantial research has been carried out to recognize and translate sign language by means of computing methods. In this paper, we propose a sign language recognition method that can adapt to video images captured by general-purpose cameras, and not by special input devices such as a depth sensor. According to this method, sign language regions are extracted from moving images by interframe difference image processing, and are recognized by using a convolution neural network (CNN). As a result of the one-person-leave-out cross validation experiment to recognize 30 words, the highest validation accuracy observed was 90.0%. Thus, the proposed method's potential for sign language recognition is manifested.</p>

<p>In this paper, we aimed at obtaining knowledge about the effect of adding a virtual glass frame and additional information to the Virtual Reality(VR) image on VR sickness. The virtual glass frame is an object model imitating a glass frame arranged in the VR space. The frame is arranged to cover the peripheral visual field when viewing a VR image using a head mounted display. We created VR contents such a spin room and carried out comparison experiments between conditions where the virtual glass frame was added to the VR image using a questionnaire and bio-information. It was found that addition of the virtual glass frame to the VR image reduces the occurrence of vection in the peripheral vision and that the evaluation value of VR sickness decreases. Moreover, adding horizontal information to the frame reduces the confusion of vestibular sensation; furthermore, it was found that the evaluation value of VR sickness further decreases.</p>

<p>Automation level 3 of an automobile is an automatic operation in the normal state;, however, the operation is switched to the driver when it is determined that operation of the system is difficult to continue. In previous study, there were cases where automatic operation cannot be smoothly switched from manual operation owing to motion other than driving, or a decrease in arousal level during automatic operation. It is considered that estimating the condition of the driver can support smooth switching. In this study, we estimated the state of the driver while automatic driving using seat surface pressure. Results indicate that the estimation was approximately 100% in the individual evaluation and 78.4% in the overall evaluation using machine learning. This suggests that it is possible to estimate the driver's state while automatic driving.</p>

<p>Recently, some measures to prevent traffic accidents have been introduced; for example, information presentation by a head up display (HUD), vibration at the handle, and sound signals. However, the index to verify these utilities has not been established yet. According to a preceding study, the relationship between brain activity and negative emotion was observed by the near-infrared spectroscopy (NIRS). Moreover, other studies show that mental stress is related to galvanic skin reaction (GSR), respiration (RSP), and blood volume pulse (BVP). In this paper, we considered danger recognition to be associated with negative emotion and mental stress situation, and examined whether the status of danger recognition can be observed by combining bio-information. We observed that danger recognition is correlated to brain activity and GSR. Additionally, the oxy-Hb and GSR were observed to change when recognizing danger. This result suggests that measuring brain activity and GSR can be used to establish whether danger is recognized or not, and the potential of evaluating presence of danger recognition by using bio-information</p>

<p>LEDs are widely used for backlighting of text indicators in vehicles. It is desirable that such text indicators have high legibility and low fatigue-inducing qualities. Suitable spectral distributions for text presentation in vehicles have not yet been clarified. This study examined legibility and fatigue when participants watched various text presentations. Participants read Japanese sentences that were presented by backlight emission, for 10 minutes in a dark room. The backlight indicators employed two color LEDs. The participants were asked to answer questionnaires to evaluate legibility and fatigue. Critical fusion frequency, brain activity and electrocardiogram were measured to evaluate fatigue objectively. Based on the questionnaire results, both colored backlight equal in legibility. Subjective evaluations showed that fatigue is more likely to be caused by high color rending white backlighting. On the other hand, objective evaluations (critical fusion frequency, brain activity and electrocardiogram) showed that fatigue is likely to be caused by high color rending white backlighting.</p>

<p>It is said that music is effective for controlling feelings. For example, we can feel relaxed to listen to slow music. This paper proposes a generation system which composes music by means of a genetic algorithm. The composition which is generated by the system consists of the melody part, the chord part, and the bass part. The system can specify the atmosphere of the generated composition by choosing adjective words; "lightness", "intensity", and "complexity". To reflect the effect of adjective points on atmosphere of generated composition, the system adjusts the parameters of an evaluation function. As a result of a composition generating experiment, similar melodies and rhythms were found in each composition. To evaluate the appropriateness of the function about reflecting the effect of adjective points on atmosphere of generated composition, we conducted a questionnaire investigation. As a result, satisfactory answers are obtained for a specific adjective word.</p>

<p>In recent years, the number of people who suffer from dementia has increased as the aging population increases. Since a standard method for the treatment of dementia has not been developed, there has been an attempt to combine the treatment of dementia with complementary and alternative medicine. Among complementary and alternative medicine methods, music therapy has attracted attention because of its low risk of side effects. Currently, group therapy is the mainstream of music therapy in Japan because of insufficient number of music therapists. As a result, care for individuals has not been sufficient. Therefore, it is necessary to provide pleasant music for each patient that can be automated. For this reason, a music selection system is required, and this system can be realized by measuring emotions using biological information such as brain activity. In this study, we analyzed brain activity while listening to classical music using near infrared spectroscopy. In our experiment, participants listened to four types of music with different characteristics. The results show that the oxy-Hb rises when participants feel uncomfortable, while the oxy-Hb decreases when participants dislike the music. This indicates that emotions and preferences while listening to music can be identified using brain activity.</p>

<p>It is argued that, in human communication, the impression of a message is determined by nonverbal information. In this paper, we evaluated the relationship between human sensibility and eye movement. We conducted two kinds of experiments to induce a change in the psychological state. We used pictures of beverages as visual stimuli. A single image was applied in one experiment type, and double images were applied in the other experiment type. We carried out a questionnaire survey regarding the interest level with respect to the displayed image and measured eye movement with a gaze measurement system. The results indicated that the gaze time related to the displayed images with higher interest was longer in the double images experiment type. The number of blinks increased or decreased according to perplexity. The number of saccades increased or decreased in response to the perplexity levels in the double images experiment, and the change of saccades related to the interest in the single image experiment. These results showed that the eye movements have a potential for evaluating the psychological state.</p>

In recent years, with the development of IoT technologies, biological information of users has been easily acquired and accumulated. It is required that user's cognitive or emotional state could be accurately identified from such biological information. Various researches have used machine learning or deep learning however, generalized methods to handle biological data have not been established. In this paper, we propose a more versatile method using convolution neural network of deep learning to extract features from individual biometric information which is multi-channels time-series data representing user&rsquo;s cognitive. We demonstrate a new method of extraction and distinguishment of features from cerebral blood flow, skin temperature, respiration and skin conductance measured during specific task conditions in the stroop experiment.

<p>We developed a new power assistance suit without the need for electricity. This suit is to support care worker's assistance behavior, e.g. lifting-up and holding motion. It is made of only rubber and nylon belts. This suit can assist the equipped person's arms and back with two kinds of equipped belts. One is for the waist, attached from the shoulder to the back of the knee via his/her back. The other is for the arms, attached from equipped person's forearm to the back of the knee via his/her back. To confirm the possibility of this suit, first, we interviewed medical doctors and physical therapists. Their comments were, it is important to adjust the length of each belt for an equipped person adequately, and it is effective for the assistance of motion in a half-sitting posture and in transfer motions between a bed and a wheel chair. Furthermore, we measured the variation of maximum output force and %MVC of the entire body muscles during lifting and holding motions using the suit and not. From the result, by using this suit, the maximum lifting weight of the equipped person was increased by over 8kg, and the arms and back of the equipped person were assisted effectively.</p>

<p>Recently, value creation from view point of service has been widespread. A service process is regarded as a kind of interaction between service providers and receivers. Sharing and managing such the knowledge on human Kansei are of growing significance to providing product's values focused on customers. However, any sufficiently-advanced technologies that realize such knowledge sharing and management have been established. One typical approach is to clarify and systematize concepts as an ontology which is a basis to describe knowledge. In the current study, we discuss a method to develop an ontology describing knowledge on human Kansei and introduce several existing ontologies. The ontology will serve as a useful reference to develop value recognition map and be of assistance to designers.</p>

<p>We developed an automatic damage diagnosis system of the gear tooth surface using a laser beam. Only attaching a laser sensor in a gear box, we can grasp the size and place of the damage on a gear tooth surface without using a rotary encoder. We confirmed the limit of this method against the velocity fluctuation. Furthermore, to utilize this system in the practical field, we developed a new system which was small, light weight, and low cost. The accuracy between new and conventional system was compared, and we confirmed the results were comparable level.</p>

<p>Recently, improvement of design capability is one of important issues in industry However, it's really difficult to share knowledge of design possessed by experts with novices, because the knowledge exists depending on their skill and degree of prociency. Thus the current study aims to excavate the experts' Knowledge. We then described design objects related to action to design electronic components.</p>

<p>In recent years, various driving support system has been developed as traffic accident prevention. However, they are unless transmitted to the intention that led to providing support, there is a possibility that the rather confusing the driver. Therefore, when the adoption of the driving assistance system, consider plenty the human visual attention to contribute significantly to cognition, without disturbing with the normal driving behavior, it is required a method for supporting recognition of the surrounding dangerous object. In this paper, we investigated the reaction time when subjected to experiments using dual task, has been presented a variety of auditory stimuli as lead cue stimuli. As a result, when using auditory stimulation with sound image movement, it tended to the reaction time of the central issue and the peripheral issues becomes earlier were observed. The result showed that the possibility of inducing an attention without considering the driver by using sound image movement as a visual assistance.</p>

<p>In recent years, the number of dementia elderly person has been increasing. Various cognitive rehabilitation is underwent in nursing homes. However, it is difficult for them living far away from their home to undergo cognitive rehabilitation. Therefore, if they undergo cognitive rehabilitation in remote locations, many of them can improve dementia. In this research, we examined cognitive rehabilitation by cooperative between remote locations. When they undergo cognitive rehabilitation in remote locations, improvement of their presence is needed and it is assumed that cooperative operation can improve their presence. We chose a maze done by two persons cooperatively as concrete cooperative operation. In our experiment, firstly we compared brain activity between doing the maze and calculating because calculation is used as cognitive rehabilitation. Secondly we examined the possible of improvement of their presence during doing the maze. The result shows that the maze and calculation caused similar brain activity and cooperative operation improve presence. It reveals effectiveness of cognitive rehabilitation by cooperative between remote locations.</p>

<p>Various walking assistance devices which promote rehabilitation and exercise for elderly have been developed. However, these devices can't support walking for exercise, because many elderly feels fatigue and bored in assisted walking. Therefore, the aim of our current study is to assist not only physically but also mentally elderly walkers to keep up their motivation. Concretely, we propose a walking promotion method that walkers listen to the beat sound adjusted according to their emotions with the ankle support assistance device. In order to confirm the effectiveness of the method, we conducted two experiments. The first experiment has four conditions where walkers have the device assist or not and the beat sound assist or not. As a result, the gait was improved with the device and sound assist where target cadences were adjusted according to their walking cycle. The second experiment has two conditions where walkers assisted sound adjusted according to their emotions or not. As a result, awakening levels perceived by the walkers remarked higher than ones in usual walking. These results showed the effectiveness of the method and possibility of the control of the emotion while walking.</p>

<p>Many traffic accidents in car driving are associated with sleepiness. It is possible to prevent these traffic accidents by detecting sleepiness of the drivers while driving their car and making them pay attention to driving. Since the possibility that cerebral blood flow is related to sleepiness has been suggested, it is assumed that it would be able to evaluate sleepiness by the changes in cerebral blood flow. In this paper, we measured changes in cerebral blood flow during driving a car using a portable near-infrared spectroscopy device. The results show that oxygenated hemoglobin that is included in the cerebral blood flow tends to decrease with the rise of the sleepiness level evaluated by facial expression and that there is the significant difference in mean oxy-Hb between the two sleepiness level. From this vital reaction, it is suggested that the possibility of sleepiness evaluation of a driver using near-infrared spectroscopy and detection of driver's sleepiness at the level that driver's consciousness become low.</p>

<p>In this research paper, we measured brain activities by near-infrared spectroscopy (NIRS) when a person is feeling the discomfort caused by vibration. In particular, we examined the effect on brain activation by the presence or absence of notice sound at the time of the start of vibration. We used a variance of oxygenated hemoglobin (oxy-Hb) as an evaluation index. We observed a difference in tendency of brain activation between the presence and absence of signal which affect participants in a surprising way. High negative correlation was observed between discomfort to vibration and brain activation around the medial prefrontal cortex, when value of surprise at the start of the vibration is small. High positive correlation is observed between discomfort to vibration and brain activation around the prefrontal cortex outer portion, when value of surprise at the start of the vibration is big. Possibility to evaluate the discomfort is suggested on the basis of brain activation which reflects the changes in the psychological state by surprise.</p>

<p>In recent years, the research in intelligent educational systems has much interest in exploring data from educational settings to understand learners behavior and mental states. Therefore, the means to grasp psychological state of learner has been demanded from a more variety of physiological indicators. In present study, it aims to grasp how the learner approach to the question. By measuring the brain activated by using the multiple-choice questions, this study is to analyze the relationship between confidence and brain activation at the time of answer the question. As a result, in the prefrontal cortex frontal pole vicinity, tended to increase oxyHb when there is high confidence of the answer. In addition, answer time tended to be quick when there is high confidence of the answer. The result shows that it may be able to grasp the state of learner by measuring combination of the change of oxyHb and answer time.</p>

<p>A number of traffic accidents have occurred due to human errors such as inattentive driving and drowsy driving. Sleepiness is considered as one of the causes of these traffic accidents. For this reason, establishment of the sleepiness evaluation method during driving a car has been required. In this paper, we measured changes in cerebral blood flow during driving a car using a portable near-infrared spectroscopy device. The results show that oxygenated hemoglobin tends to decrease with the rise of the sleepiness level evaluated by facial expression and that there are negative correlations between oxygenated hemoglobin and sleepiness level in many trials. This suggests the possibility of sleepiness evaluation of a driver using near-infrared spectroscopy.</p>

<p>高等教育においては，教育サービスの提供者と受給者が目標とする学習成果についての合意形成を行うことで，高い利用価値を共創的に実現することが期待される．本研究では，合意形成の過程を可視化するために，合意された話題の意味内容の概念化を試みる．具体的には，メンターと学習者のカウンセリング事例を取り上げ，サービス提供者，受給者，コンテンツ・チャネルとして発話内容を位置づける．</p>

<p>本研究では，テーマパークにおいて携帯情報端末を用いて混雑情報を取得することができる来場者の満足度を，マルチエージェントシミュレーションを用いてモデル化した．具体的には，期待不一致モデルをもとに，アトラクションに対する効用とアトラクションに並ぶ行列人数とのトレードオフの関係を考慮するようなモデルを構築した．その上で，来場者の満足度向上の観点からテーマパークにおける混雑情報の提供方法を検討する．</p>

Recently, view point of service has been widespread. In learning domains, service providers have been required to adopt it and to provide values co-created by teachers and learners. The value co-creation takes place in joint sphere where providers and customers make interaction, and the value is subjectively perceived by receivers (learners in learning service). That is, a mental state of leaner is one of possible value in the co-creation. The current study aims to develop a conceptual framework which describes mental states of learners for specifying values in the co-creation. As a result, we proposed ontological descriptions of learning event which consists of a teacher, a learner, learning materials, and a learning action by the learner. In concrete terms, we proposed ontological descriptions for receiver states in a learning activity, on the basis of a conceptual model of value co-creation. The ontological descriptions are useful for sharing knowledge on values perceived by learners to design learning services.

We developed a walking assistance apparatus which can be used for neuro-rehabilitation of patients. This apparatus assists only the ankle joints of the equipped person and he/she can raise his/her leg by using the stretch reflex to bi-articular muscle. To confirm the effectiveness of the apparatus, we carried out the experiment for the patients of walking with wearing the apparatus, and the angle variation of the ankle was measured. From the results, by using this apparatus, the dorsiflexion and plantarflexion of the ankle joint of the equipped person can increase while walking. Furthermore, to grasp the lifetime limit of the apparatus, we developed the fatigue test apparatus for the power transmission device of the walking assistance apparatus: from the motor to the worm-helical gearbox through the flexible shaft. This apparatus supplies the power with the motor which is the same as the motor which is used in the walking apparatus. As a result, we could define the limit to replace of the flexible shaft according to the condition of the user.

In recent years, the way to alleviate the disease by using of music that is so-called music therapy, has become popular in Japan. The music therapy is a therapeutic pedagogical technique which is carried out towards the recovery of cognitive and emotional disorders by using of music. Due to awareness of health in the super-aging society of Japan, demands for music therapy is increasing. However, it is difficult to appeal a preference of music for patients who have disorder of speech. Therefore, objective measurement method of emotion, like a measurement method of brain activity, is necessary for provision of music that fit each patients. In this paper, we conducted analysis of brain activity by using of Near-infrared spectoroscopy during listening to music. In our experiment, subjects listened to six kinds of music that combined various tempos and keys. The result shows that each combination of keys and tempos caused different brain activity. It reveals that emotions during listening to music can be identified by using of brain activity.

<p>近年日常的にエージェントが利用されるようになり、今後はエージェントとしてのロボットの普及が加速すると推測される。当該状況下では、人ーエージェント間のインタラクション時の人への負担を軽減することが、普及を促進する上で非常に重要になる。本研究では、人ーエージェント間のインタラクションを負担が軽いと推測される人間のインタラクションに近づける目的で、人が表出する特徴的キューを特定する心理実験を行った。</p>

When a company or an individual disseminate their information on advertising and sales promotion. Between several people who is a client and a designer or a designer and an illustrator and a photographer have a scene to share the visual image. In the transmission communication of visual image, onomatopoeia is able to convey information with a sense of reality. Onomatopoeia because the words that are interpreted by the sense modality. (the sense modality : Visual, Auditory, Olfactory, Gustatory, Tactual) Therefore, in this study, it was intended from association with sensory modality of onomatopoeia in the transfer communication of visual image, and to clarify the relational structure of onomatopoeia. Elucidation of the semantic structure was included ambiguity of onomatopoeia. It will enable the discussion about how to use and understandable meaning of onomatopoeia. The onomatopoeia of 200 words were investigated by the MD method and analyzed using the SS analysis and cluster analysis. And we assume that the semantic interpretation is determined by the strength of the association of the Modal. I discussed the rank structure. As a result, it is considered onomatopoeia of the group that there is a direction of understanding meaning is effective communication to convey the impression with vague. In addition, differences were observed in the technique of description and usage, selection onomatopoeia between expert and novice. It depends on skills and knowledge.

Sharing and managing knowledge on human Kansei obtained in several study fields are of growing significance to providing product's values focused on customers. However, few sufficiently-advanced technologies that realize such knowledge sharing and management have been established. One typical approach is to clarify and systematize concepts as an ontology. In the current study, we developed an ontology to describe correspondence of affective meaning spaces analyzed by semantic differential method. First, we collect literatures reveal the affective meaning evaluated through the five senses and their inter-modality. Second, we described the concepts on the basis of an ontology of color emotions. Then, we demonstrated and evaluated the usefulness of the ontology in a design situation. As a result, we confirmed that the ontology will serve as a useful reference and be of assistance to designers.

This is research on &ldquo;Pause&rdquo; in the speech language producing from a printed book by Transliteration Person: Transliteration Person produces a talking book for visually-impaired people from a printed book. &ldquo;Pause&rdquo; on the speech language is required faithful to original printed book, because Transliteration Person has to reproduce the author's intention. So we experimented in how to use &ldquo;Pause&rdquo; on the speech language with Transliteration Person. As a result, it was shown that &ldquo;Short Pause&rdquo; inserted by Transliteration Person is important to produce faithful talking book from original printed book: &ldquo;Short Pause&rdquo; is 0.5 second or less. In the future work, we would like to show clearly that &ldquo;Short Pause&rdquo; is concerned with comfortable rhythm and tempo for faithful speech language from original printed book.

<p>本研究では、多色配色に対する審美的評価ルールをシミュレートする計算モデルをBPニューラルネットワークで構築した。SD法の心理学実験を通して理論的枠組みであるHelmut Lederの心理学モデルを定量化し、それに基づき計算モデルの構造を決定し、学習率などのシステムパラメータを遺伝的アルゴリズムによって最適化した。性能テストの結果、本計算モデルは多色配色の審美的評価値に対して高い予測能力を有した。</p>

Recently, view point of service has been widespread. In learning domains, service providers have been required to adopt it and to provide values co-created by teachers and learners. The current study aims conceptualization of learners states extracted from utterance contents agreed by a teacher (mentor) and a leaner along with conversation analysis in the previous study. As a result, we proposed ontological descriptions which clarify ASIS/TOBE states perceived by teachers and learners.

<p>学習者の知識・理解状態に加えて心的状態を考慮した学習支援システムの実現のためには，心的状態に関する知識記述・管理が必要である．心理学の分野では，質問紙の評定データと統計モデルに基づいて，学習行為に関わる感情の構造を明らかにする試みがいくつかある．本研究ではそれらの知見を統一的に捉え共有することを目指し，因子分析などの統計モデルを用いて表される感情の次元に関するオントロジー記述を試みる．</p>

Some pauses are found in natural reading a book aloud, which work to support smooth communication. They are supposed to be working in transliterating service for visually impaired people. In this study, we propose types and functions of the pauses in that situation. We conducted an experiment to record speeches of 3 transliterators. As a result, 2 types of pauses were extracted; pause as visual information, and pause as easy-to-understand. In the both type usage, an expert made better use of the short pauses which put 0.5 second or less in the speeches than other transliterators. This suggests that the short pauses have an important function to support the smooth communication. In the future work, we analyze pitches and other parameters to specify the characteristics of expression with the short pauses.

<p>ヒューマンエージェントインタラクションにおいて不気味の谷は重要な課題である．不気味の谷は，人型エージェントに対して人間／非人間の二つの知覚情報処理を行う過程における情報間の齟齬により発生すると考えられるが，この齟齬が不気味さを誘発するメカニズムに関しては統一的な説明がなされていない．本研究では，知覚対象の情動評価に関わる脳部位の機能を定性表現によりモデル化することで，このメカニズムの説明を試みた．</p>

ヒューマンコンピュータインタラクションにおいて不気味の谷は重要な課題である.人間は人間に酷似したエージェントを観察する際に,観察初期の段階では人間を観察する場合と類似した知覚処理を行うが,観察時間が経過するに従って人間を観察する場合とは異なる知覚処理を行うことが実験的に示唆された.このような2つの知覚過程はこれまで感情の二重経路説によってモデル化され,不気味の谷は二つの知覚過程による情報間の齟齬によって引き起こされると解釈されてきた.本研究では,感情の二重経路に加えて海馬の働きに着目することによって,エージェントに対する不気味さ誘発メカニズムのより詳細なモデル化を試みる.

<p>学習者の知識・理解状態に加えて心的状態を考慮した支援を行う知的メンタリングシステムの実現のためには，心的状態に関する知識記述・管理が必要である．学業に関する文脈で学習者が経験する感情は包括的にAcademic Emotionと呼ばれており，学習者の特性や一時的な心的状態との概念的な関係は明確でない．そこで，知的メンタリングシステムにおける学習者モデルに記述されるべき感情の概念を整理する．</p>

Besides skills in musical expression, exact timing among music players is required in excellent performances. In order to synchronize each respective sound of music among players, they must use a variety of information. In case of four-hand performances on the piano or performances by a band, for example, information such as musical sounds, body movements and gazes is considered to play critical roles. In this study, we empirically investigated roles of the information as communication channels between music players. We conducted an experiment where pairs of players performed music with a constant tempo, one varying its tempo and one including a long rest on the piano. They played the three sets of music both in a face-to-face setting where they could watch each other and a back-to-back setting where they could not. According to the results of the experiment, we then proposed communication models of players in music performances.

<p>視線は，計算機による学習者の心的状態の理解において有益な指標となることが期待されているが，視線データの分析手法は必ずしも確立していない．本研究では，計算機を用いて学習者が多肢選択問題に回答する場面を想定し，学習者の視線データを取得する実験を実施した．そして，学習者の視線から問題回答プロセスを質的に記述し，問題回答に対する確信度を推定するための分析手法を考案を行う．</p>

<p>近年は，自信や興味などの認知，困惑などの情動といった学習者自身の心理的な側面に焦点を当てた学習者の理解が試みられている．それらに基づいた学習支援を行う場合には，あらかじめ学習者の行動と心理状態に関する概念を明示することが，具体的な支援方略の決定に役立つと考えられる．本研究では，多肢選択問題を回答する学習者の視線と心理状態を例にとり，その知識を記述するためのオントロジーの構築を試みた．</p>

In calligraphy, calligraphers express their emotions in calligraphic characters and that impress people who view the characters. Calligraphers generally acquire tacit techniques for expressing emotions by themselves and, therefore, the knowledge on the technique is difficult to be shared. We experimentally studied features of calligraphic characters expressing five different emotions, using writings created by an expert calligrapher. We classified fundamental strokes comprising the characters based on stroking methods defined in &lsquo;Eijihappo&rsquo; and then extracted knowledge on emotion expression techniques in drawing strokes.

機能指向から感性指向へと消費者にとっての価値が移り変わりつつある現在,膨大なデザイン候補から個人の嗜好に合ったものを抽出することは,情報化社会に置かれた我々の課題と言える.一方で,こうした,情報を高速かつ確実に処理するコンピュータ技術も発展を遂げている.そのため,コンピュータにより人間の感性を反映したシミュレーションを試み,その抽出精度を評価する.コンピュータ上では,人間の脳をモデルとしているニューラルネットワークにより感性評価を行うものとする.結果として,人間の脳活動において存在が示されている感性次元を適用することにより,その学習精度は十分に有効であるが,評価精度については実用レベルに至らないと言える.

<p>色刺激に対する印象はSD法および因子分析によってその因子構造が明らかになっている．また，色刺激に対する感情については，感情価，覚醒，支配という三つの次元から解釈されている．印象と感情は互いに深く関連していると考えられるため，それらに関する知見を統一的な視点から理解する必要がある．そこで本研究ではオントロジー工学の方法論を用いて，色刺激に対する印象と感情の関係構造を表現することを試みる．</p>

Human kansei is one of the important factors in recent product development. Thus, management of knowledge about consumers' feelings is an essential task in current industrial design. For the purpose of such the knowledge management, development of a descriptive method to identify feelings is needed. In this paper, we measured implicit attitudes to color combinations using affect misattribution procedure, and we analyzed proportions of positive responses to each stimulus. The results showed that chroma contrast was explained by a concept of perceptual fluency. According to the finding, we discuss significance of the perceptual fluency which links physical features and impressions from the viewpoint of ontological description. The discussion indicated a need for development of a descriptive method for correspondence between physical features and impressions explained by perceptual fluency (e.g. symmetry, contrast, prototypicality).

本稿では脳という高度な情報処理機構を手本に新しい情報処理機構を構築するという目的の下,逐次分析(SA : Sequential Analysis)を中心として強化学習(RL : Reinforcement Learning),自己組織化マップ(SOM : Self organizing Map)を組み合わせた脳型意思決定モデルを提案する.神経科学や経済学における意思決定に関する先行研究との比較と,本モデルを基にしたコンピューターシミュレーション実験によって,本モデルが人間の意思決定をどの程度再現しているか検証した.その結果から,本モデルの妥当性と,特に大脳新皮質における意思決定処理の再現性が示唆された.

<p>プロダクトやサービスに対して人間が持つ印象は，品質向上に役立てるための重要な知識である．本研究では面接調査によって抽出した印象に関する知識に対して，オントロジー工学の方法論を適用することでその記述を試みる．また，特定の対象世界において印象が言語表現によって観察されることから，印象の記述方法を提案すると共にそれらの区別について考察する．</p>

This study is to constitute a framework for understanding aesthetic sensibility by introducing a mental model called the content space which is defined as a set of all possible contents of consciousness. An aesthetic sensation is a conscious process, thus it belongs to the content space. On the other hand, all mental process must be realized by neural activities which were initially evolved in human ancestors. Because aesthetic sensation is a content of consciousness appeared only in human, it is highly probable that its origin can be explained by a correlating process of biological evolution and cultural development. As the bridge between mental and physiological processes, the emotion and mood relating aesthetic sensation are also considered in the proposed model.

本研究の成果は，大きく二つである．ひとつは色彩感情，五感の印象，および学習感情に関わるオントロジーの構築である．もうひとつは共感演出の検証である本研究で構築されたオントロジーでは，心理的属性に関わるいわば心の内容を扱ったものである．限られた文脈が限定されるものの，心理的反応のオントロジー記述としては非常に新しい試みである．このオントロジーを援用して共感を演出する際に，ドメインモデルの規約を与える概念を提供することができると考えられる．

本研究では、人間の感じる「調和感」の生成メカニズムを検討した。特に、色彩を中間言語として、様々な感性情報（視覚、聴覚、嗅覚等の感覚）を結びつけるための基礎的な研究とその実環境への応用に関する研究を行った。その成果は以下の通りである。(1)「調和感」生成に関する基礎的な知見を得たこと、（2）調和空間の構築に関する比較的普遍性の高い知見を得たこと、(3)「調和感」生成に関する脳科学な観点からの知見を得たこと、(4)「調和感」の統合的管理のためのオントロジーを構築したこと。なお、これらの成果に基づいた「調和感」の統合的管理システムの構築が今後の発展的課題である。

本研究の目的は,対象についての印象評定に関する知識を抽出し構造的に管理すると共に,それを利用した印象評価手法の開発である.知識を構造的に管理し利用するためには,オントロジー工学の方法論による知識の体系化が有用である.本年度は印象の評定に関する知識を記述するためのオントロジー構築を行い,次の成果を得た.
(1)色彩感情を測定する際に用いる評定尺度が,記述的次元の評定尺度と評価的次元の評定尺度の二つに分類されることを実験的に検証した.因子分析の結果からそれら二つの次元の評定尺度が区別されることを確認すると共に,快-不快,覚醒-睡眠,支配-服従という感情の次元を表す評定尺度は記述的次元と共通の性質を持ちながらも区別されることが示された.その上で,記述的次元と評価的次元の評定尺度においてどれほど,快-不快という評価の意味が共有されているかについて実験的に検証すると共にオントロジーとしての表現を試みた.
(2)Self-Assessment Manikin尺度を用いて単色に対する印象評価の実験を行い,構造方程式モデリングによって明度,彩度,色相という色彩の三属性と,快-不快,覚醒-睡眠,支配-服従という感情の三次元の関係を明らかにした.その上で,得られたモデル式を記述するための概念を整理した.
(3)色彩感情を表す評定値のような主観的データの記述・共有を目指し,既存の上位オントロジーに基づいて主観的な質・量に関する概念の定義を行った.また,色彩研究でよく用いられる色の表現方法である色空間モデルに関する概念と共に,主観的データの記述である統計モデルに関するオントロジー記述を行った.
以上により,印象の評定結果を記述するための概念的基盤として,オントロジーを構築することができた.このオントロジーを参照することによって,人間が抱いている印象(を表す知識)を概念レベルで管理・共有することが可能であると考えられる.