Phonetic adjustment to novel talkers in L2 listeners and in autism
Abstract: Human listeners generally perceive the speech of talkers that they have never previously encountered efficiently and without difficulty. Previous studies have shown that when listeners hear an ambiguous speech sound, they are able to disambiguate it using their existing lexical knowledge. This ability to rapidly retune native phoneme categories and adapt to the speech of a novel talker is extremely robust: it has been documented across languages, in children and older adults, and for a variety of phoneme types. Within the lab, perceptual learning may be investigated using a simple 2-part experiment (Norris, McQueen, & Cutler, 2003). First, during an initial ‘learning phase’, listeners complete a lexical decision task in which they hear spoken words and indicate whether they are words or non-words. Crucially, some of the words contain an ambiguous sound midway between two phonemes. Second, immediately following the exposure phase, participants categorise sounds along a continuum. Typically, category expansion is observed consistent with the exposure to the ambiguous sound that occurred in the learning phase. In this talk, I will present findings from a series of studies. First, I will outline the literature that has examined phonetic adjustments in bilinguals, who may show perceptual learning effects in their L1, L2 or both languages. Second, I will present recent work examining perceptual learning in autistic adults, who often show atypicalities in perceptual, social and language processing. Based on the evidence, I will systematically examine the factors that may determine whether perceptual learning effects do or do not emerge.