Aberrant inhibitory processing in the somatosensory cortices of cannabis-users

Abstract

Background:

Delta-9 tetrahydrocannabinol (THC) is a major exogenous psychoactive agent, which acts as a partial agonist on cannabinoid (CB₁) receptors. THC is known to inhibit presynaptic neurotransmission and has been repeatedly linked to acute decrements in cognitive function across multiple domains. Previous electrophysiological studies of sensory gating have shown specific deficits in inhibitory processing in cannabis-users, but to date these findings have been limited to the auditory cortices and the degree to which these aberrations extend to other brain regions remains largely unknown.

Methods:

We used magnetoencephalography (MEG) and a paired-pulse somatosensory stimulation paradigm to probe inhibitory processing in 29 cannabis-users (i.e., at least four times per month) and 41 demographically matched non-user controls. MEG responses to each stimulation were imaged in both the oscillatory and time domain, and voxel time series data were extracted to quantify the dynamics of sensory gating, oscillatory gamma activity, evoked responses, and spontaneous neural activity.

Results:

We observed robust somatosensory responses following both stimulations, which were used to compute sensory gating ratios. Cannabis-users exhibited significantly impaired gating relative to non-users in somatosensory cortices, as well as decreased spontaneous neural activity. In contrast, oscillatory gamma activity did not appear to be affected by cannabis use.

Conclusions:

We observed impaired gating of redundant somatosensory information and altered spontaneous activity in the same cortical tissue in cannabis-users compared to non-users. These data suggest that cannabis use is associated with a decline in the brain’s ability to properly filter repetitive information and impairments in cortical inhibitory processing.

Funding:

National Institutes of Health, USA.

Introduction:

Delta-9 tetrahydrocannabinol (THC), is one of the major potentially psychoactive compounds found in cannabis (Laaris et al., 2010), is an agonist of the cannabinoid (CB₁) receptor and primarily inhibits GABA mediated neurotransmission. (Paronis et al., 2012; Laaris et al., 2010). THC has been repeatedly linked to decrements in various cognitive processes both acutely and longer-term (Crean et al., 2011). For example, deficits in selective attention (Solowij et al., 1995), executive function (Cohen and Weinstein, 2018), episodic and working memory (Schoeler and Bhattacharyya, 2013; Solowij and Battisti, 2008), sustained attention (Nicholls et al., 2015), pre-pulse inhibition (Scholes and Martin-Iverson, 2009), and modulation of visual and auditory processing (Kempel et al., 2003; Schwitzer et al., 2015) have been associated with cannabis use.

Beyond cognitive domains, deficits in auditory sensory gating have been well documented in cannabis-users (Broyd et al., 2013; Edwards et al., 2009). Sensory gating is a well-studied neurophysiological phenomenon that is thought to reflect the brain’s ability to filter or inhibit redundant sensory information and thereby help preserve limited neuronal resources for processing of novel task-relevant information (Cromwell et al., 2008). Sensory gating is typically evaluated used paired-pulse designs whereby an identical pair of stimuli are presented a short time apart (e.g., 500 milliseconds apart). The brain response to the second stimulus is strongly attenuated in healthy adults and such “gating” is typically quantified using an amplitude ratio (i.e., second stimulation (S₂)/first stimulation (S₁)). As mentioned above, a large volume of literature has shown reduced auditory gating in cannabis-users (Broyd et al., 2013; Edwards et al., 2009; Patrick and Struve, 2000; Patrick and Struve, 2002; Patrick et al., 1999), with most studies using the early auditory-evoked potential (i.e., P50) as the brain response of interest. This disruption in auditory gating may be attributable to reduced inhibitory processing secondary to cannabis use, as many studies have linked gating deficits to age- and disease-related decline in inhibitory processing (Miller and Freedman, 1995; Bhattacharyya et al., 2015; Spooner et al., 2019; Light and Braff, 1999; Cheng et al., 2016b; Cheng et al., 2016a). Further, studies have shown that administration of GABA antagonists can impair sensory gating (Ma and Leung, 2011; Hershman et al., 1995). The mechanisms are not entirely clear at the neural population level, but altered dynamics in the theta range have been linked to GABAergic neurons (Buzsáki and Wang, 2012; Freund, 2003), and some have shown theta activity to be central to sensory gating deficits in the context of cannabis (Skosnik et al., 2018). GABAergic inhibitory interneurons are also known to modulate local gamma oscillations among pyramidal cells (Brunel and Wang, 2003; Bartos et al., 2007; Buzsáki and Wang, 2012; Fries et al., 2007; Fries, 2009; Fries, 2015; Singer, 1999; Salkoff et al., 2015; Vinck et al., 2013; Uhlhaas et al., 2009; Uhlhaas and Singer, 2012), and neural activity in the gamma range (i.e., 30-75 Hz) has been of major interest in previous studies of gating and inhibitory processing, especially in the context of somatosensory gating (Spooner et al., 2018; Kurz et al., 2018; Cheng et al., 2016b; Wiesman et al., 2017; Spooner et al., 2019). Unfortunately, while auditory gating has been extensively studied in cannabis-users, no study to date has probed gating in the somatosensory system of cannabis-users, despite the evidence suggesting presence of CB1 receptors in the human somatosensory cortex and animal studies showing that CB1 receptor activation results in the modulation of sensory-evoked responses (Bloomfield et al., 2019; Fortin and Levine, 2007).

To this end, we used high-density magnetoencephalography (MEG) and a paired-pulse electrical stimulation paradigm to examine the impact of cannabis use on evoked somatosensory responses, oscillatory gamma responses, and spontaneous neural activity in the somatosensory cortex. Our primary hypotheses were two-fold in that we expected cannabis-users to have both reduced sensory gating and altered spontaneous neural activity in the somatosensory cortices relative to demographically matched non-user controls. Such findings would indicate aberrant local inhibitory function and somatosensory processing in cannabis-users.

Methods

Participants:

Seventy-two healthy adults (range: 20-45 years old) were enrolled in this study, including 31 cannabis-users (9 females) and 41 nonuser controls (14 females). Participants were selected based on their cannabis use, or lack thereof, from a large ongoing MEG study of healthy and pathological aging (R01-MH103220). Briefly, the cannabis-users group included all participants who were 20-45 years old and smoked cannabis at least four times per month, while the control group included the nonusers who were in the same age range. To avoid any acute effects, participants were instructed to abstain from cannabis use prior to their visit on the day of the study. These two groups did not statistically differ in regard to mean age, sex, or handedness (all ps > .325). Exclusionary criteria for this study included any medical illness affecting CNS function (e.g., HIV/AIDS, lupus), any neurological or psychiatric disorder, history of head trauma, any other illicit substance use (besides cannabis), and the MEG laboratory’s standard exclusion criteria (e.g., ferromagnetic implants). Written informed consent was obtained from each participant after a thorough description of the study was provided, following the guidelines of the University of Nebraska Medical Center’s Institutional Review Board, which approved the study protocol.

Experimental Paradigm:

Participants were seated in a nonmagnetic chair with their head positioned within the MEG helmet-shaped sensor array. Electrical stimulation was applied to the right median nerve using external cutaneous stimulators connected to a Digitimer DS7A constant- current stimulator system (Digitimer Ltd, Garden City, UK). For each participant, we collected at least 80 paired-pulse trials with an inter-stimulus interval of 500 ms and an inter-pair interval that randomly varied between 4500 and 4800 ms. Each pulse generated a 0.2 ms constant-current square wave that was set to a limit of 10% above the threshold that was required to elicit a subtle twitch of the thumb.

MEG Data Acquisition:

All recordings were conducted in a one-layer magnetically shielded room with active shielding engaged for environmental noise compensation. With an acquisition bandwidth of 0.1-330 Hz, neuromagnetic responses were sampled continuously at 1 kHz using an Elekta MEG system (Helsinki, Finland) with 306 sensors, including 204 planar gradiometers and 102 magnetometers. During data acquisition, participants were monitored via real-time audio-visual feeds from inside the shielded room. Each MEG dataset was individually corrected for head motion and subjected to noise reduction using the signal space separation method with a temporal extension (Taulu and Simola, 2006).

Structural MRI Processing and MEG Co-registration:

Prior to MEG measurement, four coils were attached to the subject’s head and localized, together with the three fiducial points and scalp surface, with a 3-D digitizer (FASTRAK 3SF0002, Polhemus Navigator Sciences, Colchester, VT, USA). Once the subjects were positioned for MEG recording, an electric current with a unique frequency label was fed to each of the coils. This induced a measurable magnetic field and allowed each coil to be localized in reference to the sensors throughout the recording session. As coil locations were also known with respect to head coordinates, all MEG measurements could be transformed into a common coordinate system. With this coordinate system, each participant’s MEG data were coregistered with their T1-weighted structural magnetic resonance images (sMRI) prior to source space analyses using BESA MRI (Version 2.0; BESA GmbH, Gräfelfing, Germany). All sMRI data were acquired with a Philips Achieva 3T X-series scanner using an 8-channel head coil (TR: 8.09 ms; TE: 3.7 ms; field of view: 240 mm; slice thickness: 1 mm; no gap; in-plane resolution: 1.0 × 1.0 mm). All sMRI data were aligned parallel to the anterior and posterior commissures and transformed into standardized space (i.e., Talairach space). Following source imaging (i.e., beamforming), each participant’s functional MEG images were also transformed into standardized space using the transform that was previously applied to the structural MRI volume and spatially resampled.

MEG Preprocessing, Time-frequency Transformation, and Sensor-Level Statistics:

Eye blinks and cardiac artifacts were removed from the data using signal space projection (SSP), which was accounted for during source reconstruction (Uusitalo and Ilmoniemi, 1997). Epochs were of 3700 ms duration, with 0 ms defined as the onset of the first stimulation and the baseline being the −700 to −300 ms time window. Of note, we shifted our baseline away from the period immediately preceding stimulus onset to avoid potential contamination by any anticipatory responses, although there was no evidence of such anticipatory responses in our final analyses. Epochs containing artifacts were rejected based on a fixed threshold method, supplemented with visual inspection. On average, 68 trials per participant were used for further analysis, and the average number of trials accepted did not statistically differ by group (p = .460).

For the time domain (i.e., evoked) analysis, artifact-free epochs were averaged across trials to generate a mean time series per sensor, and the specific time windows used for subsequent source analysis were determined by statistical analysis of the sensor-level time series across both groups and the entire array of gradiometers. Each data point in the time series was initially evaluated using a mass univariate approach based on the general linear model. To reduce the risk of false positive results while maintaining reasonable sensitivity, a two-stage procedure was followed to control for Type 1 error. In the first stage, paired-sample t-tests were conducted to test for differences from baseline at each data point and the output time series of t-values was thresholded at p < .05 to define time-points containing potentially significant responses across all participants. In stage two, the time points that survived the threshold were clustered with temporally and/or spatially neighboring time points that were also above the threshold, and a cluster value was derived by summing all of the t-values of all data points in the cluster. Nonparametric permutation testing was then used to derive a distribution of cluster-values, and the significance level of the observed clusters (from stage one) were tested directly using this distribution (Ernst, 2004; Maris and Oostenveld, 2007). For each comparison, 10,000 permutations were computed to build a distribution of cluster values, and the time windows of time-domain evoked data that were non-exchangeable with baseline (p < .001) across all participants according to these permutation analyses were used to guide subsequent time-domain source-level analysis.

To simultaneously investigate the oscillatory responses commonly associated with somatosensory processing, we also transformed the same post-artifact-rejection epochs into the time-frequency domain using complex demodulation (Kovach and Gander, 2016), and the resulting spectral power estimations per sensor were averaged over trials to generate time-frequency plots of mean spectral density (5.0 Hz, 10 ms; range: 10-100 Hz). These sensor-level data were normalized per time-frequency bin using the respective bin’s baseline power, which was calculated as the mean power during the −700 to −300 ms time period. The time-frequency windows used for source analysis were again determined by means of a paired-sample cluster-based permutation test against baseline across all participants, and the entire frequency range (10–100 Hz). Further methodology details can be found in a recent paper (Wiesman and Wilson, 2020).

MEG Source Imaging and Statistics:

Time domain source images were computed using standardized low-resolution brain electromagnetic tomography (sLORETA; regularization: Tikhonov .01%; Pascual-Marqui, 2002). The resulting whole-brain maps were 4-dimensional estimates of current density per voxel, per time sample across the experimental epoch. These data were normalized to the sum of the noise covariance and theoretical signal covariance, and thus the units are arbitrary. Using the temporal clusters identified in the sensor-level analysis (see Results), these maps were averaged over time following each somatosensory stimulation and then grand-averaged across the two stimulations to determine the peak voxel of the time-domain neural response to the stimuli across all participants. From this peak, the sLORETA units were extracted per stimulation to derive estimates of the time-domain response amplitude for each participant.

Time-frequency resolved beamformer source images were computed using the dynamic imaging of coherent sources (DICS; regularization: singular value decomposition .0001%; Groß et al., 2001) approach, which applies spatial filters in the time-frequency domain to calculate voxel-wise power for the entire brain volume. The single images were derived from the cross-spectral densities of all combinations of MEG gradiometers averaged over the time-frequency range of interest and the solution of the forward problem for each location on a grid specified by input voxel space. Following convention, we computed noise-normalized source power for each voxel per participant using active (i.e., task) and passive (i.e., baseline) periods of equal duration and bandwidth (Hillebrand et al., 2005) at a resolution of 4.0 x 4.0 x 4.0 mm. Such images are typically referred to as pseudo-t maps, with units (pseudo-t) that reflect noise-normalized power differences (i.e., active versus passive) per voxel. MEG pre-processing and imaging used the Brain Electrical Source Analysis (version 6.1) software.

Voxel time-series data (i.e., “virtual sensors”) were extracted from each participant’s data individually using the peak voxel in the grand-averaged beamformer images. To compute the virtual sensors, we applied the sensor weighting matrix derived through the forward computation to the preprocessed signal vector, which yielded a time series for the specific coordinate in source space. Note that virtual sensor extraction was done per participant once the coordinates of interest were known. Once the virtual sensor time series were extracted, we computed the envelope of the spectral power in the frequency bin that was used in the beamforming analysis. From this time series, we computed the relative (i.e., baseline-corrected) and absolute (i.e., not baseline-corrected) response time series of each participant in both groups.

To evaluate group differences in sensory gating, we used independent sample t-tests. To evaluate other neural response parameters, 2 X 2 repeated-measure ANOVAs (stimulation by group) were conducted using SPSS (Version 23.0, IBM Analytics, Armonk, New York, USA).

Results:

All 72 participants were able to successfully complete the MEG and MRI aspects of the study. However, two male cannabis-users were excluded from the analyses due to excessive artifacts in their MEG data and/or technical problems. The remaining 70 participants (29 cannabis-users and 41 nonuser controls) had a mean age of 30.24 years old for cannabis-users and 31.56 years-old for non-user controls. This difference was not significant (p = .483). In addition, the final groups did not differ in handedness (p = .325), sex (p = .789), or on other demographic variables (Table 1).

Table 1:

Demographics & Substance Use

	Non users	Users	p value
Demographics	(n = 41)	(n = 29)
Mean age in years (SD)	31.56 (7.72)	30.24 (7.69)	p = .377
Sex, % Females	9 (31%)	14 (34.15%)	p = 1.00
% Right-handed	96.55	97.56	p = .291
Mean years of Education (SD)	17.58 (2.45)	14.78 (1.55)	p < .001
Race (frequency %)			p = .614
White	25 (60.97)	20 (68.97)
Non-white	16 (39.03)	9 (31.03)
Mean BDI score (SD)	2.78 (2.96)	4.68 (3.83)	p = .023
Substance use
Mean Cannabis use/week	NA	9 (12)
AUDIT-C (SD)	2.90 (1.79)	4.52 (2.72)	p = .003

BDI = Beck depression inventory

Notably, the two groups did significantly differ in education, alcohol use, and depression, and we conducted follow-up analyses to ensure these differences did not affect our main MEG results (see below).

Time-domain (Evoked) Analysis:

We initially time-domain averaged the sensor level data and conducted paired t-tests against baseline across all participants and gradiometers to identify time periods of interest. These analyses indicated responses from 20-100 and 520-600 ms (p < .001, corrected), corresponding to responses to stimulation one and two, respectively. Next, we applied sLORETA to compute whole-brain images of the evoked responses in each participant and then averaged these images within the time periods found in the sensor-level analysis. These images were then grand averaged across both stimulation periods and all participants to identify the peak voxel. The time series of this peak voxel was then extracted and used to examine how cannabis affects evoked response amplitude to each stimulation and the sensory gating ratio (Figure 1A). First, we computed the sensory gating ratio by dividing the mean power of the response to stimulation two by that of stimulation one in each participant. The resulting quotient, or sensory gating ratio, reflects the individual's capacity to “gate” the second stimulus in an identical pair, with smaller values indicating stronger gating (i.e., better suppression of redundant stimuli). Group differences in the sensory gating ratio were examined using an independent samples t-test, which revealed that sensory gating in the somatosensory system was significantly weaker in cannabis-users relative to nonusers, t(67) = −2.26, p = .027 (Figure 1B). To ensure that this difference in the gating ratio was not driven by group differences in response power to either stimulation, independent sample t-tests were conducted. These showed that the response power did not significantly differ for the first, t(67) = 1.16, p = .251 (Figure 1C), or second, t(67) = .35, p = .726 (Figure 1D), stimulation between cannabis-users and non-users. Addionally, for thoroughness, we examined the other response parameters using a repeated-measures 2X2 ANOVA, which showed a main effect of stimulation, F(1,67) = 49.59, p < .001, and a significant stimulation by group interaction, F(1,67) = 4.41, p = .039. The group main effect was not significant, F(1,67) = .63, p = .43.

Fig. 1.

Group differences in the time-domain somatosensory gating response. (A) The time domain averaged data of each group extracted from the peak voxel of grand-averaged sLORETA source images showed robust responses to each stimulation. The x-axis denotes time in ms, while power in arbitrary units is represented on the y-axis. The grand-averaged time-domain map, collapsed across both stimulations and groups, is shown in the top left corner. (B) Independent samples t-tests revealed a higher gating ratio (i.e., impaired gating) in cannabis users (shown in red) compared to the nonuser controls (shown in blue). Notably, the response power between the two groups did not differ significantly either to stimulation 1 (C) or 2 (D). Error bars reflect the SEM. *p < .05.

Post-hoc paired sample t-tests between the first and second stimulation revealed a weakened response to stimulation two compared to stimulation one across all participants, t(68) = 7.29, p < .001 (Figure 1). Note that the latter finding is essentially the gating effect expressed as an amplitude difference (instead of a ratio). Finally, post-hoc testing of the interaction revealed no group differences in the strength of the response to either stimulation, but that the difference in response strength between the two stimulations differed by group, t(67) = −2.10, p = .039. Again, this is just the group-wise gating difference finding without controlling for total amplitude differences.

Oscillatory Gamma Analysis:

Next, we used the time-frequency transformed data and found significant increases in many sensors near the sensorimotor and parietal regions from about 10 Hz to 90 Hz during the first 100 ms after the onset of stimulus one and stimulus two (p < 0.001, corrected; Figure 2A). The higher-frequency responses (>30 Hz) were clearly much stronger in the initial 50 ms following stimulus onset, whereas activity in the lower frequency range (< 20 Hz) was smeared across a longer time period. To evaluate the dynamics and remain consistent with the traditional gamma band, we focused our beamformer analyses on the higher 30–75 Hz frequency range and utilized two 50 ms time intervals in which the neural response to stimulation was the strongest (0–50 ms and 500–550 ms).

Fig. 2.

Time-frequency responses to somatosensory stimulation of the right median nerve. (A): Time-frequency spectrogram from a MEG sensor near the sensorimotor cortices with time indicated in ms on the x-axis and frequency in Hz on the y-axis. Stimulations occurred at 0 and 500 ms. The percent change from baseline is indicated by the color bar on the right. (B): Group averaged beamformer images (pseudo-t) for stimulation 1 (left), stimulation 2 (middle), and the group's grand-average (right). Strong increases in power were found in virtually identical areas of the contralateral hand region of the somatosensory cortex in nonuser controls (upper panel) and cannabis users (lower panel). These maps were grand averaged across both stimulations and groups to identify the peak voxel, which was followed by virtual sensor extraction and additional analyses.

These time-frequency windows were imaged using a beamforming approach, which revealed virtually identical locations in the contralateral somatosensory hand region for the first and second stimulations (Figure 2B). Next, we averaged the beamformer images across both stimulations and groups and utilized the resulting peak in this grand-averaged map to extract voxel time-series data for each participant. Using these voxel time series (i.e., virtual sensor data), we computed the amplitude envelope for the band of interest (30-75 Hz) in both relative (i.e., baseline-corrected) and absolute (i.e., not baseline-corrected) units. The relative amplitude values were averaged across the time period used in the beamformer analysis.

Following the same method as the evoked analyses, we computed the sensory gating ratio for the gamma responses in each participant and conducted an independent-samples t-test. This revealed no significant difference in gating between cannabis-users and nonusers, t(64) = −.966, p = .338. Bayesian testing provided additional support for the null hypothesis, BF₀₁ = 2.64, error % = 0.0004. The repeated-measures 2 X 2 ANOVA examining amplitude effects showed a main effect of stimulation, F(1,66) = 48.17, p < .001, and a significant stimulation by group interaction, F(1,66) = 5.64, p = .020. Again, there was no main effect of group, F(1,66) = 72.97, p = .332. Follow up paired t-tests indicated a weakened response to stimulation two compared to stimulation one across both groups t(66) = 7.13, p < .001. Thus, participants exhibited significant gating in the 30-75 Hz gamma range. Examination of the interaction term showed that the amplitude difference between the two stimulations differed by group, t(66) = −2.37, p = .011. However, since the gating ratio for gamma power did not statistically differ between users and nonusers, this interaction effect was likely biased by numerical amplitude differences between groups and should be interpreted with caution.

Finally, to evaluate whether spontaneous gamma activity immediately preceding stimulus onset differed as a function of group, we examined the absolute power time-series data (i.e., not baseline-corrected) in the same peak voxel used in the previous analysis. We computed the mean power during the baseline period (−700 to −300 ms) in each participant and conducted an independent-samples t-test, which revealed that spontaneous gamma was significantly weaker in cannabis-users relative to controls in the postcentral gyrus, t(66) = 2.04, p = .046 (Figure 3).

Fig. 3.

Spontaneous neural activity. (A): Absolute voxel time series envelope extracted from the peak voxel and averaged across the 30-75 Hz range in each group, with the gray window highlighting the baseline period. (B): Independent samples t-tests of the baseline period (gray window; −700 to −300 ms) revealed significantly reduced spontaneous gamma activity in cannabis-users (red) relative to the nonuser controls (blue). *p < .05.

Lastly, since the two groups differed statistically in years of education, depression (i.e., BDI), and AUDIT-C scores (all ps < 0.024), we correlated each with our significant MEG findings (i.e., gating ratio and spontaneous gamma) to test for their potential confound in our analyses. No significant correlations were reported (all ps > .245).

Discussion:

In the current study, we employed spatially resolved MEG and a paired-pulse electrical stimulation paradigm to investigate the impact of cannabis use on somatosensory processing and gating. In the time-domain, we observed robust evoked responses following both stimulations in the contralateral postcentral gyrus. We then computed the sensory gating ratio using the peak voxel time series, and this revealed significantly reduced gating in cannabis-users compared to nonusers. Interestingly, this gating effect appeared to be specific to evoked responses, as our gamma-band analyses indicated statistically equivalent sensory gating between groups, despite robust gamma responses and gating across both groups. Finally, we found significantly decreased spontaneous gamma activity (30-75 Hz) prior to stimulation onset in cannabis-users versus nonusers. Below, we discuss the implications of these novel findings and future directions. Sensory gating in cannabis-users has been extensively (and exclusively) studied in the auditory domain, and these previous studies have repeatedly shown reduced gating in cannabis-users, primarily utilizing electroencephalographic measures (Zachariou et al., 2008; Broyd et al., 2013; Edwards et al., 2009; Broyd et al., 2016; Rentzsch et al., 2007). This deficit in auditory sensory gating is consistent with the current findings, where we observed reduced gating in the somatosensory domain in cannabis-users. Generally, in a paired-pulse paradigm, the response to the first stimulation reflects the brain’s capacity to attend to a salient stimulus, while the response to the second stimulus is a measure of the brain’s ability to suppress the redundant information (Miller and Freedman, 1993; Miller and Freedman, 1995; Hajós, 2006). While the key mechanisms underlying such suppression are not fully understood, impaired gating has been linked to deficits in inhibitory brain function (Spooner et al., 2019; Cheng et al., 2015; Cheng et al., 2016a; Thoma et al., 2017; Light and Braff, 1999). Interestingly, previous studies using neuropsychological assessments in cannabis-users have reported impaired inhibitory processing in the cognitive domain (e.g., stop signal and Stroop tests; Greenwood and Parasuraman, 1994; Hooker and Jones, 1987; Hart et al., 2001). The fMRI literature also corroborates this overall framework by reporting aberrant neuronal functioning during response inhibition tasks (e.g., Go/No-Go task) in cannabis-users compared to controls (Smith et al., 2011; Maij et al., 2017). Though the exact mechanism remains unclear, delta-9 THC and other exogenous compounds found in cannabis are believed to modulate synaptic activity by suppressing the release of several presynaptic neurotransmitters, including GABA, glutamate, noradrenaline, dopamine, and acetylcholine (Schlicker and Kathmann, 2001; Iversen, 2003; Kim and Thayer, 2000). Such altered neurotransmission may affect inhibitory circuitry and disrupt both neural sensory gating responses, as well as behavioral performance in response inhibition tasks.

Our second major finding was that cannabis-users had reduced spontaneous neural activity during the baseline period in the contralateral somatosensory hand region of the postcentral gyrus. This finding has not been previously reported and can be tied back to the inhibitory effects of cannabinoids on GABAergic interneurons (Romero et al., 1997), which are among the primary generators of the gamma rhythm (Wyss et al., 2017; Buzsáki and Wang, 2012). Such reduced basal gamma power in cannabis-users may be central to the alterations in neural and behavioral inhibitory function. However, further investigation in this area is direly needed, and a multimodal imaging approach that includes both GABA neurotransmitter quantification via magnetic resonance spectroscopy (MRS) and MEG-derived measures of spontaneous neural activity could provide critical information and at least partially validate this framework. Given the increased legalization of recreational and medicinal cannabis use in the United States and worldwide, such studies would be very timely from a neuroscience and public health perspective.

In conclusion, we quantified sensory gating, a measure of basic inhibitory processing, in the somatosensory cortex of cannabis-users and nonusers. We observed impaired sensory gating of evoked responses to the paired-pulse stimuli, as well as altered spontaneous gamma power prior to the stimulus onset in cannabis-users compared to nonusers. Our findings suggest that the use of cannabis impairs the brain’s ability to properly filter redundant information, leading to inefficient information processing. Though this study provides novel and critical insight, it was not without limitations and future studies should work to address these. For example, we used a constant inter-pulse interval of 500 ms, which was based on our recent methodological paper (Spooner et al., 2020). While our recent methods paper noted 500 ms to be the ideal latency for gating, it is possible that other latencies would have shown larger group-wise effects. A large body of studies provides evidence that the inter-stimulus interval is directly tied to response amplitude (Ermutlu et al., 2007; Rentzsch et al., 2008; Dolu et al., 2001), but whether substance abuse or cannabis would directly affect this parameter remains unknown. Moreover, we relied on self-report metrics (i.e, questionnaires about participants’ use of cannabis) and using more objective measures in future studies could provide better reliability and validity. Future studies should also evaluate whether the age of cannabis use onset, the last time participants smoked nicotine and/or cannabis relative to the time of the MEG scan, degree of nicotine craving at scan time, amount of cannabis consumed per use episode, and/or the overall duration of use impacts inhibitory function. Given the wide variation in the potency of cannabis, such questions are often difficult to precisely answer but obviously remain important. Finally, examining possible interaction effects between cannabis use and various disease states (e.g., neuroHIV, schizophrenia) in the context of sensory gating and behavioral inhibitory function would be an interesting avenue, especially given the proposed linkage between the endocannabinoid system and etiology of schizophrenia (Hajós et al., 2008; Dissanayake et al., 2013).