Coelho, João Miguel PintoFerreira, Hugo Alexandre Teixeira DuarteSantos, João Pedro Pires de Carvalho dos2019-10-242019-10-2420192019http://hdl.handle.net/10451/39948Tese de mestrado integrado, Engenharia Biomédica e Biofísica (Sinais e Imagens Médicas) Universidade de Lisboa, Faculdade de Ciências, 2019The objective of this work is to study the effect of luminous glare on the electroencephalographic (EEG) signals of subjects that perform concentration-based tasks. The increasing access to high-power and directional light sources (such as laser pointers, but also some flashlights) has led to a growing concern with the potential effects of its use. More than the direct damaging of the retina, the focus has been directed at the effects related to the change in states of concentration on individuals performing tasks whose concentration is critical (such as helicopter pilots or heavy vehicles drivers). This effect is known as ”dazzling” and is typically a temporary deleterious effect on the ability to see or concentrate. However, while damage to the retina can be quantified, glare effects, being indirect (based on the effect on the execution of a given task), are typically qualitative (or at least of more subjective quantification). In this context, the use of brain-computer interfaces capable of analyzing the brain response to external stimuli, opens a door towards the creation of a new tool to evaluate the effects of dazzle. Its potential was evaluated by defining a set of strategies involving the illumination process, EEG signal recording and analysis. A continuous performance task commonly used as an assessment in cognitive neuroscience (N-back) was used to test the attention under the effect of dazzling, in parallel with EEG signals acquisition. Statistical data analysis was performed with the R programming language. ANOVA statistical significant results (p<0.001) for answer scores and latency were obtained for differences between the levels of difficulty, both with or without dazzling. Tukey’s test further revealed that these statistical differences were on the 0-back/2-back and 1-back/2-back pairs (p<0.005). The differences in the pair 0-back/1-back were not significant. Peak band frequency statistical tests were not significant with or without dazzling. Statistical differences were found between dazzling conditions for the frequency band power. For the 0-back and 1-back levels, with the AF7-Fp1 electrode pair, T-student tests resulted in an alpha band frequency power increase (p<0.003, in both cases). The electrode pair AF8-Fp2 resulted in an alfa and beta frequency band increase for the 1-back level (p<0.014 and p<0.029, respectively). These results suggest that concentration is affected by dazzling and can be quantified by means of measuring the change in alpha and beta frequency band power. This technique holds potential and, if further researched and developed, may constitute an effective way of measuring the degree of loss of concentration under the effect of dazzling.engConcentraçãoEncandeamento luminosoInterface cérebro-computadorSinais EEGTeses de mestrado - 2019master thesis202293025