Joint Representation of Spatial and Phonetic Features in the Human Core Auditory Cortex

Abstract

The human auditory cortex simultaneously processes speech and determines the location of a speaker in space. Neuroimaging studies in humans have implicated core auditory areas in processing the spectrotemporal and the spatial content of sound; however, how these features are represented together is unclear. We recorded directly from human subjects implanted bilaterally with depth electrodes in core auditory areas as they listened to speech from different directions. We found local and joint selectivity to spatial and spectrotemporal speech features, where the spatial and spectrotemporal features are organized independently of each other. This representation enables successful decoding of both spatial and phonetic information. Furthermore, we found that the location of the speaker does not change the spectrotemporal tuning of the electrodes but, rather, modulates their mean response level. Our findings contribute to defining the functional organization of responses in the human auditory cortex, with implications for more accurate neurophysiological models of speech processing.

Revealing the functions of supra-temporal and insular auditory responsive areas in humans

“…To test whether the sound-induced high-γ conform to a certain spatial distribution in either insular or supra-temporal areas, we calculated the Pearson correlation between the contact-level activations and latencies and their MNI coordinates, respectively. As shown in Figure 1E, a high-to-low gradient for activation and an early-to-late gradient for latency along the medial-lateral axis was found on the supra-temporal plane (HG, pSTG, and mSTG) (both p < 0.001), consistent with previous literature (Nourski et al, 2014;Patel et al, 2018). While in the InsP, a low-to-high gradient for activation a late-to-early gradient for latency along the anterior-posterior axis was found (both p < 0.001), consistent with the findings in non-human primates (Remedios et al, 2009).…”

“…The human auditory cortex was more responsive to sounds coming from the Auditory functions in human insula contralateral directions (Ackermann et al, 2011;Patel et al, 2018). However, whether the InsP also has lateral selectivity is still unclear.…”

“…Under the passive listening condition, the aSTG responded sparsely to the non-speech sound. Thus, we defined HG, mSTG, and pSTG as the auditory responsive supra-temporal areas (Nourski et al, 2014;O'Sullivan et al, 2019;Patel et al, 2018).…”

EEG-based detection of the locus of auditory attention with convolutional neural networks

“…It . https://doi.org/10.1101/475673 doi: bioRxiv preprint research (Bednar and Lalor, 2018;Patel et al, 2018;Wolbers et al, 2011) has shown that the direction of auditory attention is neurally encoded, indicating that it could be possible to decode the attended sound position or trajectory from EEG. Especially the alpha power activity could be tracked to determine the locus of auditory attention (Frey et al, 2014;Haegens et al, 2011;Wöstmann et al, 2016).…”

“…This has been investigated based on differences in the EEG entropy features (Lu et al, 2018), but the performance was insufficient for practical use (below 70 % for 60 s windows). However, recent research Lalor, 2018, 2020;O'Sullivan et al, 2019;Patel et al, 2018;Wolbers et al, 2011) has shown that the direction of auditory attention is neurally encoded, indicating that it could be possible to decode the attended sound position or trajectory from EEG. Especially the alpha power activity could be tracked to determine the locus of auditory attention (Frey et al, 2014;Wöstmann et al, 2016), though those studies employed MEG.The aim of this paper is to further explore the possibilities of CNNs for EEG-based AAD.…”