The Pathways for Intelligible Speech: Multivariate and Univariate Perspectives

Abstract

An anterior pathway, concerned with extracting meaning from sound, has been identified in nonhuman primates. An analogous pathway has been suggested in humans, but controversy exists concerning the degree of lateralization and the precise location where responses to intelligible speech emerge. We have demonstrated that the left anterior superior temporal sulcus (STS) responds preferentially to intelligible speech (Scott SK, Blank CC, Rosen S, Wise RJS. 2000. Identification of a pathway for intelligible speech in the left temporal lobe. Brain. 123:2400–2406.). A functional magnetic resonance imaging study in Cerebral Cortex used equivalent stimuli and univariate and multivariate analyses to argue for the greater importance of bilateral posterior when compared with the left anterior STS in responding to intelligible speech (Okada K, Rong F, Venezia J, Matchin W, Hsieh IH, Saberi K, Serences JT,Hickok G. 2010. Hierarchical organization of human auditory cortex: evidence from acoustic invariance in the response to intelligible speech. 20: 2486–2495.). Here, we also replicate our original study, demonstrating that the left anterior STS exhibits the strongest univariate response and, in decoding using the bilateral temporal cortex, contains the most informative voxels showing an increased response to intelligible speech. In contrast, in classifications using local “searchlights” and a whole brain analysis, we find greater classification accuracy in posterior rather than anterior temporal regions. Thus, we show that the precise nature of the multivariate analysis used will emphasize different response profiles associated with complex sound to speech processing.

Comprehending auditory speech: previous and potential contributions of functional MRI

“…Using this metric, they argued that posterior STS showed the greatest invariance to the acoustic properties of the stimuli. In another fMRI replication of this paradigm, Evans, Kyong et al (2014) used both univariate and multivariate analyses, showing the strongest univariate intelligibility response in left anterior STS but the greatest multivariate classification accuracy in local patches of cortex (searchlights) in the posterior STS. Yet, when exploring classifications in which information could be integrated across the whole temporal lobe, they found that the most informative voxels showing preferential responses to intelligible speech were located in anterior regions.…”

“…By now, a number of studies have used pattern classification as a metric for quantifying the relative contribution of different regions to acoustic and linguistic processes in the case of connected speech signals (Abrams et al, 2012;Evans, Kyong et al, 2014;McGettigan, Evans et al, 2012;Okada et al, 2010). Correia et al (2014) extended this approach using an innovative cross decoding approach (see Kaplan, Man, & Greening, 2015 for a review of cross decoding) to demonstrate the encoding of abstract word representations independent of speech acoustics.…”

Cross-language differences in the brain network subserving intelligible speech

“…Although traditional universal grammar theory argues that functional components of linguistic ability are manifest without being taught (10), recent connectionist theory within a neural network approach emphasizes interactions among primary systems of neuronal processing units that support language acquisition and use, where the weights of connections among these units are gradually changed during learning and thus highly constrained by the unique feature of a given language (9,11,12). Related with connectionist views, the dual pathway model of language based on evidence from neuroimaging and anatomical studies was also framed to interpret the neural basis of language comprehension and production (13, 14), especially in speech comprehension.Intelligible speech is processed hierarchically in human neocortex, with the anterior temporal (13,15,16) and frontal cortices (Broca's area) operating at a higher hierarchical level than the posterior region centered on superior temporal sulcus/gyrus (pSTS/pSTG, core of Wernicke's area), which itself receives inputs from primary and secondary auditory cortices (17). In the dual ventral-dorsal pathway model, the dorsal-stream pathway of speech processing assumes primary processing begins in the Significance Language processing is generally left hemisphere dominant.…”

“…Intelligible speech is processed hierarchically in human neocortex, with the anterior temporal (13,15,16) and frontal cortices (Broca's area) operating at a higher hierarchical level than the posterior region centered on superior temporal sulcus/gyrus (pSTS/pSTG, core of Wernicke's area), which itself receives inputs from primary and secondary auditory cortices (17). In the dual ventral-dorsal pathway model, the dorsal-stream pathway of speech processing assumes primary processing begins in the Significance Language processing is generally left hemisphere dominant.…”

Effects of acoustic periodicity, intelligibility, and pre-stimulus alpha power on the event-related potentials in response to speech

“…Becker, Pefkou, Michel, & Hervais-Adelman, 2013;Ding, Chatterjee, & Simon, 2014;Obleser & Weisz, 2012;Peelle, Gross, & Davis, 2013) and imaging work (e.g. Davis & Johnsrude, 2003;Evans et al, 2014;Scott, Blank, Rosen, & Wise, 2000).…”