No description available.

No description available.

No description available.

[pt] Esta dissertação insere-se na área da Psicolinguística e teve como objetivo analisar como alunos de Ensino Médio fazem a leitura das provas de espanhol do ENEM, identificando os custos associados à compreensão dos textos. A pesquisa envolveu (i) a análise linguística das provas de 2010 a 2017 e das habilidades de leitura, conforme as matrizes de referência do ENEM e do PISA e (ii) o exame das estratégias de leitura empregadas na realização da prova de 2017. Foi comparado o desempenho de 44 alunos com tempos distintos de exposição à língua espanhola (pelo menos 50 horas/aula vs. menos de 17 horas/aula). Os dados foram obtidos por meio do programa de gravação de tela Active Presenter e do rastreador ocular Tobii Pro-X3 120 Hz. Os principais aportes teóricos foram pesquisas no campo da leitura e da Psicolinguística do Bilinguismo. Os resultados indicam diferenças de complexidade entre as provas quanto às estruturas e o vocabulário, foco em habilidades de integração e interpretação de textos, e uso e função social das estruturas. A taxa de acertos foi inferior a 50 por cento, com desempenho superior para o grupo com maior exposição à língua na atividade no rastreador ocular. A média de tempo para realização da prova foi compatível com o previsto para o ENEM. Não houve diferença significativa entre os grupos quanto ao número de releituras e poucos iniciaram a leitura pelo enunciado. Em relação aos parâmetros oculares, também não há evidências de desempenho distinto. / [en] This thesis is developed under the scope of Psycholinguistics and is aimed at analyzing how high school students read the ENEM Spanish tests, in addition to identifying cost associated with reading comprehension. The research involved (i) the linguistic analysis of the 2010-2017 tests and reading skills in accordance to the ENEM and PISA reference matrices and (ii) the examination of the reading strategies employed in the 2017 test. The performance of 44 students with different language exposure time to Spanish (at least 50 hours of in-class exposure vs. less than 17 in-class exposure) was compared. The data were obtained through the Active Presenter screen recording program and Tobii Pro-X3 120 Hz eye tracker. The main theoretical frameworks were research in the field of reading and Psycholinguistics of Bilingualism. The results show differences in complexity between the tests regarding structure and vocabulary, focus on text integration and reading skills, and use and social function of the structures. The hit rate was less than 50 percent, and the group with the highest language exposure outperformed in the eye tracker activity. The average time to perform the test was compatible with the one predicted for ENEM. There was no significant difference between the groups in the number of rereading times, and few started reading by the statement. Regarding eye parameters, there is also no evidence of distinct performance.

If researchers are able to derive usability simply by analysing gaze data it provides a quick, objective and potentially automatic way of measuring the usability of an interface. In order to do that it is essential to know which traits of the gaze data that have an impact on usability. This paper investigates these traits by analysing different eye tracking metrics in the data. The goal is to see which of these metrics have a general correlation with usability. Previous research provides a clue about which metrics are useful when analysing usability. However, much of the research is based on subjective analysis or lacks in general applicability. This research provides an objective analysis that is independent of characteristics of the interface. A user study is done on 20 participants. They complete tasks on nine different ecommerce websites while their eye movements are recorded. Correlation is measured between usability and eye tracking metrics in order to investigate which metrics that are sensitive to changes in usability. The results show that fixational backtracks and number of fixations have the strongest correlation with usability. Previous research did suggest that both of these eye tracking metrics have an impact on usability. / Om det är möjligt för forskare att härleda användbarhet endast genom att analysera ögondata så tillhandahålls ett snabbt, objektivt och potentiellt automatiserat sätt att mäta användbarheten hos ett gränssnitt. För att göra detta är det avgörande att veta vilka karakteristiska drag i ögondatan som påverkar användbarheten. Denna rapport utforskar dessa drag genom att analysera olika mått i ögondatan. Målet är att se vilka av dessa mått som uppvisar en generell korrelation med användbarhet. Tidigare forskning förser oss med ledtrådar om vilka mått som är lämpliga att analysera när användbarhet ska mätas. Dock så är mycket av denna forskning baserad på subjektiv analys eller saknar generell tillämpbarhet. Denna rapport tillhandahåller en objektiv analys som är oberoende av karakteristiken hos gränssnittet. En användarstudie utförs på 20 deltagare. De utför uppgifter på nio olika webbsidor för e-handel medan deras ögonrörelser spelas in. Korrelation mäts mellan användbarhet och mått i ögondatan för att undersöka vilka mått som är känsliga för förändringar i användbarhet. Resultatet visar att tillbakabildande sackader och antal fixeringar har starkast korrelation med användbarhet. Tidigare forskning visade att båda dessa mått påverkades av användbarheten.

Research has demonstrated that everyday functional abilities are compromised in mild cognitive impairment (MCI), a transitional stage between normal cognitive aging and dementia, as well as in healthy aging. These functional changes have been shown to be strong predictors of future decline, highlighting their importance. However, early changes in everyday functioning remain poorly characterized, largely due to a scarcity of sensitive measures capable of detecting subtle disruption. Recent research suggests that eye-tracking methodology may be effective in addressing this gap. Fifty-two participants (27 younger adults and 25 non-demented older adults) completed a novel eye-tracking task involving passive viewing of a naturalistic scene and verbalization of a task goal (e.g., make coffee, pack a lunch). Participants also completed a performance-based measure of everyday action that required them to enact the same tasks (e.g., coffee, lunch) that were included in the eye-tracking paradigm, self-report measures of functional ability, and neuropsychological measures. Mixed ANOVAs were conducted to examine group (young, old) and condition (passive viewing, verbalization)/task (simple, complex) effects on eye-tracking and everyday action performance. Independent samples t-tests/Mann-Whitney U tests were conducted to examine group differences in eye-tracking and everyday action performance. Correlation analyses across all measures were conducted to evaluate the potential mechanisms of eye-tracking and everyday action results. Results showed no significant group differences in the primary eye-tracking variables, but both groups made a lower proportion of fixations to distractor (i.e., non-target) objects during task verbalization compared to passive scene viewing. Older adults made more inefficient actions during performance-based everyday task completion, particularly when task demands were high. Eye tracking and everyday action variables were related to different measures of self-reported functional ability. Finally, eye-tracking variables were primarily related to neuropsychological measures of executive functions/working memory, whereas everyday action performance was most strongly related to measures of verbal learning and memory. These findings suggest that age-related functional changes at the level of eye movements may occur after changes in behavioral performance of everyday tasks. Importantly, performance-based assessment of everyday action appears sensitive to age-related decline. Additionally, naturalistic eye movements and everyday task performance may reflect distinct components of self-reported functioning and may be driven by distinct cognitive processes. Future research with refined naturalistic eye-tracking tasks and samples with a wider range of impairment is necessary to further explore these findings and improve characterization and detection of risk for dementia. / Psychology

[pt] O rastreamento ocular (eye-tracking) possibilita rastrear a posição e a direção do olhar de uma pessoa sobre algum tipo de estímulo (e.g., imagens ou vídeos). O uso desta tecnologia permite identificar eventos inerentes à visão humana, que contém informações implícitas capazes de revelar aspectos importantes sobre o comportamento de um indivíduo durante uma determinada tarefa. Porém, identificar essas informações requer um conjunto de habilidades para interpretar os dados de rastreamento ocular e relacioná-los com conhecimentos de domínios específicos. Nesse contexto, pode-se fazer grande proveito de sistemas inteligentes para agregar os conhecimentos e experiências de especialistas junto às respostas do dispositivo de rastreamento ocular. Dessa forma, o objetivo principal deste trabalho é propor uma metodologia para criar sistemas baseados em eye-tracking, para enriquecer o processo de avaliação de um indivíduo frente a uma determinada tarefa, resultando em um modelo para representar o conhecimento dos especialistas sobre aspectos subjetivos, visando automatizar esse processo avaliativo. Portanto, o presente trabalho toma como caso de uso a avaliação da relação entre comportamento visual e eficácia de indivíduos na resolução de testes inspirados em Matrizes Progressivas de Raven. Esses testes são comumente usados na psicologia para medir inteligência e a capacidade de raciocínio abstrato a partir da visualização de imagens. Optamos por utilizar uma abordagem baseada em regras fuzzy, por permitir armazenar conhecimento de forma mais transparente e legível aos usuários finais. As regras do modelo foram desenvolvidas e validadas com o auxílio de um especialista da área da psicologia. O sistema foi testado com dados extraídos de um grupo de usuários e apresentou resultados promissores. Os achados e modelos obtidos nessa pesquisa poderão ser utilizados como alicerce para o desenvolvimento de sistemas mais robustos. / [en] Eye-tracking makes it possible to track the position and direction of a person s gaze on some stimulus (e.g., images or videos). This technology allows us to identify events inherent to human vision, containing implicit information capable of revealing essential aspects of one s behavior during a given task. However, identifying these pieces of information is a complex task that requires a set of skills to interpret the eye-tracking data and relate it to domain-specific knowledge. In this context, one can use intelligent systems to couple the knowledge and experience of specialists with the responses from the eye-tracking device. Thus, the main objective of this work is to propose a methodology to create eye-tracking-based systems to improve the assessment of subjects during specific tasks, resulting in a model that can represent the specialist s knowledge over subjective aspects to automate this process. Therefore, the present work s use case is the evaluation of the relationship between visual behavior and efficiency in solving tests inspired by Raven s Progressive Matrices. Those tests are commonly used in psychology to measure intelligence and abstract reasoning through image visualization. We chose an approach based on fuzzy rules, as it allows us to represent knowledge in a more readable way to end-users. The model s rules were developed and validated alongside a specialist in psychology. The system was tested with data extracted from users and showed promising results. The findings and models obtained in this research may be used as a foundation for the development of more robust systems.

Yes / Recent research has shown that adults with autism spectrum disorders (ASD) have difficulty interpreting others' emotional responses, in order to work out what actually happened to them. It is unclear what underlies this difficulty; important cues may be missed from fast paced dynamic stimuli, or spontaneous emotional responses may be too complex for those with ASD to successfully recognise. To explore these possibilities, 17 adolescents and adults with ASD and 17 neurotypical controls viewed 21 videos and pictures of peoples' emotional responses to gifts (chocolate, a handmade novelty or Monopoly money), then inferred what gift the person received and the emotion expressed by the person while eye movements were measured. Participants with ASD were significantly more accurate at distinguishing who received a chocolate or homemade gift from static (compared to dynamic) stimuli, but significantly less accurate when inferring who received Monopoly money from static (compared to dynamic) stimuli. Both groups made similar emotion attributions to each gift in both conditions (positive for chocolate, feigned positive for homemade and confused for Monopoly money). Participants with ASD only made marginally significantly fewer fixations to the eyes of the face, and face of the person than typical controls in both conditions. Results suggest adolescents and adults with ASD can distinguish subtle emotion cues for certain emotions (genuine from feigned positive) when given sufficient processing time, however, dynamic cues are informative for recognising emotion blends (e.g. smiling in confusion). This indicates difficulties processing complex emotion responses in ASD.

This thesis delves into eye tracking, a technique for estimating an individual’s point of gaze and understanding human interactions with the environment. A blossoming area within eye tracking is appearance-based eye tracking, which leverages deep neural networks to predict gaze positions from eye images. Despite its efficacy, the decision-making processes inherent in deep neural networks remain as ’black boxes’ to humans. This lack of transparency challenges the trust human professionals place in the predictions of appearance-based eye tracking models. To address this issue, explainable AI is introduced, aiming to unveil the decision-making processes of deep neural networks and render them comprehensible to humans. This thesis employs various post-hoc explainable AI methods, including saliency maps, gradient-weighted class activation mapping, and guided backpropagation, to generate heat maps of eye images. These heat maps reveal discriminative areas pivotal to the model’s gaze predictions, and glints emerge as of paramount importance. To explore additional features in gaze estimation, a glint-free dataset is derived from the original glint-preserved dataset by employing blob detection to eliminate glints from each eye image. A corresponding glint-free model is trained on this dataset. Cross-evaluations of the two datasets and models discover that the glint-free model extracts complementary features (pupil, iris, and eyelids) to the glint-preserved model (glints), with both feature sets exhibiting comparable intensities in heat maps. To make use of all the features, an augmented dataset is constructed, incorporating selected samples from both glint-preserved and glint-free datasets. An augmented model is then trained on this dataset, demonstrating a superior performance compared to both glint-preserved and glint-free models. The augmented model excels due to its training process on a diverse set of glint-preserved and glint-free samples: it prioritizes glints when of high quality, and adjusts the focus to the entire eye in the presence of poor glint quality. This exploration enhances the understanding of the critical factors influencing gaze prediction and contributes to the development of more robust and interpretable appearance-based eye tracking models. / Denna avhandling handlar om ögonspårning, en teknik för att uppskatta en individs blickpunkt och förstå människors interaktioner med miljön. Ett viktigt område inom ögonspårning är bildbaserad ögonspårning, som utnyttjar djupa neuronnät för att förutsäga blickpositioner från ögonbilder. Trots dess effektivitet förblir beslutsprocesserna i djupa neuronnät som ”svarta lådor” för människor. Denna brist på transparens utmanar det förtroende som yrkesverksamma sätter i förutsägelserna från bildbaserade ögonspårningsmodeller. För att ta itu med detta problem introduceras förklarbar AI, med målet att avslöja beslutsprocesserna hos djupa neuronnät och göra dem begripliga för människor. Denna avhandling använder olika efterhandsmetoder för förklarbar AI, inklusive saliency maps, gradient-weighted class activation mapping och guidad backpropagation, för att generera värmekartor av ögonbilder. Dessa värmekartor avslöjar områden som är avgörande för modellens blickförutsägelser, och ögonblänk framstår som av yttersta vikt. För att utforska ytterligare funktioner i blickuppskattning, härleds ett dataset utan ögonblänk från det ursprungliga datasetet genom att använda blobdetektering för att eliminera blänk från varje ögonbild. En motsvarande blänkfri modell tränas på detta dataset. Korsutvärderingar av de två datamängderna och modellerna visar att den blänkfria modellen tar fasta på kompletterande särdrag (pupill, iris och ögonlock) jämfört med den blänkbevarade modellen, men båda modellerna visar jämförbara intensiteter i värmekartorna. För att utnyttja all information konstrueras ett förstärkt dataset, som inkorporerar utvalda exempel från både blänkbevarade och blänkfria dataset. En förstärkt modell tränas sedan på detta dataset, och visar överlägsen prestanda jämfört med de båda andra modellerna. Den förstärkta modellen utmärker sig på grund av sin träning på en mångfaldig uppsättning av exempel med och utan blänk: den prioriterar blänk när de är av hög kvalitet och justerar fokuset till hela ögat vid dålig blänkkvalitet. Detta arbete förbättrar förståelsen för de kritiska faktorerna som påverkar blickförutsägelse och bidrar till utvecklingen av mer robusta och tolkningsbara modeller för bildbaserad ögonspårning.

Mental health is essential for overall well-being, focusing emotional, psychological, and social aspects. Assessing and managing mental health requires understanding mental state parameters, including cognitive load, cognitive impairment, and emotional state. Advanced technologies like eye tracking provide valuable insights into these parameters, transformed mental health evaluation and enabled more targeted interventions and better outcomes. Thesis focused towards developing intelligent system to monitor mental health, focusing on cognitive load, cognitive impairment, and emotional state. The research has three main objectives, including creating four eye-tracking-based unimodal datasets and a multimodal dataset to address the lack of publicly available mental health assessment datasets. Each dataset is designed to study cognitive load, cognitive impairment, and emotional state classification using varied stimuli. In addition to dataset creation, the thesis excels in feature extraction, introducing novel features to detect mental state parameters and enhancing assessment precision. High-level features such as error rate, scanpath comparison score, and inattentional blindness are incorporated, contributing to find cognitive impairment scores. Five models are developed to detect mental states by separately monitoring the mental state parameters, cognitive load, cognitive impairment, and emotional state. The models employ statistical analysis, machine learning algorithms, fuzzy inference systems, and deep learning techniques to provide detailed insights into an individual's mental state. The first two models, Eye-Tracking Cognitive Load models (ECL-1 and ECL-2) focus on cognitive load assessment during mathematical assessments and Trail Making Test tasks. ECL-1 model utilizes statistical analysis to understand the correlation between eye tracking features like pupil diameter and blink frequency with the cognitive load while performing mathematical assessments. With the identification of relevant features while performing Trail Making Test (TMT), the ECL-2 model effectively classifies low and high cognitive load states with a notable 94% accuracy, utilizing eye-tracking data and machine learning algorithms. The third model, the ETMT (Eye tracking based Trail Making Test) model, uses a fuzzy inference system and adaptive neuro-fuzzy inference system to detect mental states associated with cognitive impairment. It provides detailed scores in visual search speed and focused attention, important for understanding the exact cognitive deficits of a patient. This greatly aids in understanding the cognitive states of an individual and addresses deficits in executive functioning, memory, motor function, attentional disengagement, neuropsychological function, processing speed, and visual attention. The fourth model, PredictEYE, utilizes a deep learning time-series univariate regression model based on Long Short-Term Memory (LSTM) to predict future sequences of each feature. Machine learning-based Random Forest algorithm is applied on the predicted features for mental state prediction and identifying the mental state as calm or stressful based on a person's emotional state. The personalized time series methodology makes use of the power of time series analysis, identifying patterns and changes in data over time to enable more precise and individualized mental health assessments and monitoring. Notably, PredictEYE outperforms ARIMA with an accuracy of 86.4%. The fifth model introduced in this study is based on a multimodal dataset, incorporating physiological measures such as ECG, GSR, PPG, and respiratory signals, along with eye tracking data. Two separate models, one based on eye tracking data and the other based on all other physiological measures developed for understanding the emotional state of a person. These models demonstrate comparable performance, with notable proficiency in binary classification based on arousal and valence. Particularly, the Binary-Valence model achieves slightly higher accuracy when utilizing eye tracking data, while other physiological measures exhibit stronger classification performance for the Binary-Arousal model. The thesis makes substantial progress in mental health monitoring by providing accurate, non-intrusive evaluations of an individual's mental state. It emphasizes mental state parameters such as cognitive load, impairment, and emotional state, with AI-based methods incorporated to improve the precision in detection of mental state.