Early detection of mild cognitive impairment (MCI) is an important challenge in aging societies. Electroencephalography (EEG) and eye-tracking have independently been explored as potential biomarkers; however, their integrative effects remain insufficiently examined. This exploratory study investigated whether combining EEG spectral features with gaze variability may provide complementary information for MCI discrimination. EEG signals were recorded using the 10–20 system, and spectral power features were extracted. We compared three models: (a) high-dimensional EEG features, (b) L1-regularized feature selection (LASSO), and (c) integration of the selected EEG features with gaze variability. Performance was evaluated using leave-one-out cross-validation and area under the ROC curve (AUC). Model (a) yielded limited discrimination (AUC = 0.52). Feature selection increased AUC (0.64), and additional integration of gaze variability further increased AUC (0.78). These preliminary findings suggest potential complementarity between neural and behavioral variability measures.

Chronological changes in microsaccade (MS) rates and pupillary responses were analysed as metrics of cognitive workload across five distinct tasks. Particular focus was placed on the impact of MS extraction parameters, specifically the velocity threshold, on the sensitivity of these metrics. It was found that the velocity threshold significantly affects the rank order of MS rates during cognitive load comparisons. Latent attention levels were estimated using both metrics, confirming that measurement deviations influence the plausibility of the assessment. Methodological limitations and considerations for future workload analyses are also discussed.

The feasibility of evaluating cognitive workload during the oral reading of single strings is examined using microsaccade (MS) rates and pupillary changes, in addition to the reported conventional ocular metrics. This study utilises an open dataset containing eye movement measurements for reading aloud four types of character strings under both short and long length conditions. Both binocular MS rates and pupillary sizes are chronologically summarised at 500 ms intervals. The resulting temporal changes are analysed to discuss the underlying cognitive dynamics during the tasks.

This survey reviews AI goals and tools for: (1) preparing educational materials, (2) interacting with teachers and students, and (3) assessing the results and providing feedback with (semi-)automatic methods.As a summary, we provide the crucial challenges to be tackled and discuss the associated ethical concerns.

The assessment of the quality of automatic Japanese speech transcription systems is influenced by orthographic variation and homonyms, though character and word error rates are often used as evaluation metrics. Detailed assessments are usually conducted using evaluations and ratings by humans, which limits the survey conditions of the experiment. In the case of this experiment, a paired comparison method was introduced in order to conduct a simple review of 100 sets of transcriptions of short text presentations. The texts were converted into transcriptions using three models of automatic Japanese speech transcription systems. The ratings of the three models were compared using 10 experiment participants and a LLM with three types of prompts. The possibility of aligning the result with human preferences by adjusting the prompts given to the LLM was studied. Also the differences between the judgements made by participants and the LLM were examined and discussed.

Microsaccades (MSs) are often used as an index of mental activity and they may be classified as binocular or monocular using the synchronicity of their appearance in both eyes. The possibility of detecting binocular MSs was examined using the characteristics of eye features. A classification procedure was conducted and performance was evaluated using measurements taken at 500Hz and 1KHz. The validity of classification using monocular eye tracking, and the dependency of data characteristics measured during the experimental tasks were examined. Also, the influence of sampling rates of eye trackers is discussed by comparing the characteristics of binocular MSs measured using two types of eye trackers. Possible applications were also discussed.

Route search behaviour using metro maps was analysed by taking measurements of eye movement. A procedure for extracting areas of interest (AOIs) from eye fixation data with Gaussian distribution using the same 9 metro maps as in the previous study was developed, and the AOI sequences which found appropriate routes were evaluated. Viewer’s search behaviour was examined based on the Viewed features of the AOIs. The three stages of a search, which were suggested in the previous work, were classified using the extracted AOI sequences, and the viewing characteristics of each stage were examined. The most suitable form of illustration for train maps was also discussed.

The International Society for Clinical Eye Tracking (ISCET) serves as a global platform for promoting international consensus on open standards on clinical eye tracking. Originally formed in March 2023, ISCET was created to facilitate collaboration, knowledge exchange, and advancements in clinical eye tracking applications, with the ultimate goal of fostering interdisciplinary research and improving clinical outcomes. Through collaborative and interdisciplinary efforts, ISCET’s current mission is to provide guidance on conducting eye-tracking tasks in clinical settings, unify clinicians’ voices, and maintain reference datasets for normative comparisons. ISCET now has 80 members, spread across the globe. In the Europe/Africa region meeting on January 24, 2024, ISCET established subcommittees to address specific needs. This paper outlines the rationale behind ISCET’s formation, its mission, objectives, ongoing initiatives, and organizational structure. Ongoing work focuses on surveying current international clinical eye tracker usage to inform standards development.

The International Society for Clinical Eye Tracking (ISCET) serves as a global platform for promoting international consensus on open standards in the application of eye tracking to clinical areas of focus. Formed in March 2023, ISCET facilitates collaboration, knowledge exchange, and advancements in clinical eye-tracking applications, with the goal of fostering interdisciplinary research and improving clinical outcomes. Through collaborative and interdisciplinary efforts, ISCET aims to provide guidance on conducting eye-tracking tasks in clinical settings, unify clinicians’ voices, and maintain reference datasets for normative comparisons. Convening during its latest meeting for the Europe/Africa region on January 24, 2024, ISCET established subcommittees to address specific needs. This paper outlines the rationale behind ISCET’s formation, its mission, objectives, ongoing initiatives, and organizational structure. Ongoing work focuses on surveying current international clinical eye tracker usage to inform standards development.

This poster presents a study on detecting Alzheimer’s disease (AD) using deep learning from gaze data. In this study, we modify an existing pre-trained deep neural network model, gazeNet, for transfer learning. The results suggest the possibility of applying this method to mild cognitive impairment screening tests.

Early detection of Alzheimer’s disease (AD) is important but difficult. Screening for AD using neuropsychological tests such as mini-mental state examination (MMSE) is time-consuming and burdensome for patients. Recently, several methods have been reported for detecting AD based on eye movements. However, analyzing eye movements requires considerable effort. Although machine learning from eye movement data is a strong candidate for labor-saving, it requires large datasets. In this study, we modify an existing pre-trained deep neural network model, gazeNet, for transfer learning. For evaluation, we exclusively used data from one participant and fine-tuned the model using data from all the remaining participants. We repeated this procedure separately for each of the 14 participants. The results of eye movement during the antisaccade task were not satisfactory for the discrimination of AD, and detailed analysis suggested that the data might potentially have a correlation with MMSE scores in the mild cognitive impairment range.

The mental workload of older motorists driving various routes on public roads in their own cars was evaluated using estimations of driver’s attention levels based on saccade rates and pupillary changes. Levels of attention while driving were calculated using a Bayesian modelling technique. This was employed due to the extensive amount of data measured over the road course, which was lengthy. Mean attention levels of 5 categories of driving routes were summarised, and routes with turns required additional attention resources. Individual attention levels and confidence intervals correlate with some ratings of NASA-TLX factors, and with the factor for Frustration in particular.

Change of reward expectation during a competitive card game was evaluated using pupil responses. In order to simulate interactive play of participants in a poker game, the game process was fully controlled by the computer player. As a result, pupil response based on reward expectation was examined. Also, participant’s pupil size dilates in response to the change of reward expectation when the computer player provides some expectation.

In order to improve the accuracy of a procedure for detecting symptoms of dementia, features of ocular oscillation are examined, in addition to features of waveforms of pupil light reflexes (PLRs) of both eyes in response to blue or red light pulses directed toward either eye. An experiment was conducted using elderly people with three levels of cognitive performance: Alzheimer's disease, mild cognitive impairment, and a normal control group. In order to classify participants, two regression functions using extracted variables were introduced to the diagnosis. The results confirm the effectiveness of measuring features of ocular oscillation as a method of classifying patients. As PLR responses of each eye were different during light pulse irradiations of the eye on the left or right side, overall performance may be improved by using two independent classifications.

The possibility of measuring temporal change in cognitive workload during tasks is examined using microsaccade (MS) rates and pupillary change. The experimental task was designed as a search for a specific figure, where the task difficulty and reaction accuracy were controlled during the trials. Individual cognitive workload was measured after the experimental sessions were completed, using ratings for NASA-TLX scale. Since the source may be a common one, changes in latent attention resources required for the task were estimated using a designated state-space model, using the observation data in order to synthesise measurement of MS rates and pupillary change. The predicted levels of attention resources correspond to the activity during the performance of the experimental tasks during the trials, and reflected some of the rating scores for workload scales. Also, the ranges of confidence intervals for attention resources correlate significantly with the ratings for information processing at the stage where visual stimulus is presented during tasks.

The possibility of predicting individual scores of critical thinking ability from estimated ratings of essay texts was examined using machine learning techniques. First, the feasibility of predicting critical thinking ability scores from scores of personality and literacy of science and technology surveys was confirmed. Scores for personal characteristics were estimated using automated assessments which were based on a pre-trained language processing model of two types of essay texts. Second, variables employed in the surveys used to predict scores of critical thinking ability were replaced step by step with estimated values from the essay texts. Prediction performance decreased gradually, and the accuracy with all estimated values was above the level of performance which was predicted using the essay text analysis. As a result, the possibility of predicting scores from essay text assessments was examined.

We applied state-space modelling technique to estimate the cognitive workload of a speech-in-noise (SIN) recall task, based on participants’ oculo-motor responses to speech signals. We estimated common latent attention levels in 15 time bins and observed temporal changes between pupillary dilations and saccade frequencies, given that the both conditions were independent. We also compared two speech type factors (natural vs. synthetic) and three levels of signal-to-noise (-1dB, -3dB, and -5dB) using the estimated parameter distribution. The comparison of experimental factors provided us with insights into differences in participants’ processing of spoken information during a SIN recall task.

A procedure for calculating the probability of levels of dementia is proposed using features of binocular pupil light reflex (PLR) to a chromatic light pulse on either eye. The PLR waveforms of 101 elderly participants, consisting of patients with Alzheimer’s disease (AD) and mild cognitive impairment (MCI) or belonging to a normal control group (NC), were measured during four experimental conditions. Three factor scores were calculated from the PLR waveform features for each response. Responses were summarised and the differences between the two eyes were calculated in order to detect asynchronicity of PLRs in response to a light pulse on either eye. Pupillary oscillation was also measured separately without light pulses, and frequency powers were evaluated. Probabilities for the level of dementia of each participant were predicted using two types of regression functions for MCI+AD or AD patients, which were optimised using a variable selection procedure for extracted features. In the results, some variables which represent asynchronous measurement and pupillary oscillation were selected, and the requirement of binocular measurement was confirmed. A prediction procedure for levels of dementia of participants using the optimised functions was proposed, and performance was evaluated.

Visual attention is estimated as temporal latent resource levels using a state-space model and microsaccade (MS) frequency measurements taken during a dual task. In order to optimise the model, MS frequency was evaluated every 0.5 seconds. As a result, the contribution of the accuracy of the response to the task was confirmed by the estimated resource levels.

To improve program learning tools with eye tracking technology, it is crucial to understand fixation points and individual information, providing personalized navigation cues for different levels of expertise programmers. Meanwhile, long short-term memory (LSTM) and clustering techniques revealed important characteristics for eye movement data regarding comprehension performance. This paper is about a spatial analysis by co-clustering with the gaze location among different levels of expertise programmers. Then it predicts the next fixation point based on the human eye movement data by LSTM. Finally, combining the individual background information and the eye movement information, it generates a new indicator, named ‘code comprehension index’, to indicate the current code understanding level on real-time gaze information for individuals, which can be applied to improve the program navigation tools’ effectiveness and efficiency for different levels of expertise programmers.

Temporal visual attention levels and directional bias are evaluated using behavioural responses and microsaccade (MS) frequency measurements taken during a dual task. The behavioural responses were analysed, and the directional bias in the peripheral field of vision extracted using a modelling technique. The observed MS frequency recorded every 0.5 seconds was applied to a state-space modelling technique to extract visual attention levels. The optimised modelling results contribute to the accuracy of the responses and to the estimated levels of attention resources. Also, the positional bias of the resource levels in the peripheral field of vision changes in response to the clusters of the behavioural responses.

The possibility of predicting anticipated scores representing student's characteristics from their essay texts using a machine learning technique was examined. Student's characteristics include critical thinking disposition ability, disaster-prevention conscientiousness, and personality. First, the potential for predicting essay assessment scores from essay texts using machine learning with each set of essay reports or essay comments was confirmed. However, prediction performance of one other set of essays and other scores of student's characteristics was insufficient. When the training procedure was re-designed using both sets of essays, performance for some scores improved. The limited performance was examined and further improvement procedure was discussed.

The relationship between ocular metrics and factor ratings for mental workloads are examined using a Bayesian statistical state-space modeling technique. During a visual search task experiment, microsaccade frequency and pupil size were observed as measures of mental workload. Individual mental workloads were measured using a 6 factor NASA-TLX rating scale. The models calculated generalized temporal changes of microsaccade frequency and pupil size during tasks. The contributions of factors of mental workload are examined using effect size. Also, chronological analysis was introduced to detect the reactions of the metrics during tasks. The results suggest that the response selectivity of microsaccades and pupil size can be used as factors of mental workload.

A procedure for detecting cognitive impairment in senior citizens is examined using pupil light reflex (PLR) for chromatic light pulse and a portable measuring system.Features of PLRs of blue and red light pulses are compared.PLRs of elderly subjects were studied in order to develop a procedure for detection of the symptoms of cognitive function impairment using a dementia evaluation test.PLRs of both eyes were measured using blue and red light pulses aimed at either of the two eyes.The features of PLR waveforms for each eye were remained in comparable level for every group of participant.Three factor scores were calculated from the features, and a classification procedure for determining the level of dementia in a subject was created using regression analysis.As a result, the contribution of factor scores for blue light pulses according to a participant's age was confirmed.

An emotional intensity regression estimation model was created using calculated perceived intensity values and deep learning. In our previous study, we considered emotional intensity using 10 categories and estimated emotional intensity by categorization, but the flexibility of the method was insufficient. In order to solve this problem, an emotional intensity estimation model which takes into account differences in the perceptual intensity value of each category of emotional intensity was used in this study. For this purpose, two types of perceived intensity values were calculated for a Japanese speech corpus of sounds uttered in an emotion perception rating experiment. In the results, the average correlation coefficient between the estimated intensity value and the set intensity value of the sounds was 0.73 for the emotional intensity estimation model when perceived intensity values were used. These results suggest the possibility of successfully estimating emotional intensity using regression.

In order to improve the readability of metro maps, reading behavior for route search has been analyzed using eye movements in the previous studies. The factors of the numbers of route and transfers on the route were examined. Since the metro map images are different between cities, effectiveness of another factor for image features on eye movements was examined using features of saliency maps. The 48 metro maps were classified into three clusters, the difference of feature of eye movements was extracted. As the results, relationship between fixation time and values of the saliency was discussed.

Relationships between features of oculo-motors and perceptual impressions of Valence and Arousal are analysed using viewer's reactions to 67 emotion-evoking photographs. Individual rating scores are compensated for using item response theory, and chronological changes of oculo-motor indices are analysed in response to two-dimensional ratings. These reactions are summarised as regression models, and predicted emotional categories based on oculo-motor reactions are evaluated. Prediction performance is also evaluated using mean similarities for the predicted categories of emotion. While performance improved when these features were added, individual reactions to features should be included in order to improve prediction performance for each participant. Also, temporal features of oculo-motors for Valence and Arousal are selected independently of their contribution to prediction.

Sentences in student's essay reports were analysed to examine the feasibility of evaluating ability of critical thinking disposition in order to develop this ability during a fully online course in a university class. Features of essay reports, dependency of terms, and the representation of selected terms between levels of critical thinking disposition which were measured using a questionnaire were compared. The results show that the frequency of dependencies reflects the level of factor scores of critical thinking disposition and that multi-dimensional scales of these frequencies can illustrate the relationships between dependencies and levels of ability. Also, senses which contained affirmative or negative contents are evaluated using a corpus of affirmative and negative words used in essay reports. These frequencies are influenced by the level of ability as well.

In order to evaluate cognitive performance and the level of task mental workload of older drivers, some features of oculo-motors such as eye movement and pupillary response were extracted during car driving on a public road. The individual cognitive performance of 11 selected older drivers was measured in advance using some conventional tests including Manoeuvre and the useful field of view (UFOV). Changes in the extracted features deviated along with the test route, which was classified into 5 groups according to the type of route. The regression relationships between the features of oculo-motors and the cognitive test scores was created using the LASSO technique, where fitness and feature selections are evaluated. The predicted scores for driver's cognitive performance and dependency of route group were evaluated and the overall possibility of the estimating driver's conditions was confirmed. These results suggest that older driver's eye movements during driving reflect their cognitive abilities and level of mental workload.

Ocular information was observed during a set of dementia tests involving participants with Alzheimer's Disease (AD), with a mild level of cognitive impairment (MCI), or in a control group.The number of participants was 26.Features of changes in pupil size and in the central position of both eyes of participants of all three types were compared.There are significant differences in some of the metrics between the types, in earlier test sessions.The possibility of classification was confirmed using the extracted features, and the contributions of some features were examined.

Recently, there have been many studies of emotion estimation using non-linguistic speech data. In addition, there have been a few studies on emotional intensity. However, failure to read this emotional intensity correctly can lead to errors in the responses humans and machines should make when communicating with each other. In this paper, we developed three models for emotional intensity estimation using deep learning, and examined the accuracy of emotional intensity estimation of a Japanese speech corpus, which resulted in 52.4% accuracy of emotional intensity estimation. We also investigated the correlations between acoustic features and analyzed the properties of acoustic features in order to improve estimation accuracy, and found that the differentiation of gammatone cepstral coefficients varied significantly between intensities.

Pupillary change due to emotion-evoking stimuli and the relationship between pupil responses and emotion-rating behaviours were analysed carefully. Chronological pupil responses were compared by level of emotional impression, which was standardised using item response theory. The results show that mean pupil size changes according to the level of emotional impression, and the maximum size depends on the classification level, such as middle-group negative on the valence scale. Also, the duration of appearance of the significant difference between the various levels of the groups depends on the number of levels. These results confirm the results of some of the previous studies. Therefore, they suggest that pupil response consists of the composite reactions mentioned above. The importance of assessment of the individual rating responses was confirmed in a detailed analysis.

The progress of development of the disposition toward critical thinking in blended and fully online learning environments is compared. Participant’s individual characteristics, including responses to a questionnaire about critical thinking disposition, were measured during a bachelor level university course. Though the measured scores are comparable, the relationships between the scores in the two surveys during the course are different according to learning style. The contributing variables were extracted using linear regression analysis, and the weights assigned to them depended on the style of learning. In particular, variables which contributed to transforming the participant’s ability are discussed. Also, cluster patterns of scores of literacy of science and technology were extracted, and the contributions of the cluster scores was discussed.

We discuss the problem of setting the best number of peers to which a given evaluation job should be assigned, in a Peer Assessment setting. The Peer Assessment is supposed to happen in a large scale class, such as in the case of Massive Open Online Courses. We use a dataset that simulate a large class (1000 students), based on Gaussian distributions of the Student Model features. Such features are related to the student's proficiency, and assessment capability. The number of peers assigned to the same evaluation job was controlled from 3 to 50 in 6 steps using 10-point scale. The abilities of participants were estimated using Item Response Theory. All parameters of IRT models, which is called as Generalized Partial Credit Model, such as "ability", "consistency", and "strictness", were estimated well using MCMC technique; their standard deviation errors gradually decrease with the number of peers. As a preliminary result of optimisation, an appropriate number of peers was 15 as comparing the stadardised errors across the conditions.

In order to extract the relationship between head-eye movements and visual impressions while using Head-mounted displays (HMD) to view images, two types of response task experiments involving visual stimuli were conducted, using several visual angles. Both head movements and visual impressions increased with the visual angles during the observation of moving visual stimuli. Vestibulo-ocular reflex (VOR) gains as a ratio of eye and head movements decreased as the visual angle increased because the eye movement deviations were relatively small. When the viewer's reaction required an action, such as rotating the head or aiming at a target, the strength of the viewer's impression increased, though VOR gains remained at the same level.

The frequency of microsaccades is often used as a measurement of eye movement in order to estimate the level of a cognitive mental workload during an experimental task. When the experiment includes several levels of manipulation tasks, ocular information including microsaccades, saccades and pupil diameters can be measured and compared. The results show that the microsaccade rate reflects a participant's subjective impressions. In addition to measurements of oculomotors, microsaccade rates also reflect the level of effort required by a mental workload, and the mutual relationships between these rates are analyzed using causal analysis.

The management of a fully online course is a key to maintaining learning effectiveness. Some factors of participants and their learning activities were analysed. Features of students who preferred joining fully online courses were extracted and compared across several years. Note taking activity during the course was lexically analysed and the features of notes were extracted. The relationships between these features and learning performance was compared. Five typical note taking styles were extracted, and reflections of learning performance and participants' characteristics were examined.

Tables of values rarely provide enough information to make deep understandings about the underlying data. For millennia, charts, graphs, illustrations and diagrams have been used to visualise the underlying structure of correlated data. This paper demonstrates in brief how diagrams can be used to visualise complex interrelationships using Structural Equation Modeling of IQ.

The relationship between features of an elderly driver’s eye fixations and cognitive performance was analysed. The participants were 29 motorists aged over 60 years who drove frequently. Their eye movements were observed during 16 specific situations along a test course. Their cognitive performance was measured using computer based tests prior to driving. Since some features of eye fixations during driving correlate with the scores of cognitive tests, the possibility of predicting the cognitive scores was examined using features of eye fixations and a multiple regression analysis technique. Prediction performance was also affected by specific features of the fixations and their behaviour.

Oculomotor indicies in response to emotional stimuli were analysed chronologically in order to investigate the relationships between eye behaviour and emotional activity in human visual perception. Seven participants classified visual stimuli into two emotional groups using subjective ratings of images, such as "Pleasant" and "Unpleasant". Changes in both eye movements and pupil diameters between the two groups of images were compared. Both the mean saccade lengths and the cross power spectra of eye movements for "Unpleasant" ratings were significantly higher than for other ratings of eye movements in regards to certain the duration of certain pictures shown. Also, both mean pupil diameters and their power spectrum densities were significantly higher when the durations of pictures presented were lengthened. When comparing the response latencies, pupil reactions followed the appearance of changes in the direction of eye movements. The results suggest that at specific latencies, "Unpleasant" images activate both eye movements and pupil dilations.

A diagnostic procedure to predict the probability of diagnosing a patient with Alzheimer's Disease (AD) was developed using features of pupil light reflex (PLR) waveforms.15 features of PLRs for three colours of light pulses at two levels of brightness were measured.Participants were 12 AD patients and 7 control group subjects.A logistic regression analysis was introduced to identify AD patients using two factor scores of features of PLR.The prediction performance of combinations of factor scores for features of PLRs were then evaluated using a test of fitness.An MCMC technique was introduced to estimate the parameters of the regression functions.The model provides a distribution of the probability of diagnosis of AD patients and control group subjects.

Lexical feature patterns of notes students took during a blended learning course were classified and the student's transitions between clusters were analysed using a formative evaluation. The lexical features were extracted from relationships between the contents of notes students took and the contents presented by the lecturer. Five clusters were extracted using four dimensional features and cluster analysis, and most participant's note taking patterns in the first half of the course transformed from one of two clusters to one of five clusters, which included three additional new clusters, in the second half. This was due to note taking skills instruction given by the lecturer. The contributions of participants' improved note taking skills and their reflections were evaluated quantitatively, and the mean features of notes taken and mean learning performance between clusters were also compared.

The need to know how to code for university graduates has become crucial if they are to achieve their goals of employment after graduation. Post-graduation jobs increasingly require some understanding of how programs work and are developed. The project reported here used HTML and JavaScript to introduce students to the core structure of coding. It uses code that produces a visual result which helps non-IT students overcome their fear of coding. Using an online HTML and JavaScript editor to create a 2D game for a mobile deice introduces students to some basic programming skills in a fun and creative way. This paper reports on a project that tracked before and after sentiment towards learning coding skills. It shows a shift in sentiment from a fear of coding to a better understanding of the potential of coding to produce creative outcomes. However, despite the shift in sentiment across the project, there was still too much reliance on external help - students were not able to initiate programming without some assistance. This suggests, along with the method described in this paper, other methods should also be explored.

Participant's emotional factors for a fully online course were measured twice during course using some metrics which were used for a blended learning course. Self efficacy and reflection during learning courses with instruction of note-taking skills were analysed. The contribution of courses and survey points were not significant for gathered responses from participants except "learning hours". These responses of participant's reflection were affected by their characteristics and self efficacy. The relationships across survey metrics were compared across two courses using a causal analysis technique. Also, changes of metrics during the course positively affected to participant's reflections.

Game development activity using Team-based Learning (TBL) was investigated in order to identify factors contributing to the usability of the product. In this study, three teams from two different countries are compared. As the related factors, the followings were examined to analyze the relationships with usability scores: (1) learning reflection, (2) social media communications within teams, and (3) participants' characteristics and information literacy. Usability scores were conveyed using a System Usability Scale (SUS) by evaluations from the other teams. The participants' characteristics and information literacy were measured before starting the project as a pre-test. The discussions and communications via social media of each group were categorized as: Proposal, Permission, Encouragement, and Acknowledgment, using protocol analysis to examine their contributions towards the usability scores. After completing the study project, a learning reflection questionnaire was completed by all participants to evaluate efficacy, satisfaction and achievement of learning, and difficulties.

To determine the relationship between brain activity and eye movements when activated by images of facial expressions, electroencephalograms (EEGs) and eye movements based on electrooculograms (EOGs) were measured and analyzed. Typical facial expressions from a photo database were grouped into two clusters by subjective evaluation and designated as either "Pleasant" or "Unpleasant" facial images. Regarding chronological analysis, the correlation coefficients of frequency powers between EEGs at a central area and eye movements monotonically increased throughout the time course when "Unpleasant" images were presented. Both the definite relationships and these dependencies on images of facial expressions were confirmed.

Audio information plays a key role in the learning environment. The effectiveness of audio information on note-taking and learning performance was examined using survey data from a fully online course. First, metrics of note-taking activities were measured using the relationships between these lexical features, and their contributions to learning performance were then examined. In particular, the effectiveness of learning materials presented as slide contents or audio information was compared. Second, some relationships between these metrics and the scores of learning performance evaluations were analysed, and the effectiveness of audio information was qualified using regression analysis. The results indicate that audio information significantly affects both the contents of notes taken and learning performance.

The Human-Computer Interaction (HCI) environment, which includes topics taught, teaching resources and evaluation procedure, is changing rapidly. As the body of knowledge about HCI is periodically redefined, course content should be adapted in response to these changes. This position paper presents trends in topics taught, course content and assessment procedures using examples of HCI courses. Methods which encourage students to develop their learning activities and the assessment procedures are discussed in regards to course content and the current situation in the HCI learning environment.

To assess the transformation of a lecturer's term-based concepts presented to students, both the lecturer's presentation and the student's notes in a blended learning course were lexically analyzed and represented in graphs. The contents of these notes were mathematically compared and evaluated. Graphs were then created using the noun co-occurrences in the notes, and the differences in distances between the graph of the lecturer's presentation and the graph of the student's notes were calculated. In the results, students recorded additional nouns to replace the nouns given by the lecturer, and introduced new nouns of their own as substitutes as well. Also, the terms students wrote in their notes were not recorded separately, but were linked with other terms.

In this paper, we propose a new approach for the comparison and matching of documents written in different languages. Our approach is devoted to situations where documents can be conceptually represented by abstract graphs. Our approach builds on previous work yet it addresses some limitations left open by previous work. The benefits of our approach is twofold. First, our approach makes it possible to compare original multi-lingual documents without requiring a translation in a pivot language contrary to other approaches. Second, we propose a highly visual and interactive environment so that human experts can perform both comparison of documents and lexical matching in a seamless way contrary to other approaches where matching and comparison are often considered separately.

In theory, course curriculum, also known as course syllabus, should play an important role in the success of learning. However, in practice course syllabus are often far from playing this role. In this paper, we analyze several course syllabus in order to better characterize learning expectations in the field of HCI (Human Computer Interaction) and Computer Science. We start by analyzing variations of course syllabus descriptions in that field. We report the results of a case study using simple statistics to automatically extract relationships from syllabus documents of HCI courses described online. We further provide some requirements for future syllabus descriptions that seem to be missing in all the syllabus descriptions analyzed. Our aim with this work is to urge further discussions and future work to enhance the content and practice related to course syllabus.

This paper is concerned with the presentation of a collaborative recommendation system that implements a cascade of strategies in order to support the learning process. Similarities between learners are determined by taking advantage of the underlying implicit or explicit personalization and of the non-personalised modes of interaction. In the personalised approach implicit profiles are based on the patterns of behaviour of learners, while explicit profiles are generated from the results of a questionnaire on learning style. The non-personalisation approach relies on the cumulative intervention of a community of learners implied by the recorded frequency of the usage of objects by learners, and by the expert rating of objects by teachers. Content-based and collaborative approaches are combined into a hybrid model that widens the range of objects to which a learner may be exposed. The quality of service of the recommendation system is evaluated by considering the accuracy of its predictive capability on a publicly available data set.

The lexical features of notes taken during blended and fully online courses were analysed to improve learning activity, as the contents of notes can be used to facilitate course evaluation. The lexical contents of student's notes was used to analyse the note taking behaviour in both blended and fully online courses. Features such as the number of terms, the word ratios of student's notes and the degree of coverage of the lecturer's notes were compared. The relationships between learning performance and note-taking behaviour were also analysed. The causal relationships between student's characteristics and note assessments were statistically compared, and contributing factors were extracted.

In order to visualize and analyze the learning process of students who participated in a fully online engineering course, note-taking activity was used and lexical analysis of notes taken by students was made. The features of notes taken were compared with the contents of the lecturer's presentation slides and oral instructions. Also, the latent semantic indexing technique was applied to the student's notes and the lecturer's presentations, and the lexical transformation of the course content was analyzed.

To analyze a chronological accommodation process for viewing objects, viewer's VEP, eye movements and pupil oscillations were measured while six sizes of Landolt circle were presented for 1s, with reviews being asking and then not being asking for their responses right afterwards. The significant features of the metrics were extracted to illustrate the perception process and resulting sizes of visual stimuli. The stimulus sizes affect P1 and N1 delay times of VEP, mean of eye movement lengths and pupil diameters and oscillations during the second 500 {milliseconds}. The feasibility of prediction for stimulus sizes viewed is been confirmed using the indices and support vector machines. Also, both indices of eye movements and pupil oscillations are determined to be significant features for predictions. This result suggests that oculo-motor indices reflect the size of objects being viewed and can be an index of prediction for the viewing sizes.

A fully online learning environment requires support procedures in order to promote pro-active learning. This paper presents the causal relationships between student's characteristics, note-taking skills, learning experience, note assessment and test scores. A structural equation modeling technique is used to trace student's learning activity as note-taking occurs. The results of this modeling technique suggest that key factors and their contributions to course performance can be measured. Also, the factors which contribute to note-taking activity were examined.

In order to create reading support systems, the feasibility of using viewed size predictions was evaluated using features of EEG waveforms. Viewer's occipital single channel potentials were measured while six sizes of Landolt circles were shown. Features such as series of potentials, latencies of P1 and N1 peaks, and 5 vectors of Fourier descriptors (FD) were extracted from the EEG waveforms. The performances of 6-classes were classified and compared across sets of each feature and combinations of these. The EEG waveforms were transformed when accumulated. The five-fold cross validation rate for all sets of features increased with the number of EEGs accumulated. The accuracy of a feature combination for P1, N1 and FD is significant when the waveform is measured twice. This result suggests that the size of objects being viewed can be predicted using a set of features which can be extracted from a single channel EEG waveform recorded at an occipital electrode.

Contextual understanding, which consists of memory, reasoning and recall, is a key process of human-computer interactions and interfaces. To determine the possibility of predicting the recall accuracy of reading and memorizing tasks using features of eye movements for a targeted text, a contextual understanding task experiment was conducted. The relationship between eye movement during memorization and recall performance was hypothesized. Viewer's eye-movements during the reading of definition statements were observed, and twelve sector features of these eye-movements across a two-dimensional space were measured. The prediction procedure was developed using Support Vector Regressions with Gaussian kernel. The correlational relationship between predicted and experimental recall accuracies was significant; therefore the hypothesis was supported. Also, it was found that prediction performance depended on a combination of features of eye movements.

To estimate viewer's contextual understanding, features of their eye-movements while viewing question statements in response to definition statements, and features of correct and incorrect responses were extracted and compared. Twelve directional features of eye-movements across a two-dimensional space were created, and these features were compared between correct and incorrect responses. The procedure of estimating the response was developed with Support Vector Machines, using these features. The estimation performance and accuracy were assessed across combinations of features. The number of definition statements, which needed to be memorized to answer the question statements during the experiment, affected the estimation accuracy. These results provide evidence that features of eye-movements during reading statements can be used as an index of contextual understanding.

Eye-movements can be an index of usability, though there is no significant relationship between a subjective usability score and indices of eye-movements, however [Nakayama and Katsukura 2007]. A possible reasons for this is the shortness of the observational duration. As another approach, power spectrum density (PSD) and cross spectrum density (CSD) of eye-movements can be used as an index of mental workload [Nakayama and Shimizu 2004]. This paper addresses the possibility of assessing usability using frequency analysis of eye-movements.

To determine the performance of estimating the degree of "strength of belief" (SOB) of responses using eye-movements, the features of eye-movements were extracted while subjects answered questions and reviewed their own responses to multiple choice tasks. The estimation was conducted using Support Vector Machines (SVM) with features of eye-movements between two successive fixations. In the results, the overall estimation performances were significant when the prediction was calculated using certain combinations of features for answering and reviewing sessions respectively. In comparing feature combinations of eye-movements for significant estimations between answering and reviewing sessions, significant combinations components of features between the two sessions did not coincide.

To examine the feasibility of estimating the degree of strength of belief (SOB) of responses using eye movements, the scan paths of eye movements were analyzed while subjects reviewed their own responses to multiple choice tasks. All fixation points of eye movements were classified into visual areas, or cells, which corresponded with the positions of answers. Two estimation procedures are proposed using eye-movement data. The first one is identifying SOB using scan-path transitions. By comparing subject's reports of high and low SOB and eye-movement estimations, a significant correct rate of discrimination of SOB was observed. When the threshold of discrimination was controlled, a high rate of correct responses was obtained if it was set at a low level. The second procedure is conducting SOB discrimination using support vector machines (SVM) trained with features of fixations. Subject's gazing features were analyzed while they reviewed their own responses. A discrimination model for SOB was trained with several combinations of features to see whether performance of a significant level could be obtained. As a result, a trained model with 3 features (which consist of interval time, vertical difference, and length between fixations) can provide significant discrimination performance for SOB. These results provide evidence that strength of belief can be estimated using eye movements

A lot of multimedia learning materials and software have been developed as learning tools. Their usability for learning has not often been discussed, however. The issue of system usabilty is often considered regarding various other processes, but learning materials should also be evaluated. The indices of oculo-motors can be used for usability tests [Nakayama et al. 2002], so they may also indicate the usability of learning materials. To examine the feasibility of using oculo-motors to evaluate multimedia learning materials, an exercise was tested under experimental conditions.

This paper describes the influence of eye blinks on frequency analysis and power spectrum difference for task-evoked pupillography and eye-movement during an experiment which consisted of ocular following tasks and oral calculation tasks with three levels of task difficulty: control, 1×1,and 1×2 digit oral calculation.The compensation model for temporal pupil size based on MLP (multi layer perceptron) was trained to detect a blink and to estimate pupil size by using blinkless pupillary change and artificial blink patterns. The PSD (power spectrum density) measurements from the estimated pupillography during oral calculation tasks show significant differences, and the PSD increased with task difficulty in the area of 0.1 - 0.5Hz and 1.6 - 3.5Hz, as did the average pupil size.The eye-movement during blinks was corrected manually, to remove irregular eye-movements such as saccades. The CSD (cross spectrum density) was achieved from horizontal and vertical eye-movement coordinates. Significant differences in CSDs among experimental conditions were examined in the area of 0.6 - 1.5 Hz. These differences suggest that the task difficulty affects the relationship between horizontal and vertical eye-movement coordinates in the frequency domain.

The oculo-motor re ects the viewer s ability to process visual information. This paper examines whether the oculo-motor was affected by two factors: rstly task dif culty and secondly eye-movement frequency. In this paper, oculo-motor indices were de ned as measurements of pupil size, blink and eye-movement. For the purpose of this study, two experiments were designed based on previous subsequential ocular tasks were subjects were required to solve a series of mathematical problems and to orally report their calculations.The results of this experiment found that pupil size and blink rate increased in response to task dif culty in the oral calculation group. In contrast however both the saccade occurrence rate and saccade length were found to decrease with the increased dif culty of the task. The results suggests that oculo-motor indices respond to task dif culty. Secondly, eye-movement frequencies were elicited by the switching frequency of a visual target. Pupil size and the saccade time were found to increase with the frequency however, blink and gazing time were found to decrease in response to the frequency. There was a negative correlation between blinking and gazing time. Additionally, the correlation between blinking and saccade time appeared in the higher frequencies.These results indicate the oculo-motor indices are affected by both task dif culty and eye-movement frequency. Furthermore, eye-movement frequency appears to play a different role than that of task dif culty.

Eye movement reflects a viewer's visual information process. This study examines whether eye-movement responds to the viewer's cognitive load. It is already known that pupil size and blink can use as an indicator of mental workload. Saccades are rapid eye movements to turn a fovea to a focusing target. For this process, saccade was extracted to observe viewing process.