Many people exposed to sinewave analogues of speech first report hearing them as electronic glissando and, later, when they switch into a dspeech modeT, hearing them as syllables. This perceptual switch modifies their discrimination abilities, enhancing perception of differences that cross phonemic boundaries while diminishing perception of differences within phonemic categories. Using high-density evoked potentials and fMRI in a discrimination paradigm, we studied the changes in brain activity that are related to this change in perception. With ERPs, we observed that phonemic coding is faster than acoustic coding: The electrophysiological mismatch response (MMR) occurred earlier for a phonemic change than for an equivalent acoustic change. The MMR topography was also more asymmetric for a phonemic change than for an acoustic change. In fMRI, activations were also significantly asymmetric, favoring the left hemisphere in both perception modes. Furthermore, switching to the speech mode significantly enhanced activation in the posterior parts of the left superior gyrus and sulcus relative to the non-speech mode. When responses to a change of stimulus were studied, a cluster of voxels in the supramarginal gyrus was activated significantly more by a phonemic change than by an acoustic change. These results demonstrate that phoneme perception in adults relies on a specific and highly efficient left-hemispheric network, which can be activated in top-down fashion when processing ambiguous speech/non-speech stimuli.

& Investigating the degree of similarity between infants ’ and adults ’ representation of speech is critical to our understanding of infants ’ ability to acquire language. Phoneme perception plays a crucial role in language processing, and numerous behavioral studies have demonstrated similar capacities in infants and adults, but are these subserved by the same neural substrates or networks? In this article, we review event-related potential (ERP) results obtained in infants during phoneme discrimination tasks and compare them to results from the adult literature. The striking similarities observed both in behavior and ERPs between initial and mature stages suggest a continuity in processing and neural structure. We argue that infants have access at the beginning of life to phonemic representations, which are modified without training or implicit instruction, but

Mapping from acoustic signals to lexical representations is a complex process mediated by a number of different levels of representation. This paper reviews properties of the phonetic and phonological levels, and hypotheses about how category structure is represented at each of these levels, and evaluates these in light of relevant electrophysiological studies of phonetics and phonology. The paper examines evidence for two alternative views of how infant phonetic representations develop into adult representations, a structure-changing view and a structure-adding view, and suggests that each may be better suited to different kinds of phonetic categories. Electrophysiological results are beginning to provide information about phonological representations, but less is known about how the more abstract representations at this level could be coded in the brain.

I present LIBPHON, a nonparametric regression-based model of phonological acquisition that induces a generalised and productive pattern of vowel harmony—including opaque and transparent neutrality—on the basis of simplified formant data. The model quickly learns to generate harmonically correct morphologically complex forms to which it has not been exposed. 1 Explaining phonological patterns How do infants learn the phonetic categories and phonotactic patterns of their native languages? How strong are the biases that learners bring to the task of phonological acquistion? Phonologists from the rationalist tradition that dominated the past half-century of linguistic research typically posit strong biases in acquisition, with language learners using innatelygiven, domain-specific representations (Chomsky and Halle, 1968), constraints (Prince and Smolensky, 2004) and learning algorithms (Tesar and rules or constraint rankings from which they can classify or produce novel instances. In the last decade, however, there has been a shift toward empiricist approaches to phonological acquisition, use and knowledge. In this literature, eager learning algorithms (Aha, 1997), in which training data are used to update intensional representations of functions or categories then discarded, have been the norm. 1 However, research in related fields—particularly speech perception— indicates that speakers ’ knowledge and use of language, both in production and comprehension, is at least partly episodic, or instance-based

Perception of acoustically complex phonological features in vowels is reflected in the induced brain-magnetic activity