There are different aims in this thesis. The primary aim is to investigate visual perception in Chinese orthography, from its fundamentally distinct unit, characters, to sentence reading. The first aim of the thesis is to investigate how a single Chinese character is processed in many ways. We have looked into an effect called orthographic satiation/decomposition (Cheng & Wu, 1994). It refers to the feeling of uncertainty about the composition of some characters when staring at a character for too long. Lee (2007) extended Cheng and Wu’s study (1996), and the results have showed that orthographic satiation occurred faster in females than males. We replicated Lee’s study (Experiment 1) (Chapter 3), and have found that: (1) there was no significant difference between male and female and (2) a radical that can stand alone as a single character makes characters in which it occurs resistant to satiation. Following orthographic satiation, in Chapter 4, we explored the preference for eye/hemisphere visual pathways in Chinese characters (Experiment 2 & 3) and words (Experiment 4). In English, researchers have reported a contralateral preference when four-letter words were presented very quickly using a haploscope (Obreg´on & Shillcock, 2012) . It raises the question of whether presenting Chinese characters and words will show similar results considering the complexity and the special characteristics of Chinese orthography. We presented Chinese characters and words to participants using a haploscope. Our results showed that: (1) the contralateral visual pathway was preferred in perceiving right-left structured Chinese characters and two-character words, (2) when a semantic radical is projected to the LH, participants are able to recognise the semantic component better, (3) neighbourhood size (NS) (Tsai, Lee, Lin, Tzeng & Hung, 2006) affects how participants recognise words, and(4) males do better than females recognising characters but not words. After investigating the recognition of Chinese characters and words, we analysed the eye-movements in Chinese and English reading corpora. The processes of reading are intuitively thought to be more complex than perceiving a single character or words. The last studies in the thesis focused on the reading behaviours in Chinese and English. The eye movement differences and similarities between reading Chinese and English were investigated. In Chapter 5, we showed that reading Chinese elicits more divergence of the eyes within a fixation, compared with reading English. We interpreted these data in terms of recent demonstrations that apparent size causes increases in visual sensitivity (Arnold & Schindel, 2010) and engages more cortical resource in V1 (Kersten & Murray, 2010). Our analyses were based on movement within exactly temporally synchronized binocular fixations in the reading of Chinese and English 5000-word multi-line texts, using monocular calibration, with EyeLink-2 technology. When faced with visually complex orthography, the oculomotor system ‘tricks’ the rest of the visual system into ‘zooming in’ on the text. We consolidated the relevant theorizing into the ‘Divergence Affects Reading’ (DOLLAR) Theory. In Chapter 6, we reported that (1) vertical movements within a fixation tend to be smaller than horizontal ones, and (2) vertical movements within a fixation tend to be upwards. We speculated that it is appropriate for the earlier part of the fixation to be associated with visual recognition and for the later part of the fixation to be associated with executive action. The tendency to move upwards also suggested that in real-world reading, the upper part of words/characters were informative. In the last chapter analysing the reading corpus (Chapter 7), we reported corrective saccades after return sweeps. We found that in English, there were more corrective saccades after return sweeps than in Chinese. We interpreted these data in terms of spatial coding (Kennedy & Murray, 1987). In terms of Chinese and English differences, the stimuli that were used in our corpus show that the length for each line was different in English. The length for each line in Chinese was less different. Though the first character of each line was at the same place for two languages, it was more difficult for English subjects to locate the correct place after return sweeps because the length of return sweeps was different. In short, this thesis investigated visual perception in Chinese orthography, in terms of characters, words, and real-world reading. Moreover, we compared the differences and similarities between languages (English and Chinese). Despite the fact that the orthographies of English and Chinese are very different, we still found similar effects (e.g., contralateral preference) between them. This thesis thus has contributed to a better understanding of the differences and similarities between English and Chinese in terms of the orthographies.

Orientador: Daniela Micheline dos Santos / Co-orientador: Marcelo Coelho Goiato / Banca: Eduardo Piza Pellizzer / Banca: Aldiéris Alves Pesqueira / Resumo: A perda do bulbo ocular compromete não só a estética, mas também a tonicidade muscular da região facial do paciente, uma vez que com a ausência do globo ocular os músculos orbiculares dos olhos podem sofrer atrofia. Desse modo, o objetivo do presente estudo foi verificar a atividade elétrica dos músculos orbiculares, antes e após a instalação de próteses oculares em pacientes que foram submetidos à enucleação unilateral do bulbo ocular. Foram selecionados, por meio de anamnese e exame clínico, 12 pacientes voluntários com indicação de prótese. O sinal eletromiográfico foi realizado com o auxílio do eletromiógrafo, em quatro situações clínicas: Repouso (R), Abertura e Fechamento Normal das Pálpebras (AFN), Abertura e Fechamento Rápido das Pálpebras (AFR) e Apertamento (A). Esses registros foram realizados antes da instalação da prótese ocular, e após 7, 30 e 60 dias da instalação e uso da mesma. Os mesmos ensaios foram realizados no músculo orbicular do olho sadio do paciente, resultados que serviram como controle do estudo. Os dados obtidos foram submetidos à análise estatística pelo programa SPSS (p<0.05) e o t-teste foi utilizado para comparar os músculos superior e inferior por período de tratamento (inicial, 7, 14, 30 e 60 dias), para as quatro condições clínicas. Nas quatro condições clínicas avaliadas foi verificado diferença estatisticamente significativa em relação ao período inicial e após 7 dias da instalação da prótese. O fascículo superior do músculo orbicular do olho apresentou maiores valores de atividade elétrica em relação ao fascículo inferior em todas as situações clínicas avaliadas. Os menores valores de atividade elétrica foram observados durante o período inicial para a condição de repouso (OS 8.418 / OI 5.933) e os maiores após 60 dias na condição... / Abstract: The eye loss besides affecting patient's aesthetics, it compromises the muscle tone of the facial region owing to the atrophy of orbicular muscles. Thus, although the use of ocular prosthesis does not return patient's vision, it fills the anophtalmic cavity restoring the cosmetic and muscle tone. The aim of this present study was to evaluate the electrical activity of orbicular muscles before and after ocular prosthesis insertion of patients who underwent unilateral enucleation of eyeball. The electrical activity of the orbicular muscles was assessed through the Myosystem BR1 electromyograph in four clinical situations: (1) rest, (2) normal opening and closing of the eyelid, (3) fast opening and closing of the eyelids, and (4) clenching. The electrodes were placed in the fascicles of upper (UO) and lower (LO) orbicular muscles. Electromyographic examinations were performed before and after 7, 14, 30 and 60 days of prosthesis insertion. T-test (p<.05) was used to compare the upper and lower orbicular muscles for each period of evaluation in all clinical conditions. A total of 12 patients of both genders were treated and they aged from 42 to 80 years. Several factors were the cause of anophthalmia and the trauma during job accident was the main reason. A statistical significant difference in the electromyographic data was observed for all four clinical conditions when comparing the baseline with the 7-day prosthesis insertion periods. The UO exhibited higher values of electrical activity than LO for all clinical situations. The lowest electrical activity was noted for the baseline period during the rest condition (UO 8.418 /LO 5.933), while the greatest one after 60 days of prosthesis insertion during clenching (UO 131.504 / LO 117.123). After ocular prosthesis insertion, a significant increase in the electrical activity values of the orbicular muscles was observed. / Mestre

Made available in DSpace on 2014-06-11T19:23:42Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-01-20Bitstream added on 2014-06-13T18:19:51Z : No. of bitstreams: 1 toni_mc_me_jabo.pdf: 258996 bytes, checksum: 041759d510733c85331d9b8673133b75 (MD5) / Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Com o presente estudo pretendeu-se conhecer os valores médios para as estruturas intraoculares e compará-los entre cães com diferentes conformações cranianas por meio da ultrassonografia ocular com transdutor de 20 MHz. Foram selecionados 30 cães, adultos, peso corpóreo até 10 kg, machos ou fêmeas, hígidos, atendidos no Hospital Veterinário Governador Laudo Natel - Faculdade de Ciências Agrárias e Veterinárias – UNESP – Câmpus de Jaboticabal. Os cães foram distribuidos em três grupos: G1 – 10 braquicefálicos, G2 - 10 mesocefálicos e G3 - 10 dolicocefálicos. Na análise estatística realizada pelo teste t Student com nível de significância de 5%, não houve diferença entre os olhos direito e esquerdo dos cães do mesmo grupo (p > 0,05). Valores médios e desvios padrão foram obtidos para profundidade da câmara anterior (D1), espessura da lente (D2), profundidade da câmara vítrea (D3) e comprimento axial do bulbo do olho (D4) nos três grupos. G1: D1 2,92mm (± 0,37), D2 7,03mm (± 0,61), D3 9,48mm (± 0,74) e D4 19,43mm (± 0,60). G2: D1 3,08mm (± 0,67), D2 7,02mm (± 0,87), D3 8,76mm (± 1,23) e D4 18,86mm (± 1,45). G3: D1 3,07mm (± 0,65), D2 7mm (± 0,81), D3 9,75mm (± 0,98) e D4 19,82mm (± 0,95). Para comparação dos valores médios de D1, D2, D3 e D4 entre os grupos, foi utilizado o teste de Tukey, com nível de significância de 5%. Na análise de D3, foi observada diferença (p < 0,05) na comparação entre G2 e G3. Na análise de D4, foi observada diferença (p < 0,05) na comparação entre G2 e G3 / The present study aimed to know the average values of the intraocular structures and compare among different skulls’ conformations in dogs using ocular ultrasonography with 20 MHz probe. They were selected 30 healthy adult dogs, body weight 10 kg, male or female, assisted at Governador Laudo Natel Veterinary Teaching Hospital - Faculdade de Ciências Agrárias e Veterinárias – UNESP – Campus of Jaboticabal. Animals were divided into three groups: G1 - 10 brachycephalic dogs, G2 – 10 mesocephalic dogs and G3 - 10 dolicocephalic dogs. The Student t test at 5% significance level was used to compare difference between right and left eyes of the dogs in the same group and no difference was found (p > 0.05). Mean and standard deviations were obtained for anterior chamber depth (D1), lens thickness (D2), vitreous chamber depth (D3) and axial length of the eye globe (D4) in the three groups. G1: D1 2.92mm (± 0.37), D2 7.03mm (± 0.61), D3 9.48mm (± 0.74) and D4 19.43mm (± 0.60). G2: D1 3.08mm (± 0.67), D2 7.02mm (± 0.87), D3 8.76mm (± 1.23) and D4 18.86mm (± 1.45). G3: D1 3.07mm (± 0.65), D2 7mm (± 0.81), D3 9.75mm (± 0.98) and D4 19.82mm (± 0.95). The Tukey test at 5% significance level was used to compare the average of D1, D2, D3 and D4 among groups. Differences between G2 and G3 were found (p <0.05) when comparing D3 and D4

Made available in DSpace on 2018-08-02T00:00:40Z (GMT). No. of bitstreams: 1 tese_10810_Hamilton Rivera Flor20171019-95619.pdf: 4725252 bytes, checksum: 16042ed4abfc5b07268db9f41baa2a83 (MD5) Previous issue date: 2017-03-17 / Automated reading and analysis of human emotion has the potential to be a powerful tool to develop a wide variety of applications, such as human-computer interaction systems, but, at the same time, this is a very difficult issue because the human communication is very complex. Humans employ multiple sensory systems in emotion recognition. At the same way, an emotionally intelligent machine requires multiples sensors to be able to create an affective interaction with users. Thus, this Master thesis proposes the development of a multisensorial system for automatic emotion recognition. The multisensorial system is composed of three sensors, which allowed exploring different emotional aspects, as the eye tracking, using the IR-PCR technique, helped conducting studies about visual social attention; the Kinect, in conjunction with the FACS-AU system technique, allowed developing a tool for facial expression recognition; and the thermal camera, using the FT-RoI technique, was employed for detecting facial thermal variation. When performing the multisensorial integration of the system, it was possible to obtain a more complete and varied analysis of the emotional aspects, allowing evaluate focal attention, valence comprehension, valence expressions, facial expression, valence recognition and arousal recognition. Experiments were performed with sixteen healthy adult volunteers and 105 healthy children volunteers and the results were the developed system, which was able to detect eye gaze, recognize facial expression and estimate the valence and arousal for emotion recognition, This system also presents the potential to analyzed emotions of people by facial features using contactless sensors in semi-structured environments, such as clinics, laboratories, or classrooms. This system also presents the potential to become an embedded tool in robots to endow these machines with an emotional intelligence for a more natural interaction with humans. Keywords: emotion recognition, eye tracking, facial expression, facial thermal variation, integration multisensorial

Neste trabalho foi desenvolvido um novo instrumento para monitoramento computadorizado da curvatura da região central anterior da córnea humana durante cirurgias refrativas. Através da projeção de um disco de Plácido na córnea, imagens dos reflexos são digitalizadas e processadas. Algoritmos baseados em técnicas de visão computacional e óptica geométrica determinam a curvatura da região central (-7 mm em diâmetro), com alta precisão e desempenho. Mapas coloridos com códigos de cor em dioptrias (proporcionais ao inverso do raio de curvatura) são gerados para auxiliar o oftalmologista durante a cirurgia. / In this work we have developed a new instrument for computerized monitoring of corneal central curvature during surgery. By projecting Placido Rings on the cornea, images of the reflections are digitized and processed. Algorithms based on computational vision and optical geometry determine the central curvature (-7 mm in diameter), with high performance and precision. Color coded maps in diopters (proportional to the inverse of the radius of curvature) are generated to aid the ophthalmologist during surgery.

The aim of this thesis is to predict targets in a Multiple Object Tracking (MOT) task, in which subjects track multiple moving objects. We processed and analyzed data containing object and gaze position information from 1148 MOT trials completed by 20 subjects. We extracted multiple features from the raw data and designed a machine learning approach for the prediction of targets using neural networks and hidden Markov models. We assessed the performance of the models and features. The results of our experiments show that it is possible to train a machine learning model to predict targets with very high accuracy. 1

As online fashion retailing has grown immensely over the last fifteen years, retailers’ website designs have had to evolve to meet customers’ increasing demands, needs and expectations in order to attract and retain them. The purpose of this thesis is to provide a greater understanding of how different ages of consumers respond to retailers’ holistic website designs through the use of eye tracking technology. Insights gained from this research will enable retailers to enhance their website design in accordance with their target market’s behaviour, habits and expectations. Eye tracking is a key tool for analysing and understanding human-computer interaction and can be used alongside qualitative in-depth interviews to provide a detailed understanding of users’ responses to the website design. This study uses the Stimulus–Organism–Response (S-O-R) paradigm to investigate how the fashion retailer’s holistic website design influences users’ visual, cognitive and behavioural responses, which leads to their approach or avoidance behaviour. The study also examines whether the shopper’s motivation or age will have an influence on their responses to the website design. It is important to analyse the differences in behaviour between age groups because more mature consumers are now shopping online, and academic research on this topic is limited. A sample of 50 participants, spanning ages 20 to 70 and consisting of regular users of the website, participated in the study. The study was conducted on the fashion retailer’s live website and results were recorded in real time, making the findings more valid. The study found that users were very focused on looking for products when shopping online and that the product listings page is arguably the most important page on the website as users visit it the most and spend the most time on it. Furthermore, the study found that navigation and customisation design stimuli were the most important to users and that they did not to pay much attention to the lower sections of the web pages. The study also found that users’ responses to the website can change according to their motivation as users spent less time on the website if they were goal-directed shopping and had more focused viewing patterns than if they were browsing. Moreover, the study also found that different age groups had different responses to the website design, as older users took longer to navigate the website and had slower viewing times than younger users, and also looked at the thumbnail images that were further down the page a lot less. This research has filled a gap in the academic literature and provided a detailed understanding of eye fixations, and how they can be interpreted in website design studies. Future academic studies on retailers’ website design can assume that long fixation durations indicate positive attention. Also, the findings enable retailers to understand how the design stimuli affect different ages of users’ shopping experiences and, as a result, can design their website according to their target markets’ behaviour. Furthermore, this study highlights the areas that are not capturing users’ attention on the website, and that need to be improved by retailers. Overall, the findings suggest that retailers should focus on providing an easy, efficient and quick shopping experience for users when designing their website.

M.Phil. (Optometry) / The accommodative-convergence accommodation (AC/A) ratio is a commonly used relationship in the practice of optometry. Many practitioners make use of the AC/A ratio as an aid to diagnostic and prognostic decisions. It is not perfectly clear what effects positive and negative lenses have on the AC/A ratio. A number of questions remain relating to linearity, symptomatology and equality of lens effects on the AC/A ratio. This experiment was designed and conducted in an attempt to answer some of these unanswered questions and forms the foundation of this thesis. A literature review of available knowledge related to this study is detailed. An instrument is described which was constructed to measure the necessary accommodation and vergence changes induced by various stimuli. A sample of first year students at the Rand Afrikaans University department of Optometry was selected according to various acceptance criteria. A total of 109 students were screened , of which a group of 26 students was subjected to the experimental investigation. The data were then subjected to a statistical analysis in an attempt to reveal correlations, linearity and group formations.

M.Phil. (Optometry) / Despite many years of research, diabetic retinopathy (DR), and diabetic macular edema (DME) remain difficult to diagnose, prevent, and treat. The complicated nature of the disease, the limited information on DR and DME and the increasing prevalence of diabetes mellitus (DM) in South Africa, provided motivation for this study. To the best of my knowledge, this is the first study in our country to identify demographic, medical and visual aspects ofDR and DME collectively. A further incentive was the availability in optometry of recently developed computer software based upon multivariate statistics, which provided a unique opportunity to analyze, for example, tri-variate contrast sensitivity acuities using stereo-pair scatter plots. All refractive status measurements were also analyzed and compared with the same method. Together, the results from this study provide a broader clinical and research perceptive on DR and DME. In this cross-sectional study, 202 diabetic patients at the Helen Joseph Hospital in Johannesburg were recruited. Demographic variables included age, gender, race, age of diagnosis, duration of DM, and social habits. Medical variables included systemic conditions present, blood pressures, body mass index (BMI), lipid profiles, glycerated haemoglobin (HbAlc), and other available biochemical data (for example cholesterol, urea and creatinine levels). Visual variables included, distance, pinhole and near visual acuities, contrast sensitivity acuities, refractive status measured with autorefraction, colour vision, Amsler grid, intraocular pressures (lOP), and fundus photography. Administration of the Impact of Visuallmpainnent (IVI) questionnaire provided new information concerning the restrictions in daily living participation caused by DR or DME. The predominant characteristics of the study population consisted of Type 1DM among female Coloured subjects. Approximately 66% of all subjects had also been diagnosed with hypertension. The mean age ofthe subjects was 52 (± 14) years, age of diagnosis 41 (± 13) years, and duration ofDM 10.8 (± 9.7) years. Mean blood pressures (136/81 ± 20.5/11 mmHg) and glycated haemoglobin (HbAlc, 9.9 ± 3.4%) values were slightly higher than the recommended control levels (BP= 120/80 mmHg and HbAlc = 6 to 7%).

M.A.(Clinical Social Work) / In South Africa, trauma is a vivid reality for many children. Unfortunately due to a lack of resources and knowledge, many children in early childhood who experience trauma symptoms are left untreated. Children in this developmental phase of early childhood, have a limited vocabulary, which adds to the challenge of effectively treating trauma through alternative talk therapies. Neurotherapies like Eye Movement Integration Therapy (EMI), which does not rely on the verbal ability of the child, has therefore gained a lot of interest. The goal of this study was to explore whether EMI can be a useful intervention in treating trauma in early childhood. The objectives included to, i) determine whether or not a change in trauma symptoms was evident from pre- to post-EMI intervention, using the Trauma Symptom Checklist for Young Children (TSCYC); ii) explore the perceptions of parents/caregivers regarding EMI’s effectiveness in the reduction of trauma symptoms; and iii) formulate conclusions and recommendations regarding EMI’s implementation as a trauma intervention with children in early childhood. The researcher followed an exploratory design. The one-group pre-test/post-test design was utilised for conducting the study. The study made use of the Trauma Symptoms Checklist for Young Children (TSCYC), a parent/caregiver report that measures the prevalence and intensity of trauma symptoms like anger, anxiety, dissociation, post-traumatic stress intrusion, post-traumatic stress avoidance, post-traumatic stress arousal, post-traumatic stress total and sexual concerns, in order to determine if a single EMI session could produce a change in trauma symptoms. The group was measured prior to the administration of one EMI session, which according to Beaulieu (2004) is sufficient to result in a measurable change in trauma symptoms. Two weeks later the group’s symptoms were re-measured, using the same instrument. The prescribed EMI protocol was followed. Although the findings from studies of EMI with adults and teenagers appear promising, the usefulness of EMI with young children has not been explored. The results from the study indicated that all of the symptoms as measured by the TSCYC reduced significantly (p<.05) after a single EMI session. It would therefore appear as if EMI might be a useful intervention strategy to treat trauma experienced during early childhood.