Deguchi, C. 1 , Besson, M. 2 , Boureux, M. 1 , Sarlo, M. 1 & Colombo, L. 1
1 Dipartimento di Psicologia Generale, Padova University, Padova, Italy
2 Institut de Neurosciences Cognitives de la Méditerranée, CNRS, Marseille, France
We investigated attentive and non-attentive auditory processing of familiar and unfamiliar vowels in Italian native speakers using the Event-related potentials (ERPs). An auditory odd-ball sequence including 4 vowels (/u/ as standard, and /o/, /y/ and /ø/ as deviants), each spoken by different speakers, was presented, while participants watched a silent movie (Passive condition) and discriminated deviant vowels (Active condition). Importantly, /u/ and /o/ belong to the phonemic inventory of Italian, and are phonologically distinctive, while /y/ and /ø/ exist only as dialectal allophones. In the F1-F2 acoustic space, the deviant vowels differed from the standard in different dimensions; F1 for /o/, F2 for /y/ and both F1/F2 for /ø/. Previous studies showed that the mismatch negativity (MMN) indexes pre-attentive categorization of vowels based on long-term memory traces for the native phonemes (Näätänen et al., 1997; Deguchi et al., 2010). We were interested in investigating if the patterns of components elicited by familiar and unfamiliar vowels, in the passive condition, or under attentional focus in the active discrimination task, differ, and how (Sussman et al., 2004).
Results showed no clear MMN, but in the Passive condition the P3a and RON were elicited by unfamiliar vowels (/y/ and /ø/) reflecting involuntary attention switch (Escera et al., 2000). In the Active condition, discrimination accuracy was higher for /ø/ than the other deviants showing contribution of both acoustic dimensions (F1 and F2). The N2, P3 and N4 components differed in amplitude and latency between vowels reflecting complex interactions between the native phonemic inventory and perceptual discriminability of speech sounds. Overall results suggest that the sequence of familiar and unfamiliar vowels including acoustically variable tokens did not induce automatic linguistic-mode categorization, but unfamiliar vowels evoked novelty effect requiring additional resources and different cognitive processes compared with familiar vowels.