Aferição de usabilidade de interfaces assistivas sob o prisma da computação afetiva

• Ano de defesa: 2022
• Autor(a) principal: Mezencio, Rovilson
• Orientador(a): Lemos, Rodrigo Pinto
• Banca de defesa: Lemos, Rodrigo Pinto, Gomide, Renato Sousa, Alves, Ricardo Henrique Fonseca, Guedes, Leonardo Guerra de Rezende, Vasques, Thiago Lara
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Goiás
• Programa de Pós-Graduação: Programa de Pós-graduação em Engenharia Elétrica e da Computação (EMC)
• Departamento: Escola de Engenharia Elétrica, Mecânica e de Computação - EMC (RG)
• País: Brasil
• Palavras-chave em Português: Computação afetiva; Software assistivo; Testes psicométricos; Interface homem máquina; Comunicação alternativa e aumentativa
• Palavras-chave em Inglês: Affective computing; Assistive software; Psychometric tests; Man-machine interface; Alternative and augmentative communication
• Área do conhecimento CNPq: ENGENHARIAS::ENGENHARIA ELETRICA::TELECOMUNICACOES
• Link de acesso: http://repositorio.bc.ufg.br/tede/handle/tede/12069

The usability of interfaces is connected to the perception and emotion of each user. The way each individual feels the world around them is their own. Assess or test an interface from a cognitive point of view and measure with a simple note whether or not it can be improved in a generic way without discovering if groups of individuals need special attention or if they react in a similar way, it is already an ancient science. Affective computing arises to study whether these perceptions, emotions, attitudes and behaviors can be measured and how this can improve interfaces so that there is greater personalization in the use of interfaces. Assistive Software interfaces need to be more personalized than interfaces are general-purpose. Each user has their specific difficulty and different personalities. Psychometric tests in their variety and scientificity have been used to classify sets of users and verify if their needs can be grouped. The idea now is to find out if people within groups with similar skills, competences, personalities, way of acting and thinking may have similar technological needs when using some technology. This work raised the techniques and methodologies for usability assessment of assistive or non-assisted interfaces. Then, a bibliographic research was carried out to understand which would be the best personality classifier available in Portuguese. Furthermore, methodologies were sought in Affective Computing to assess emotions activated in the use of an available and free assistive tool. Then, an attempt was made to assess the quality of an interface, as well as the grade or level it reached Also, the perceptions caused by its use were verified and if there is a way to make a grouping based on the personalities mapped by an inventory based on the Big Five. A survey was carried out with 96 (ninety-six) participants who were grouped into four different personalities in order to verify if different personalities have similar cognitive and emotional perceptions when performing a task in an assistive tool. Compiling the data and using specific statistical techniques for the correlation study, it is concluded that there are indications that people with the same personality, classified using the Personality Characteristics Inventory-ICP, a national method based on the Big Five , may have similar cognitive and emotional perceptions.