Adaptação e especificidade ao estímulo em respostas evocadas auditivas a vogais

• Ano de defesa: 2015
• Autor(a) principal: Daniel Márcio Rodrigues Silva
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Vogais; P2; Potenciais evocados auditivos; N1; Categorização; Adaptação; Atenuação por repetição; MMN; Neurociências; Complexo N1-P2; Negatividade de incrongruência
• Link de acesso: http://hdl.handle.net/1843/BUBD-9WMUJA

Long-latency evoked potentials in response to vowel sounds were obtained in two studies designed to investigate neurophysiological manifestations of within-category and acrosscategory distinctions in passive listening conditions. The sound stimuli were synthesizedvowel sounds selected from a continuum crossing the boundaries between categories /i/ and /e/ and between /e/ and //. In Study I, N1 and P2 responses were investigated in an adaptation paradigm in which pairs of successive sounds were presented in order to examine the effect of the first sound (S1) presentation on the response to the second sound(S2). For each participant (N=26), three exemplars of the /i/ category and three exemplars of the /e/ category formed a series, in equal steps, along the vowel continuum: i1; i2; i3; e1; e2; e3. Each trial consisted of the presentation of one of six possible pairs: i3-i3; e1-e1; i1-i3; e3-e1;e2-i3; i2-e1. Adaptation effects (S2 response attenuation) on N1 and P2 were observed for all pairs. Responses to S2 in identical sounds pairs (i3-i3; e1-e1), within-category pairs (i1-i3; e3-e1) and across-category pairs (i2-e1; e2-i3) were compared and no significant differences wereobserved N1 responses being particularly similar across pairs. These results indicate that the observed adaptation effects are not stimulus-specific. Systematic N1/P2 amplitude variations associated to differences between S1 sounds suggest that underlying neuralpopulations do not respond to vowels in a non-specific manner. In Study II, the MMN response was elicited in a passive oddball paradigm. For each participant (N=16), three sounds (i; e1; e2) varying in equal steps along the continuum were selected in such a way that the difference between the first two sounds crosses the /i/-/e/ category boundary. Giventhe across-category i-e1 pair and the within-category e1-e2 pair, two oddball sequences were generated from each pair such that the roles of deviant and standard stimulus were inverted. In this way, the sound e1 occurred as deviant and as standard in both within- and acrosscategoryconditions. In agreement with previous results, larger MMN amplitudes inresponses to the e1 deviant in the across-category than in the within-category condition indicate that the MMN reflects representations of vowel categories. Taken together, the results of studies I and II suggest that MMN generation involves different mechanisms from the ones manifested in adaptation paradigms such as the one employed in Study I and, therefore, provide no support to proposals according to which MMN can be entirely explained by N1 stimulus-specific adaptation.