Temporal pain processing in the primary somatosensory cortex and anterior cingulate cortex

Abstract

Pain is known to have sensory and affective components. The sensory pain component is encoded by neurons in the primary somatosensory cortex (S1), whereas the emotional or affective pain experience is in large part processed by neural activities in the anterior cingulate cortex (ACC). The timing of how a mechanical or thermal noxious stimulus triggers activation of peripheral pain fibers is well-known. However, the temporal processing of nociceptive inputs in the cortex remains little studied. Here, we took two approaches to examine how nociceptive inputs are processed by the S1 and ACC. We simultaneously recorded local field potentials in both regions, during the application of a brain-computer interface (BCI). First, we compared event related potentials in the S1 and ACC. Next, we used an algorithmic pain decoder enabled by machine-learning to detect the onset of pain which was used during the implementation of the BCI to automatically treat pain. We found that whereas mechanical pain triggered neural activity changes first in the S1, the S1 and ACC processed thermal pain with a reasonably similar time course. These results indicate that the temporal processing of nociceptive information in different regions of the cortex is likely important for the overall pain experience.

Fig. 1

Design of a multi-region LFP-based neural interface for pain. A We recorded local field potentials (LFPs) form the rostral area of the anterior cingulate cortex (ACC) and primary somatosensory cortex (S1) using silicon probe arrays in rats. B Raw LFP signals were processed to compute three band-limited LFP power features ${\{\mathbf{Y}}_k^{\mathrm{ACC}}\}$ and ${\{\mathbf{Y}}_k^{\mathrm{S}1}\}$ form ACC and S1 channels (low gamma (30–50 Hz), high gamma (50–100 Hz), and ultra-high frequency (300–500 Hz)), and sent to an automated decoder based on a state space model (SSM), which independently inferred the latent variables ${\{z}_k^{\mathrm{ACC}}\}$ and ${\{z}_k^{\mathrm{S}1}\}$. C In the presence of noxious stimulus, the SSM determined the relative change in Z-score from baseline and used this change as a proxy for acute pain signals

Fig. 2

Pain-evoked event-related potentials (ERPs) and Z-score analysis of the pain decoder showed temporal sequence in cortical response to a mechanical stimulus. A Illustration of raw LFP trace of ACC and S1. ERPs are marked by black triangles. Onset of a noxious mechanical stimulus (pin prick, PP) is marked by the black vertical line. The red curve indicates LFP signals in the ACC and the blue curve indicates LFP signals in the S1. The arrow marks the time of the ERP peak latency. B Comparison of the ERP peak latency between the ACC and S1 (n = 96 trials from 5 rats). On average, the ERP peak latency in the ACC was longer than that of the S1 (n = 96; ****p < 0.0001, paired t test). C Illustration of Z-score curve of the SSM-based pain decoder in the ACC and S1 (see Methods for details). Z-score peak is marked by the black triangle. The onset of PP is marked by the black vertical line. The red curve indicates the Z-score curve in the ACC and the blue curve indicates Z-score curve in the S1. The arrow marks the time of the Z-score peak latency. D Comparison of Z-score peak latency between the ACC and S1 (n = 96 trials from 5 rats). The latency in the ACC was longer than the latency in the S1 (n = 96; **p = 0.0014, paired t test)

Fig. 3

Pain-evoked ERPs and pain decoder did not show clear temporal sequence in cortical processing of a thermal pain stimulus. A Illustration of Z-score curve of the ACC and S1. The onset of the thermal stimulus is marked by the black vertical line, whereas the paw withdrawal time is marked by the brown vertical line. We set the maximum absolute value of the Z-score curve between the thermal stimulus onset and paw withdrawal as the Z-score peak value, which is marked by the black triangle. The red curve indicates the Z-score curve in the ACC, and the blue curve indicates the Z-score curve in the S1. The arrow marks the time of the Z-score peak. B There is no significant difference in the Z-score peak latency between the ACC and S1 (n = 48 trials from 5 rats; p = 0.1551, paired t test)