Publicação

Beyond words : a study of local versus global shape processing in dyslexic readers

Ver documento

Detalhes bibliográficos
Resumo:Two processing styles are involved in visual identification of letters/words and objects: either part-based (analytic, local) or holistic (global) processing. Compared with typical readers, dyslexic readers tend to have difficulties implementing local processing to written letters and words. This thesis aimed to explore if these difficulties are restricted to the verbal domain or are more general, affecting the processing of non-linguistic objects. To do so, two experiments were conducted with dyslexic adults and age-matched typical readers to examine object recognition. In Experiment 1, participants performed an object naming task in which we orthogonally manipulated the objects configuration (objects recognized from their global shape vs. objects whose recognition depended on their internal detail) and their visibility (blurred vs. non-blurred images). In Experiment 2, a subset of the same participants performed a difficult object decision task and a superordinate object categorization task (same stimuli) which was assumed to require less object individuation. Relative to controls, dyslexic readers’ performance was especially poor (longer RTs) for “internal detail” objects, exacerbated under difficult encoding contexts (blurred images; Experiment 1), and in a task where (presumably) a coarse global shape processing strategy did not suffice for successful object recognition and consequently participants needed to base also on local shape information (i.e., object decision task; Experiment 2). These results seem to suggest that dyslexics process objects in a different manner than controls, and specifically that they could be impaired at part-based processing for object identification. This extends previous findings for written letters/words, hence, are not domain-specific.
Autores principais:Dias, Diana Sofia Gonçalves
Assunto:Dislexia Reconhecimento visual Processos cognitivos Dissertações de mestrado - 2021
Ano:2021
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Lisboa
Idioma:inglês
Origem:Repositório da Universidade de Lisboa
Descrição
Resumo:Two processing styles are involved in visual identification of letters/words and objects: either part-based (analytic, local) or holistic (global) processing. Compared with typical readers, dyslexic readers tend to have difficulties implementing local processing to written letters and words. This thesis aimed to explore if these difficulties are restricted to the verbal domain or are more general, affecting the processing of non-linguistic objects. To do so, two experiments were conducted with dyslexic adults and age-matched typical readers to examine object recognition. In Experiment 1, participants performed an object naming task in which we orthogonally manipulated the objects configuration (objects recognized from their global shape vs. objects whose recognition depended on their internal detail) and their visibility (blurred vs. non-blurred images). In Experiment 2, a subset of the same participants performed a difficult object decision task and a superordinate object categorization task (same stimuli) which was assumed to require less object individuation. Relative to controls, dyslexic readers’ performance was especially poor (longer RTs) for “internal detail” objects, exacerbated under difficult encoding contexts (blurred images; Experiment 1), and in a task where (presumably) a coarse global shape processing strategy did not suffice for successful object recognition and consequently participants needed to base also on local shape information (i.e., object decision task; Experiment 2). These results seem to suggest that dyslexics process objects in a different manner than controls, and specifically that they could be impaired at part-based processing for object identification. This extends previous findings for written letters/words, hence, are not domain-specific.