Carga cognitiva e sua correlação com piscadas durante o uso de um simulador de cadeira de rodas

• Ano de defesa: 2020
• Autor(a) principal: Salgado, Débora Pereira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Biomédica
• Palavras-chave em Português: Carga cognitiva; Simulador de cadeira de rodas; NASA-TLX; Piscadas; EEG; Tecnologia Assistiva; Presença; Realidade Virtual; Cognitive Load; Wheelchair Simulator; Blinks; Assistive Technology; Usability; Presence; Virtual Reality; CNPQ::ENGENHARIAS::ENGENHARIA BIOMEDICA; CNPQ::ENGENHARIAS::ENGENHARIA BIOMEDICA::BIOENGENHARIA::PROCESSAMENTO DE SINAIS BIOLOGICOS; Cadeiras de rodas; Realidade virtual
• Link de acesso: https://repositorio.ufu.br/handle/123456789/29909; http://doi.org/10.14393/ufu.di.2020.626

Driving a wheelchair is difficult and complex, where it requires motor, visual and cognitive skills. There are simulators aimed mainly at users without or with little experience of using this assistive technology as a form of safe training. However, the real wheelchair and the simulator need to be developed or improved so that the users feel adapted and motivated to use them. Developing technologies to assist people without using assessments that help to understand and bring users closer to developers leads to giving up and discouraging the use of these tools. Many of these assessments are made using subjective metrics, such as interviews or questionnaires; and objective metrics are often related to user performance. Generally, these subjective metrics are classified as explicit, which are the responses that the user is aware of. These metrics can often not reflect people’s genuine opinions and feelings. In contrast, implicit measures infer mental content based on tasks, that can be evaluated continuously while the task is being performed. In this perspective, the present work aimed to assess the cognitive load during the use of the virtual reality training tool, the wheelchair simulator. The study proposes the use of implicit metrics, specifically, the blink indexes derived from an EEG signal. In summary, the study analysed implicit and explicit metrics of a task performed in the wheelchair simulator in a virtual environment. The explicit metrics were derived from questionnaires on usability, presence, cognitive load and performance. The implicit metrics, on the other hand, were the characteristics of the frequency, duration and velocity of the blinks collected during the experiment. The results were consistent with the reports found in the literature, where, the cognitive load that requires a mental-visual demand is related to the decrease in the frequency, duration and velocity of the blink. The results of the correlation analyses between the NASA-TLX scores (subjective cognitive load assessment) and the blink parameters were moderate and weak. Thus, at the moment, for the application presented, it is concluded that evaluating only the blinks is not enough to infer the cognitive load, however it is a metric that applied in conjunction with questionnaires can show more relevant information than just the application of questionnaires.