Inter-individual variability in dorsal stream dynamics during word production

Abstract

The current standard model of language production involves a sensorimotor dorsal stream connecting areas in the temporo-parietal junction with those in the inferior frontal gyrus and lateral premotor cortex. These regions have been linked to various aspects of word production such as phonological processing or articulatory programming, primarily through neuropsychological and functional imaging group studies. Most if not all the theoretical descriptions of this model imply that the same network should be identifiable across individual speakers. We tested this hypothesis by quantifying the variability of activation observed across individuals within each dorsal stream anatomical region. This estimate was based on electrical activity recorded directly from the cerebral cortex with millisecond accuracy in awake epileptic patients clinically implanted with intracerebral depth electrodes for pre-surgical diagnosis. Each region's activity was quantified using two different metrics-intra-cerebral evoked related potentials and high gamma activity-at the level of the group, the individual and the recording contact. The two metrics show simultaneous activation of parietal and frontal regions during a picture naming task, in line with models that posit interactive processing during word retrieval. They also reveal different levels of between-patient variability across brain regions, except in core auditory and motor regions. The independence and non-uniformity of cortical activity estimated through the two metrics push the current model towards sub-second and sub-region explorations focused on individualized language speech production. Several hypotheses are considered for this within-region heterogeneity.

Keywords: HGA; fronto-parietal network; iERP; speech production; stereotactic electroencephalography.

FIGURE 1

Anatomical sampling for 16 patients. Each horizontal bar represents the number of recording contacts that were identified to lie in each of the anatomically defined brain regions of interest. Each colour is a patient. For example, superior temporal gyrus was sampled 21 times across three patients, each contributing 8, 7 and 6 contacts, respectively.

FIGURE 2

Correlation of neural responses across the anatomical regions of interest. (a) Scatterplot of within region correlations computed on the time courses averaged per patient (x‐axis) vs. averaged per contact (y‐axis), based on intra‐cerebral event‐related potentials (iERPs) (red) and high gamma activity (HGA) (green) for the different regions of interest (ROIs) (dots). Each dot is an ROI. The metrics computed across patient and across contacts yield very similar results for iERP and HGA. (b) Focus on correlations across contacts, now directly compared for iERP (x‐axis) vs. HGA (y‐axis) time courses. Each dot is an ROI. The two metrics show substantial disparities across regions: Both sig = both metrics show a significant correlation across contacts; HGA sig. = only HGA does; iERP sig. = only iERP does; Neither sig. = neither metric shows a significant correlation across contacts.

FIGURE 3

Different levels of inter‐patient variability of intracerebral activity during word production in four representative regions of the left hemisphere. Left column: intra‐cerebral event‐related potential (iERP) time courses and their significance mask in z score against baseline; middle column: high gamma activity (HGA) time courses and their significance mask in z score against baseline; right column: single trial activity in HGA in z score against baseline, black dots represent response times (where available). The colour of the region names boxes corresponds to those in Figure 2b. (a) ‐Superior temporal gyrus showed highly correlated time‐courses both in the iERP and the HGA metrics. (b) Intra parietal sulcus showed uncorrelated iERP time courses, but correlated HGA time courses. (c) ‐For Pars Opercularis, in the inferior frontal gyrus, the opposite pattern was observed: correlated iERP but uncorrelated HGA. (d) In supramarginal gyrus, neither metric was correlated across patients.

FIGURE 4

Recordings from the supra sylvian area of supra marginal gyrus (SMG) in the left (Rp electrode) and right hemisphere (R electrode) and the left retro‐sylvian area of SMG (Xp electrode) in Patient 16. Note that the activation post vocalization is recorded bilaterally only in the supra‐sylvian area and not in the retro‐sylvian area.

FIGURE 5

Consistency of intra‐cerebral event‐related potentials (iERPs) recorded during a word naming tasks across regions of the left dorsal stream. The regions are ordered by decreasing iERP correlation across contacts and patients to highlight the gradient of variability. From left to right columns: grand average, patient averages and contact‐level statistical masks in z score against baseline

FIGURE 6

Consistency of high gamma activity (HGA) recorded during a word naming tasks across regions of the left dorsal stream. The regions are ordered by decreasing HGA correlation across contacts and patients to highlight the gradient of variability. From left to right columns: grand average, patient averages and contact‐level statistical masks in z score against baseline

FIGURE 7

Correlations (values on the x‐axis) of neural responses across contacts for regions (y‐axis) that were sampled in both hemispheres in the current dataset. L = left; R = right.

FIGURE 8

Distribution of the depth electrode coverage across patients in the fronto‐parietal dorsal network. The colour of each dot indicates whether the corresponding cortical region showed significantly consistent activities in intra‐cerebral event‐related potentials (iERPs) and high gamma activity (HGA) (green), in HGA only (orange), in iERPs only (blue), or in neither of these measures (red).