Processing of auditory novelty across the cortical hierarchy: An intracranial electrophysiology study

Abstract

Under the predictive coding hypothesis, specific spatiotemporal patterns of cortical activation are postulated to occur during sensory processing as expectations generate feedback predictions and prediction errors generate feedforward signals. Establishing experimental evidence for this information flow within cortical hierarchy has been difficult, especially in humans, due to spatial and temporal limitations of non-invasive measures of cortical activity. This study investigated cortical responses to auditory novelty using the local/global deviant paradigm, which engages the hierarchical network underlying auditory predictive coding over short ('local deviance'; LD) and long ('global deviance'; GD) time scales. Electrocorticographic responses to auditory stimuli were obtained in neurosurgical patients from regions of interest (ROIs) including auditory, auditory-related and prefrontal cortex. LD and GD effects were assayed in averaged evoked potential (AEP) and high gamma (70-150 Hz) signals, the former likely dominated by local synaptic currents and the latter largely reflecting local spiking activity. AEP LD effects were distributed across all ROIs, with greatest percentage of significant sites in core and non-core auditory cortex. High gamma LD effects were localized primarily to auditory cortex in the superior temporal plane and on the lateral surface of the superior temporal gyrus (STG). LD effects exhibited progressively longer latencies in core, non-core, auditory-related and prefrontal cortices, consistent with feedforward signaling. The spatial distribution of AEP GD effects overlapped that of LD effects, but high gamma GD effects were more restricted to non-core areas. High gamma GD effects had shortest latencies in STG and preceded AEP GD effects in most ROIs. This latency profile, along with the paucity of high gamma GD effects in the superior temporal plane, suggest that the STG plays a prominent role in initiating novelty detection signals over long time scales. Thus, the data demonstrate distinct patterns of information flow in human cortex associated with auditory novelty detection over multiple time scales.

Keywords: Averaged evoked potential; Electrocorticography; High gamma; Human auditory cortex; Predictive coding.

Figure 1.

Local/global deviant experimental paradigm. a: Schematic of the four experimental stimuli. b: Stimulus sequences. c: Comparisons between trials to characterize local and global deviance responses. Adapted from Strauss et al. (2015).

Figure 2.

GD stimulus target detection task performance. Summary of data from six subjects. a: Timing of button presses in each subject. Hits and false alarm responses are shown as long and short vertical lines, respectively. Schematic of the 11-minute-long experimental block is shown on top; recorded instructions and LGD stimulus sequences (see Fig. 1b) are represented by white and gray rectangles, respectively. Numbers represent OAA/S scores, as assessed before and after the experimental block. b: Hit rates (% correctly detected target stimuli), sensitivity (d’) and reaction times (RTs) (upper, middle and lower panel, respectively) for each of the six subjects. Box-and-whiskers RT plot depicts median values and 10th, 25th, 75th and 90th percentiles; median values for each subject are shown inside boxes. Dashed line corresponds to the grand median value across all subjects and hit trials (0.463 s).

Figure 3.

Cortical responses to standard and deviant stimuli along the auditory processing hierarchy. a: MRI top-down view of superior temporal plane and side view of the hemispheric surface showing the locations of four representative recording sites (sites A, B, C, D in HG, STG, MTG and inferior frontal gyrus [IFG], respectively) in subject R369. Colors represent different ROIs used in the present study. b: AEP and high gamma responses (cyan/turquoise and magenta/purple waveforms, respectively) recorded from the four cortical sites in response to standard and deviant stimuli (lighter and darker waveforms, respectively). Lines and shading represent mean values and the 95% confidence intervals, respectively. LD and GD effects are presented on the left and right, respectively. Horizontal bars underneath response waveforms highlight significant differences between responses to standard and deviant stimuli (q<0.05, non-parametric cluster-based permutation test, false discovery rate-adjusted).

Figure 4.

Spatial distribution of sites exhibiting LD and GD effects (panels a and b, respectively) in a representative subject (R369). Sites that feature significant differences between responses to standard and deviant stimuli, as measured by AEP and high gamma responses, are shown in cyan and magenta, respectively. Sites that didn’t exhibit either significant difference are shown in white. Dashed outlines indicate absence of response to the initial portion of the stimuli (“No response”). Sites that were not included in the analysis are marked with X. Insets: MRI coronal section through the right temporal lobe (section plane indicated by a dashed line) showing location of depth electrode contacts in the right amygdala that exhibited AEP local and deviance effects.

Figure 5.

Spatial distribution of sites exhibiting significant LD and GD effects as measured by the AEP and high gamma activity (panels a and b, respectively). Summary of data from six subjects, plotted in MNI coordinate space and projected onto FreeSurfer average template brain. Top-down views of the right superior temporal plane are plotted underneath side views of the right lateral hemispheric convexity, aligned with respect to the y_MNI coordinate. Sites exhibiting significant LD effects only, both LD and GD effects, and GD effects only, are depicted in the left, middle and right column, respectively. Sites are color-coded based on their ROI assignment in each individual subject.

Figure 6.

Temporal properties of LD and GD effects. a: Onset latencies of LD and GD effects (left and right column, respectively) across ROIs. Data from six subjects. Onset latencies of AEP and high gamma effects are shown in cyan and magenta, respectively. Box and whiskers plots (across-contact medians, quartiles, 10th and 90th percentiles) are shown for ROIs with at least ten sites across the six subjects exhibiting significant deviance effects. Gray box represents RTs across the six subjects. b: Time course of LD and GD effects (left and right column, respectively). Total number of sites across all subjects exhibiting significant effects is plotted as a function of time after the fifth vowel onset for AEP and high gamma in cyan and magenta, respectively. Dashed lines represent four exemplar time points depicted in panel c. c: Spatial distribution of LD and GD effects (upper and lower row) at four representative time points (left to right: 100, 225, 400 and 700 ms after fifth vowel onset). Sites that exhibited significant deviance effects at these time points are shown in cyan and magenta for AEP and high gamma, respectively. Data from six subjects, plotted in MNI coordinate space and projected onto FreeSurfer average template brain.