Warning: file_put_contents(/opt/frankenphp/design.onmedianet.com/storage/proxy/cache/7752d186b6b839ea153ab6629b0747e5.html): Failed to open stream: No space left on device in /opt/frankenphp/design.onmedianet.com/app/src/Arsae/CacheManager.php on line 36

Warning: http_response_code(): Cannot set response code - headers already sent (output started at /opt/frankenphp/design.onmedianet.com/app/src/Arsae/CacheManager.php:36) in /opt/frankenphp/design.onmedianet.com/app/src/Models/Response.php on line 17

Warning: Cannot modify header information - headers already sent by (output started at /opt/frankenphp/design.onmedianet.com/app/src/Arsae/CacheManager.php:36) in /opt/frankenphp/design.onmedianet.com/app/src/Models/Response.php on line 20
Emotion Recognition in Individuals with Cocaine Use Disorder: The Role of Abstinence Length and the Social Brain Network - PMC Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 1.
Published in final edited form as: Psychopharmacology (Berl). 2021 Jun 5;239(4):1019–1033. doi: 10.1007/s00213-021-05868-x

Emotion Recognition in Individuals with Cocaine Use Disorder: The Role of Abstinence Length and the Social Brain Network

Rachel A Rabin 1, Muhammad A Parvaz 1,2, Nelly Alia-Klein 1,2, Rita Z Goldstein 1,2
PMCID: PMC8689230  NIHMSID: NIHMS1762896  PMID: 34089343

Abstract

Rationale:

Emotion recognition is impaired in drug addiction. However, research examining the effects of cocaine use on emotion recognition yield mixed evidence with contradictory results potentially reflecting varying abstinence durations.

Objectives:

Therefore, we investigated emotion recognition and its neural correlates in individuals with cocaine use disorder (CUD) parsed according to abstinence duration.

Methods:

Emotion recognition performance was compared between current cocaine users (CUD+, n=28; cocaine-positive urine), short-term abstainers (CUD–ST, n=23; abstinence<6-months), long-term abstainers (CUD-LT, n=20; abstinence≥6-months) and controls (n=45). A sample subset (n=73) underwent structural magnetic resonance imaging to quantify regional gray matter volume (GMV) using voxel based morphometry.

Results:

CUD+ demonstrated greater difficulty recognizing happiness than CUD-ST and controls, and sadness and fear compared to controls (p<0.01). For fear, CUD-ST also performed worse than controls (p<0.01), while no differences emerged between CUD-LT and controls. Whole-brain analysis revealed lower GMV in the bilateral cerebellum in CUD+ compared to CUD-LT and controls; a similar pattern was observed in the amygdala (CUD+<CUD-LT) (pFWE<0.01). Collapsed across all participants, poorer recognition for happiness was associated with lower right cerebellar GMV (pFWE<0.05).

Conclusions:

Emotion recognition is impaired with current cocaine use, and selective deficits (in fear) may persist with up to 6-months of abstinence. Lower cerebellar GMV may underlie deficits in positive emotion recognition. Interventions targeting emotional-social-cognitive deficits, especially among active users, may enhance treatment success for individuals with CUD.

Keywords: Cocaine, Cocaine Use Disorder, Social Cognition, Emotion Recognition, Social function, Abstinence, Amygdala, Cerebellum

1. INTRODUCTION

The six basic emotions, happiness, surprise, sadness, anger, fear, and disgust, are universally recognized (Ekman 1972; Ekman et al. 1987) and evidence suggests that the recognition of each emotion may be modulated by different neural substrates (LeDoux 2000; Phillips et al. 2003). Identifying these emotions in facial expressions is a critical element of social cognition as they guide appropriate social responses and interactions. Emotion recognition dysfunction constitutes a hallmark of substance use disorders (Castellano et al. 2015) and a contributing factor to real-life social difficulties that may lead to diminished emotional support and social isolation (Preller et al. 2014). Given the significance of social connectedness in recovery (Bathish et al. 2017), ameliorating emotion recognition deficits may contribute to improved treatment outcomes in addiction.

The evidence for an association between drug use, especially psychostimulants, and emotion recognition has been mixed. On the one hand, studies demonstrate that, compared to occasional users and controls, stimulant users who use at least monthly are selectively impaired in fear recognition (Kemmis et al. 2007; Morgan and Marshall 2013; Verdejo-Garcia et al. 2007), while other studies show that this impairment also extends to different negative emotions such as anger (Ersche et al. 2015) and sadness (Kim et al. 2011). On the other hand, several studies suggest comparable general ability for identifying basic facial expressions between stimulant users and controls (Fox et al. 2011; Payer et al. 2008; Woicik et al. 2009). Results from pharmacological studies are also mixed. Bedi et al (2016) showed that smoked cocaine had no effect on emotion recognition in weekly cocaine users. Other pharmacological challenge studies conducted in recreational stimulant users demonstrated positive associations between stimulant use and emotion recognition. For example, a placebo-controlled study found that d-amphetamine enhanced the ability to identify both positive and negative emotions in previous recreational drug users (Wardle et al. 2012). Similarly, experimental cocaine administration in current recreational poly-drug users improved the ability to identify negative emotions, namely fear, anger, disgust, and sadness (Kuypers et al. 2015). In general, this pattern of mixed results, or the absence of a direct relationship between stimulant use duration and emotion recognition performance among users (Kemmis et al. 2007; Kim et al. 2011), suggests a critical role for other factors. One such factor may be recency of drug use and the importance of accounting for abstinence length in relation to cognitive function has repeatedly been highlighted in the addiction literature (Basterfield et al. 2019).

Evidently, no specific pattern of relationships between emotion recognition performance and cocaine recency emerged from the above reviewed studies, potentially reflecting between-study heterogeneity in the methodologies employed [challenge studies, cross-sectional designs, functional magnetic resonance imaging (MRI) paradigms] as well as variability in the included participant groups (regular cocaine users, polysubstance users, individuals with cocaine use disorders). Moreover, objective markers that corroborate self-reported cocaine abstinence were not included in all studies while the range of abstinence length within and between studies was extensive (1 day to 12-months); the latter factors are crucial given that abstinence duration may affect outcomes in a non-linear fashion. In a review of studies in chronic cocaine users, Potvin et al (2014) demonstrated that intermediate abstainers (≤12 weeks abstinent) showed greater impairments across multiple cognitive domains compared to current cocaine users (urine-positive for cocaine) and long-term abstainers (≥20 weeks abstinence) (Potvin et al. 2014). This inverted U-shaped trajectory may reflect the impact of recent cocaine use in masking underlying cognitive impairments (Pace-Schott et al. 2008; Woicik et al. 2009), whereby active cocaine users may self-medicate with cocaine, a stimulant, to acutely relieve underlying/pre-existing cognitive impairments (e.g., attention, executive function) (Jovanovski et al. 2005), which may also improve social-cognitive impairments such as emotion recognition. The underlying neurobiological mechanism of this effect, that could also be subsumed under a negative “withdrawal” effect on cognition, may encompass dopamine depletion [resulting from down regulation of dopamine synthesis and release in response to chronic cocaine administration (Dackis and Gold 1985)] and its normalization by cocaine administration (that acutely increases firing in the mesolimbic and mesocortical dopaminergic tracts (Gawin 1991)], together temporarily “normalizing” cognitive function (Hoff et al. 1996). Alternatively, this inverted U-shaped trajectory may reflect the phenomenon of incubation of craving (i.e., where craving increases over the first several weeks of abstinence, remaining high for an extended period of time) (Parvaz et al. 2016), as previously suggested by our group.

Neuroimaging studies propose a specialized social brain network, comprised of the medial prefrontal cortex, orbitofrontal cortex, insula, amygdala, and fusiform gyrus, that supports social cognition (Adolphs 2009; Choudhury et al. 2006; Johnson et al. 2005) including emotion recognition (Dal Monte et al. 2013). Given its extensive interconnectedness with limbic regions of the brain, the cerebellum, specifically the posterior region, has also been implicated in social cognitive processing (Baillieux et al. 2008; Van Overwalle et al. 2014). Notably, the structural integrity of the social brain network extending to the cerebellum has been associated with emotion recognition performance in clinical populations (Adolphs et al. 1994; Ersche et al. 2015; Kipps et al. 2007; Pera-Guardiola et al. 2016) and with social functioning outcomes (Bickart et al. 2011; Noonan et al. 2018). Structural compromises of these same brain regions are also consistently reported in chronic cocaine users. Using voxel based morphometry (VBM), studies demonstrate that individuals with current cocaine use disorders (CUD) had lower gray matter volume (GMV) in prefrontal, insular, and cerebellar regions compared to healthy controls (Alia-Klein et al. 2011; Ersche et al. 2011; Rando et al. 2013; Sim et al. 2007). Importantly, abstinence-mediated GMV increases have been reported in stimulant users such that longer periods of sobriety were associated with greater GMV in the frontal cortex, insula, and, cerebellum as accompanied by improvements in attention, working memory, and executive function (Connolly et al. 2013; Hirsiger et al. 2019; Parvaz et al. 2017). Morphological recovery with drug abstinence may reflect increased dendritic arborization, neuronal cell body volume, and/or neuroplasticity in the form of metabolic recovery (Kalivas 2007; Siemsen et al. 2019; Zahr et al. 2010). Few attempts, however, have been made to relate GMV in these regions to emotion recognition in individuals with CUD while concurrently accounting for length of cocaine abstinence.

Therefore, using a cross-sectional design, we performed a secondary analysis comparing active cocaine users (CUD+, cocaine-positive urine), recent abstainers (CUD-ST, cocaine-negative urine with ≤6-months of abstinence), long-term abstainers (CUD-LT, cocaine-negative urine with >6-months of abstinence), and healthy controls on emotion recognition. A subset of participants underwent a T1-weighted MRI scan to quantify whole-brain GMV. We predicted that, compared to all other groups, CUD-ST would have the poorest emotion recognition performance for negative emotions (Hypothesis A). In addition, we predicted that, compared to CUD-LT and healthy controls, CUD+ and CUD-ST would have lower GMV in regions in the social brain network and the cerebellum (Hypothesis B). Lastly, the exact nature of the interdependence between emotion recognition and GMV as well as their relative contribution to social function remains unclear; therefore, as an exploratory analysis we investigated the mediating role of emotion recognition in the relationship between GMV of the social brain network and social function.

2. METHODS AND MATERIALS

2.1. Participants

Cocaine users and healthy controls were recruited through local advertisements, treatment facilities, and by word-of-mouth. Participants who met our inclusion criteria (see below) were selected from previous or ongoing neuroimaging protocols conducted in our lab between 2013 and 2017 [previous results were published in e.g., (Bachi et al. 2018; Gan et al. 2019; Moeller et al. 2020; Moeller et al. 2018)]. Studies were approved by the Icahn School of Medicine at Mount Sinai Institutional Review Board. Participants provided written informed consent and received compensation for their participation.

The sample consisted of 71 individuals with CUD and 45 healthy controls (CTL) (Table 1). Cocaine-using participants met criteria for CUD as assessed by the Structural Clinical Interview for DSM-IV/DSM-5 and identified cocaine as their primary drug of choice administered by smoking, n=51, nasally, n=17, a combination, n=2, or intravenously, n=1. Participants with CUD were stratified a priori according to their cocaine urine status and extended abstinence duration: active users, those urine-positive for cocaine to index recent use (<72 hours) (CUD+; n=28); individuals with short-term abstinence defined by negative urine for cocaine and self-reported abstinence of less than 6-months (CUD–ST, n=23); and individuals with long-term abstinence defined by negative urine for cocaine and self-reported abstinence for 6-months or longer (CUD-LT, n=20). The 6-month cut-off was based on previous studies demonstrating the trajectory of cognitive function (Potvin et al. 2014) and GMV recovery associated with protracted abstinence (Parvaz et al. 2017). Note that self-reported abstinence, in combination with urine analysis, is considered a valid and reliable measure for similar scientific and clinical types of research in addiction, as recommended by NIDA (Campbell et al. 2012) and that markers such as hair samples do not provide precise information about abstinence length (Cooper et al. 2012). Note also that the clinical interviewers were blind to these abstinence/urine-based groupings. Exclusion criteria for all participants were (i.) diagnosis of major psychiatric disorders (other than substance use disorders in the cocaine group); (ii.) positive urine screens for psychoactive drugs/metabolites (amphetamine, methamphetamine, phencyclidine, benzodiazepine, cannabis, opiates, barbiturates or inhalants) with the exception of cocaine for CUD+; (iii.) history of head trauma or loss of consciousness (>30 minutes) or other neurologic diseases; (iv.) verbal IQ<75 measured by age-corrected scores on the Wide Range Achievement Test-3 (Wilkinson 1993); (v.) medical diseases requiring hospitalization/regular monitoring; (vi.) medication in the previous six months affecting the central nervous system; (vii.) positive urine pregnancy test in women; and (viii.) MRI contraindications.

Table 1.

Demographic and Clinical Characteristics of the Study Sample

N=116 CUD+ (n=28) CUD-ST (n=23) CUD-LT (n=20) CTL (n=45) p-value
Sex (M/F) * 20/8 22/1a 18/2a 30/15c,d 0.02
Race (Caucasian/AA/Other) 0/26/2 2/20/1 2/13/5 7/32/6 0.10
Age * 49.6±6.0a 47.7±8.0a 45.8±7.4a 39.4±9.2b,c,d <0.01
Education 12.7 ±2.5 12.6±1.7 12.7±1.5 13.6±2.9 0.29
Verbal IQ (WRAT) * 95.5±9.4 93.7±10.9a 98.5±10.5 100.9±9.5c 0.02
Non Verbal IQ (Matrix Reasoning) 9.5±3.4 10.3±3.2 9.5±2.9 10.7±2.8 0.36
Depression (BDI) 5.3±3.9 6.2±8.3 6.8±7.6 4.1±5.9 0.46
Hours of sleep/night † 6.4±2.1 6.9±1.8 6.8±1.5 6.5±1.2 0.69
Sleep quality (very badly/badly/fairly well/well/very well) † 0/4/8/9/6 0/0/9/7/6 0/1/4/6/9 0/4/12/11/18 0.49
Duration of cocaine abstinence (in days) * 2.0±1.4d (range 0-5 days) 35.9±40.6d (range 23-120 days) 1603.6±1487.8b,c (range 182-5840 days) n/a <0.01
Cocaine age of onset 23.1±7.2 23.1±5.0 23.2±5.8 n/a 0.99
Lifetime cocaine use (years) * 22.4±7.6d 19.5±8.2 15.8±6.7b n/a 0.02
Days of cocaine use in past 30 Days * 13.6±7.8c,d 2.70±4.4b 0.0b n/a <0.01
Withdrawal (CSSA) * 20.4±11.1c,d 10.8±9.4b 13.0±14.8b n/a 0.01
Craving (CCQ) * 26.7±11.3c,d 13.6±10.1b,d 4.5±7.3b,c n/a <0.01
Cigarette smokers (current/past/never) * 26/0/2a 17/6/0a 11/7/2a 11/5/29b,c,d <0.01
CPD of current smokers 7.1±5.6 6.8±5.5 8.3±5.9 4.2±2.5 0.41
Lifetime alcohol (years) * 21.3±9.7a 17.4±10.4 20.6±10.7a 11.1±11.3b,d <0.01
Lifetime alcohol use to Intoxication (years) * 3.9±6.7d 5.0±7.9 9.8±10.0a,b 1.5±3.7d <0.01
Days of alcohol use in past 30 Days * 6.7±5.5a,c,d 1.4±2.3b 1.3±3.6b 1.3±2.5b <0.01
Alcohol to intoxication in past 30 days 0.8±2.61 0.4±1.3 0.0±0.0 0.1±0.6 0.14
Current AUD (n) 2 3 1 n/a 0.61
Past SUD (n) 15 14 17 n/a 0.07
Open in a new tab
*

p<0.05

a

, mean differs from CTL;

b

, mean differs from CUD+;

c

, mean differs from CUD-ST;

d

, mean differs from CUD-LT;

AA, African American; AUD, alcohol use disorder; BDI, Beck Depression Inventory; CCQ, Cocaine Craving Questionnaire; CPD, cigarettes per day; CSSA, Cocaine Selective Severity Assessment; CTL, control; CUD+, participants testing positive for cocaine; CUD-ST, participants testing negative for cocaine and with <6-months abstinence; CUD-LT, participants testing negative for cocaine and with >6-months abstinence; F, female; M, male; n, number of participants; SUD, substance use disorder; WRAT, Wide Range Achievement Test.

†

missing data, CUD+, n=1 and CUD-ST, n=1

2.3. Measures

2.3.1. Clinical and Substance Use Measures

Participants underwent a clinical interview conducted by trained clinicians and/or research staff, which included the Structured Clinical Interview for DSM for Axis I Disorders (American Psychiatric Association 2013) and the Addiction Severity Index (ASI) (McLellan et al. 1992).

As an exploratory approach, we used four separate measures from the ASI family-social subscale as a proxy for social functioning: (i.) the interviewer severity rating (a subjective rating by the interviewer of their overall general impression of the participant); (ii.) the composite score (a summary score computed from the ASI family-social subscale indexing outcome from the participants’ perspective); and the following two questions: (iii.) “With whom do you spend most of your free time? Family, friends, or alone?”; and (iv.) “How many close friends do you have?” The ASI was also used to quantify cocaine use duration (years) and abstinence (days). The Cocaine Selective Severity Assessment Scale (Kampman et al. 1998) and the Cocaine Craving Questionnaire (Tiffany et al. 1993) assessed cocaine withdrawal and craving, respectively, over the past 24 hours; the Beck Depression Inventory II assessed depressive symptoms in the previous two weeks (Beck 1996). Sleep patterns were assessed (on the day of the MRI for the VBM subsample and on the day of ERT administration for all other participants), by asking participants about the quantity of sleep they had the previous night and to grade its quality on a 5-point scale (1 = very badly, 2 = badly, 3 = fairly well, 4 = well, 5 = very well).

2.3.2. Emotion Recognition Test

The ERT, from the Cambridge Neuropsychological Test Automated Battery, was used to assess the ability to correctly identify the six basic emotions in facial expressions: happiness, sadness, fear, anger, surprise, and disgust. Computer-morphed images derived from the facial features of real individuals, each showing a specific emotion, were displayed on the screen one at a time for 200 milliseconds. Participants selected the emotion that the face displayed from the six emotions. Both accuracy and speed of responding were emphasized.

2.4. Statistical Analysis for Emotion Recognition

All statistical analyses were performed in SPSS Statistics 24.0. Group differences in demographic, clinical, and drug variables were analyzed using chi-square tests for categorical variables and one-way analysis of variance (ANOVA) for continuous variables. Two mixed 6 (emotion: happiness, surprise, sadness, anger, fear, disgust) × 4 (group: CUD+, CUD-ST, CUD-LT, CTL) repeated-measures ANOVAs (which are robust to violations of assumptions of normality) were conducted, one for ERT accuracy (percent correct), and one for ERT reaction time. Significant main effects of emotion and group were followed by t-tests. To identify the source of significant interactions, we conducted separate one-way AN(C)OVAs for the six emotions and four groups, and paired t-test comparing each of the six emotions within each group and comparing groups for each emotion. Post-hoc results of simple effects were Bonferroni-corrected to account for multiple comparisons [for emotion: p<0.00083=p≤0.05/60 (15 paired comparisons × 4 groups) and group: p<0.0083=p≤0.05/6 (6 comparisons)]. In analyses where Mauchly’s test showed a violation of the sphericity assumption, Greenhouse-Geisser corrections were used.

Demographic and clinical variables that showed between-group differences (see Table 1) were correlated with ERT using Spearman correlations [given that demographic and clinical variables were not normally distributed (with the exception of age and WRAT where Pearson correlations were employed)] across all participants. Cocaine variables were evaluated exclusively in CUD participants (both within subgroups and across all CUD participants). If the variable significantly correlated with ERT outcomes (p<0.05 with a correlation coefficient of >0.40; note the reduced thresholds as compared to our main analyses, i.e., incorporating the magnitude of the coefficient as a better indicator for potential confounds as compared to the standard and more conservative Bonferroni correction used elsewhere in the manuscript), it was included as a covariate for that analysis. Effect sizes were determined using partial eta-squared and Cramer’s V as appropriate. These analyses were performed to test Hypothesis A.

Group differences in our exploratory social function outcome were analyzed using non-parametric Kruskal-Wallis tests due to non-normality. Significant group differences were further explored with Mann-Whitney U tests. For categorical variables, chi-square tests were employed. Post-hoc tests were corrected for multiple comparisons [p<0.002=p≤0.05/24 (4 ASI outcomes × 6 comparisons)].

2.5. Voxel Based Morphometry

2.5.1. MRI Data Acquisition and Processing

A subset of individuals (N=73: CUD+, n=14; CUD-ST, n=16; CUD-LT, n=16; CTL, n=27) was scanned within approximately one month (range: 0-35 days) of ERT administration. For the CUD+ group, all but one participant had a positive urine drug screen for cocaine on the scan day; the one participant who tested negative for cocaine was scanned within one week of ERT administration (when the participant had a positive urine test). There were four participants in the CUD-ST group who tested positive for cocaine on scan day; these scans were performed within 10 days of ERT. All other subjects across the other groups tested negative for cocaine and any other drugs on scan day.

T1-weighted MRI scans were collected with a 3T Skyra (Siemens, Erlangen, Germany) using a vendor provided 32-channel head coil, and a 3D MPRAGE sequence [FOV 256 × 256 × 179 mm3, 0.8 mm isotropic resolution, TR/TE/TI=2400/2.07/1000 ms, flip angle 8° with binomial (1, −1) fat saturation, bandwidth 240 Hz/pixel, echo spacing 7.6 ms, in plane acceleration (GRAPPA) factor of 2, total acquisition time ~7 min]. Images were preprocessed using the “HCP PreFreeSurfer structural pipeline” (based on FSL 5.0.6 and FreeSurfer 5.3.0-HCP) to align the origin to the anterior–posterior commissure line and correct for image distortions (bias-field inhomogeneities).

2.5.2. Voxel Based Morphometry

Statistical Parametric Mapping 12 (SPM12; Wellcome Department of Cognitive Neurology, University of London, London, UK) and the Computational Anatomy Toolbox (CAT12, Christian Gaser, Department of Psychiatry, University of Jena, Jena, Germany) running on MATLAB 2017 (MathWorks, MA, USA) were used to preprocess the structural images with default parameters provided by the CAT12 toolbox as in previous studies (Kaag et al. 2018; Lotze et al. 2019). Each image was first displayed in SPM12 to screen for artifacts. The preprocessing steps involved the following: (i.) T1 images were spatially normalized to the Montreal Neurological Institute space; (ii.) whole-brain structural data were segmented into gray matter, white matter, and cerebrospinal fluid; (iii.) resulting images were checked for sample homogeneity; (iv.) gray matter tissues were modulated by multiplying by the Jacobian determinants generated during spatial normalization to compensate for expansion or contraction due to the nonlinear part of the transformation; and (v.) modulated images were smoothed using a 4-mm full width at half maximum Gaussian kernel. Total intracranial volume was computed as the sum of the extracted total gray and white matter volumes and cerebrospinal fluid for each participant and used to account for overall head size on regional GMV

2.6. Statistical Analyses for Voxel Based Morphometry

Whole-brain statistical analyses were performed on the smoothed gray matter images. Using a voxel-wise probability threshold of p<0.001 and a posteriori family-wise error (FWE) cluster-level correction of p<0.05, AFNI MonteCarlo simulation (3dClustSim) identified a minimum significant cluster size of 267 voxels. An ANCOVA was used to investigate group main effects (F-test) and pairwise contrasts were employed for post hoc analyses while controlling for multiple comparisons at p<0.004 [p<0.05/12 (group comparisons in both directions)]. The Marsbar tool in SPM (http://marsbar.sourceforge.net) was used to extract GMV from clusters of voxels that showed significant between group differences for display purposes and further analyses in SPSS. These analyses were performed to test Hypothesis B.

Whole-brain multiple regression models were used to assess the relationship between GMV and the ERT emotions that demonstrated significant between group differences; corrections for multiple comparisons were applied. Given that VBM subgroups were of modest size, regressions analyses were performed across all participants for increased power; cocaine variables were inspected in CUD participants only. We also tested these models omitting the control group. We also explored associations between extracted GMV and ERT (separately for accuracy and reaction time) using partial correlations in SPSS and correcting for multiple comparisons [p<0.002=p≤0.05/24 (6 emotions × 4 GMV regions). In addition, given that this relationship supports our overall working hypothesis, we conducted whole-brain and region-of-interest regression analyses between duration of cocaine abstinence and GMV. Lastly, we assessed the robustness of these relationships against potential outliers using the percentage-bend correlation function in the Robust Correlation Toolbox (Pernet et al. 2012) in MATLAB. We applied the default bending constant (0.2). Age and total intracranial volume were included as covariates of no interest in all VBM analyses to control for their influence on morphometric outcomes (Barnes et al. 2010; Pell et al. 2008).

We reran both the ANOVA and the regression models using extracted values in SPSS to test whether recent alcohol use (last 30 days) and lifetime years of alcohol use (to intoxication) influenced the VBM findings in cocaine users as suggested in the literature (Alia-Klein et al. 2011; Rando et al. 2013); negative results are reported in the Supplementary Material.

2.6.1. Mediation Analysis

An exploratory mediation analysis was carried out to study the relationship between GMV (independent variable) and social function (dependent variable) with emotion recognition performance as the mediating variable; age and total intracranial volume were included as covariates. These analyses were conducted using the SPSS macro PROCESS (Preacher and Hayes 2008) to obtain estimates of the indirect mediation effect using 5000 bootstrap samples and 95% confidence intervals. Mediation was tested in the full sample and within the CUD group.

3. RESULTS

3.1. Demographic and Substance Use Characteristics

Data were screened for outliers (defined as deviating ≥3 standard deviations from the mean) and normality on the ERT. The cognitive dataset of N=116 was derived from an original set of N=123 after removing seven outliers (CUD+, n=2; CUD-ST, n=3; CUD-LT, n=1; CTL, n=1). These participants were excluded from all ERT analyses; however, participants who had MRI data were retained for the VBM analyses (n=4). See Table 1 for descriptive characteristics.

The CUD+ group had the shortest duration of cocaine abstinence compared to both CUD− subgroups, although given the high variability, differences in abstinence only reached significance when comparing CUD+ and CUD-ST to CUD-LT. Nevertheless, CUD+ participants differed from both CUD− subgroups in frequency of recent cocaine use (past 30 days), and withdrawal and craving (CUD+>CUD−; craving also differed between CUD-ST>CUD-LT). Compared to the CUD-LT, CUD+ had greater duration of cocaine use. With respect to alcohol consumption, CUD+ and CUD-LT had greater years of use compared to healthy controls. However, CUD+ consumed more alcoholic beverages in the previous 30 days compared to the three other subgroups. Sleep quantity and sleep quality did not differ between the groups (see Table 1 and Supplementary Material). Correlations conducted between the 11 variables that showed significant group differences (Table 1) and ERT showed no significant associations (at the a priori p<0.05 and of a correlation coefficient r>0.40). Of note, this included duration of lifetime cocaine use (p>0.1). Therefore, no covariates were used in the subsequent analyses.

3.2. Emotion Recognition Performance

3.2.1. ERT: Percent Correct

A repeated-measures ANOVA revealed significant main effects for emotion [happiness>surprise>sadness>disgust=anger>fear; F (4.15, 464.9)=150.51, p<0.001, η2 =0.57] and group [CUD+<CTL and CUD+<CUD-ST; F (3,112)=10.01, p<0.001, η2 =0.21] and the interaction between emotion and group [F (12.43, 464.9)=2.69, p=0.016, η2 =0.07]. Post-hoc tests revealed that the interaction effect was driven by significant group differences for happiness (CUD+<CTL: p<0.001; and CUD+<CUD-ST: p=0.0011), sadness (CUD+<CTL: p<0.001), and fear (CUD+<CTL: p<0.001; and CUD-ST<CTL: p<0.001) (Table 2; Figure 1). No group differences emerged for anger, disgust, or surprise.

Table 2.

Group Differences in Emotion Recognition and Social Function Outcomes

N=116 CUD+ (n=28) CUD-ST (n=23) CUD-LT (n=20) CTL (n=45) p-value
ERT Accuracy (percent correct)
Happiness** 69.79 (19.5)a,c 83.48 (8.6)a,b 76.00 (16.5) 82.81 (12.6)b 0.001
Sadness** 45.00 (21.0)a 56.67 (16.2) 54.00 (15.4) 60.96 (12.4)b 0.001
Anger 35.11 (15.6) 46.67 (16.2) 41.83 (14.8) 48.29 (15.9) 0.006
Disgust 35.59 (21.6) 47.83 (22.4) 43.83 (21.7) 56.22 (23.0) 0.010
Fear** 22.73 (17.7)a 23.19 (16.4)a 28.67 (16.4) 41.63 (19.6)b,c <0.001
Surprise 62.14 (13.6) 73.04 (11.9) 66.50 (19.8) 63.77 (14.0) 0.047
ERT Reaction Time
Happiness 1414.77 (632.3) 1751.70 (1149.6) 1262.99 (374.1) 1347.13 (566.5) 0.100
Sadness 1843.28 (765.4) 1914.41 (677.6) 1964.55 (680.9) 1845.79 (767.2) 0.923
Anger 2144.73 (1350.2) 2133.97 (956.1) 2476.57 (2571.7) 2190.01 (1511.0) 0.867
Disgust 1780.03 (706.8) 2153.85 (1156.1) 1979.92 (1044.7) 1856.76 (878.9) 0.504
Fear 1788.29 (704.1) 1800.36 (791.3) 1859.20 (723.8) 2104.24 (1231.00 0.460
Surprise 1682.41 (667.2) 1592.66 (664.1) 1575.14 (482.3) 1676.88 (731.4) 0.907
ASI Outcomes
Interviewer Severity Score† (0/1/2/3/4) 20/2/1/3/2 16/5/2/0/0 19/0/1/0/0 43/2/0/0/0 0.004
Composite Score 0.07±0.1 0.08±0.1 0.1±0.2 0.07±0.09 0.817
Free time (alone/family/friends) 13/7/8 14/5/4 7/6/7 8/17/20 0.028
# of Friends 3.4±3.5 3.9±3.1 3.7±3.7 4.2±3.0 0.395
Open in a new tab
a

, mean differs from CTL;

b

, mean differs from CUD+;

c

, mean differs from CUD-ST;

d

, mean differs from CUD-LT

**

p<0.004;

*

p<0.0125;

for ERT, p-value denotes significance for follow-up one-way ANOVAs for the group x emotion interaction for emotion; for ASI outcomes, p-values reflects significance from Kruskal-Wallis tests;

†

, post-hoc groups difference showed a trend towards significance.

ASI, Addiction Severity Index; CTL, control; CUD+, participants testing positive for cocaine; CUD-ST, participants testing negative for cocaine and with <6-months abstinence; CUD-LT, participants testing negative for cocaine and with >6-months abstinence; ERT, emotion recognition task.

Figure 1.

Figure 1.

Open in a new tab

Group Main Effects on ERT Performance and Gray Matter Volume. Compared to CTL, CUD+ had poorer performance for recognizing (A) happiness, (B) sadness, and (C), fear. For fear, CUD-ST also showed worse performance compared to CTL (C). Error bars represent standard deviations. Significant between group differences in GMV were present bilaterally in the cerebellum and amygdala (D). More specifically, CUD+ had reduced GMV in both the bilateral cerebellum and bilateral amygdala (pFWE-corr≤0.001) compared to CUD-LT, and reduced bilateral cerebellum compared to CTL (pFWE-corr=0.003). F-maps are displayed using a voxel-wise FWE-corrected threshold of p<0.001. The color bar represents the corresponding F-score. Figure 1E shows the corresponding group differences in GMV in the bilateral cerebellum and bilateral amygdala. Error bars represent standard deviations.

3.2.2. ERT: Reaction Time

A repeated-measures ANOVA revealed a significant main effect for emotion [happiness<sadness>surprise<disgust=anger=fear; F (2.98, 334.19)=14.44, p<0.001, η2 =0.11], but not for group [F (3, 112)=0.12, p=0.946, η2 =0.003] or the interaction between emotion and group [F (8.95, 334.19)=1.08, p=0.377, η2 =0.03] (Table 2). See Supplementary Material for post-hoc test results for main effects for ERT accuracy and reaction time.

3.3. Reports of Social Function

Exploratory analyses revealed significant group differences for the interviewer severity score [χ2(12)= 29.33, p=0.004, Cramer’s V=0.29]. Planned post-hoc tests revealed that CUD+ and CUD-ST had greater severity ratings compared to controls at trend levels, p=0.03 and p=0.009, respectively. There were no group differences in the composite score [H(3)=0.23, p=0.97, η2=0.002] or for reported number of friends [H(3)=4.61, p=0.20, η2=0.04]. However, a trend emerged also for how free time was spent (p=0.028, Cramer’s V=0.25), driven by reports of more time spent alone by the CUD-ST compared to the CTL group.

3.4. Voxel Based Morphometry

3.4.1. Subsample Characteristics

See Supplementary Material for demographic, clinical and substance-using characteristics of the VBM subsample (N=73) (Supplementary Table 1; these variables did not differ substantially from the main cohort).

3.4.2. Between-group Differences

Whole-brain analyses revealed significant group differences in the bilateral cerebellum (pFWE<0.001) and bilateral amygdala (pFWE<0.01). Post-hoc pairwise comparisons revealed that compared to CTL and CUD-LT, CUD+ had smaller bilateral GMV in the cerebellum (pFWE<0.001). In addition, compared to CUD-LT, CUD+ demonstrated lower GMV in the bilateral amygdala (pFWE≤0.001) (Table 3 and Figure 1). Correlations conducted between the demographic and clinical variables that showed between group differences (Supplementary Table 1) and GMV of these 4 regions of interest (bilateral cerebellum and bilateral amygdala) rendered no significant relationships that survived multiple comparisons (r<0.42, p>0.003), with the exception of duration of cocaine abstinence, see section 3.4.5. Therefore, no additional covariates were used. Similarly, there were no significant relationships between duration of lifetime cocaine use and GMV in the bilateral cerebellum and bilateral amygdala (p>0.2).

Table 3.

Brain Regions with Significant Between Group Differences in Gray Matter Volumes

Anatomical Label Hemisphere Coordinates (x,y,z) Z-score Cluster Size p FWE-corr
Group Main Effects
Cerebellum Right 36, −54, −53 4.26 966 <0.001
Cerebellum Left −36, −48, −57 4.55 622 <0.001
Amygdala Right 17, −2, −20 4.20 376 <0.001
Amygdala Left −17, 0, −21 3.93 279 0.008
Post-hoc between-group comparison
CUD+<CUD-LT
Cerebellum Right 27, −66, −56 4.80 1998 <0.001
Cerebellum Left −36, −48, −57 4.78 2097 <0.001
Amygdala Right 17, −2, −20 4.84 768 <0.001
Amygdala Left −17, −2, −20 4.51 527 <0.001
CUD+<CTL
Cerebellum Right 35, −66, −54 4.28 628 <0.001
Cerebellum Left −35, −48, −57 4.41 689 <0.001
Open in a new tab

x, y and z coordinates given in MNI space (Montreal Neurological Institute).

No other group comparisons were significant (CUD+ vs. CUD-ST; CUD-ST vs. CUD-LT; CUD-ST vs. CTL; CUD-LT vs. CTL).

3.4.3. Relationships between ERT Performance and Gray Matter Volume

Whole-brain analyses revealed a significant positive association between ERT percent correct for happiness (but not sadness and fear) and GMV in the right cerebellum (x=38, y=−53, z=−59, Z=3.85, k=497, pFWE=0.002) (Figure 2) and a similar trend for the left cerebellum (x=−30, y=−51, z=−51, Z=3.80, k=358, pFWE<0.07). The robust correlation method confirmed the significant relationships between ERT percent correct for happiness and GMV in the right cerebellum [Bend r=0.25, p=0.03, 95% CI = (0.01, 0.46)] and GMV in the left cerebellum [Bend r=0.33, p=0.006, 95% CI = (0.09, 0.53)]. Removing controls, a whole-brain regression between ERT percent correct for happiness and GMV did not reach significance; however, a correlation between ERT percent correct for happiness and extracted GMV in the right cerebellum (r=0.30, p=0.06) and the left cerebellum (r=0.34, p=0.03) were significant at a trend level. Using a whole-brain approach, no significant associations emerged between ERT accuracy for any emotion and GMV in the amygdala. Similarly, no significant correlations emerged with extracted amygdala GMV and ERT accuracy for any emotion (r<0.09, p>0.31). Given the lack of significant group effects for ERT reaction time, correlations between GMV and ERT reaction time were not inspected.

Figure 2.

Figure 2.

Open in a new tab

Relationship between Emotion Recognition and Right Cerebellar GMV (N=69). There was a positive relationship between percent correct for happiness on the ERT and GMV in the right cerebellum (pFWE-corr=0.002), but nor for sadness or fear. Values for GMV are presented as standardized residuals to correct for age and TIV. The structural images (on the left) show the right cerebellum in an axial, coronal and sagittal view. T-maps are displayed using a voxel-wise FWE-corrected threshold of p<0.001. The color bar represents the corresponding T-scores.

3.4.5. Relationships between Duration of Cocaine Abstinence and Gray Matter Volume

Whole-brain analyses revealed significant positive associations between duration of cocaine abstinence and GMV in the bilateral cerebellum (right: x=48, y=−68, z=−47, Z=4.54, k=632, pFWE=0.005; left: x=−45, y=−68, z=−47, Z=4.70, k=774, pFWE=0.001) and right amygdala (x=17, y=−2, z=−14, Z=4.69, k=401, pFWE=0.03) as demonstrated in Figure 3. The robust correlation method confirmed the significant relationships between duration of cocaine abstinence and GMV in the right cerebellum [Bend r=0.64, p<0.01, 95% CI = (0.39, 0.78)], left cerebellum [Bend r=0.61, p<0.01, 95% CI = (0.39, 0.76)], and right amygdala [Bend r=0.60, p<0.01, 95% CI = (0.39, 0.75)].

Figure 3.

Figure 3.

Open in a new tab

Relationship between Duration of Cocaine Abstinence and GMV (N=46).

Across CUD participants, greater duration of cocaine abstinence was associated with greater GMV in the bilateral cerebellum (pFWE=0.001) and right amygdala (pFWE=0.03). Values for GMV are presented as standardized residuals to correct for age and TIV. Values for days abstinent are transformed using a square root function. The color bar represents the corresponding T-scores.

Results did not change when the ANOVA and the regression models were performed with recent alcohol use (last 30 days) and lifetime years of alcohol use (to intoxication) as covariates of no interest; see Supplementary Material.

3.5. Mediation Analysis

As an exploratory analysis, we tested whether emotion recognition for happiness mediated the relationship between GMV of the right and/or left cerebellum and social function (using the ASI composite score). Results were not significant for the model that included the right cerebellum [β=−0.002, 95% CI= (−0.82, 0.83)] or the left cerebellum [β=−0.006, 95% CI= (−0.12, 0.09)]. Note that using the interviewer severity ratings as a continuous measure to index social function produced similar results. Models were re-tested in CUD only, and results still did not reach significance. In addition, we tested the potential mediating role of abstinence length; here relationships were tested for a potential mediating effect between GMV and social function, GMV and emotion recognition, and emotion recognition and social function. None of these models reached significance.

4. DISCUSSION

The primary goal of this study was to determine if emotion recognition performance was associated with regional brain volumes of the social brain network in individuals with CUD parsed according to length of cocaine abstinence. Partially in line with our hypothesis (Hypothesis A), and despite similar group latencies to respond to emotional faces, CUD+ demonstrated greater difficulty recognizing happiness compared to CUD-ST and healthy controls and greater difficulty recognizing negative emotions (sadness and fear) compared to healthy controls; CUD-ST also showed deficits in recognizing fear as compared to healthy controls. Overall, this is the first study to report deficits in positive emotion recognition in active CUD users. The results in the CUD-ST group suggest that short-term abstinence (<6-months) may be associated with temporary normative performance for happiness recognition that then plateaus (a conclusion that should be interpreted with caution due to the cross-sectional nature of this study and because differences were only significant when compared to CUD+, but not CUD-LT or CTL). Importantly, results were not simply driven by group differences in craving/withdrawal symptoms or lifetime cocaine use. In line with our results for negative emotion recognition, Kuypers et al (2015) observed that cocaine administered to recreational cocaine users impaired sadness recognition and others showed that regular cocaine users who used cocaine within the previous 5 days had difficulties with fear recognition (Ersche et al. 2015; Kemmis et al. 2007; Morgan and Marshall 2013). Of note, there were no group differences between CUD+ and CUD-LT in ERT accuracy, suggesting that emotion recognition may not improve with protracted abstinence, an interpretation supported by the lack of association between abstinence duration and ERT outcomes. Taken together, this pattern of results for emotion recognition accuracy is consistent with a working hypothesis derived from our previous studies, and with the self-medication hypothesis, whereby current cocaine use in individuals with CUD may enhance cognitive functions (e.g., memory) (Woicik et al. 2009) at the expense of impaired emotional processing (Parvaz et al. 2015).

Our findings also provide insight into the neurobiological underpinnings associated with abnormal emotion recognition and cocaine abstinence in CUD. Compared to the other groups (CUD-LT and controls, or CUD-LT only, respectively), CUD+ had lower GMV in the bilateral cerebellum and amygdala, providing partial support for Hypothesis B; of note, results were not driven by group differences in duration of lifetime cocaine use. Other studies have similarly reported an association between GMV in the cerebellum and the amygdala and cocaine use (Makris et al. 2004; O’Neill et al. 2001; Sim et al. 2007), with more pronounced cerebellar deficits in current cocaine users compared to abstainers (mean abstinence=244 days) (Hanlon et al. 2011). Interestingly, in our sample the cerebellar findings were localized to the posterior lobe, a region coined the “limbic cerebellum” signifying its role in higher-level cognitive-emotional processing (Schmahmann et al. 2007; Stoodley and Schmahmann 2010; Van Overwalle et al. 2014). It has been suggested that the posterior cerebellum is involved in mediating the perception of others’ emotional states (Ferrari et al. 2019) and that its damage contributes to emotional processing impairments (Stoodley and Schmahmann 2010), specifically in emotion recognition (D’Agata et al. 2011). Accordingly, and beyond group effects, we showed that GMV of the bilateral posterior cerebellum increased linearly with accuracy for happiness recognition, supporting earlier findings in psychopathy patients (Pera-Guardiola et al. 2016). Further support for the relationship between emotional processing and the structural integrity of the cerebellum derives from a study conducted in healthy controls that found a negative association between white matter volume of the cerebellum and the ability to recognize facial expressions (Uono et al. 2017). Of note, the relationships between emotion recognition and GMV may, in part, be driven by group differences in the latter (highest in controls and CUD-LT and lowest in CUD+). Nevertheless, we included the healthy control subjects to increase power and to show a pattern of results that transcended all our subjects; results remained largely the same when controls were excluded from these analyses.

We also found that longer periods of abstinence in CUD were associated with greater cerebellar GMV, suggesting that with prolonged sobriety these GMV deficits may normalize, which may translate into improved positive emotion recognition. Similar to the cerebellum, greater abstinence duration was also associated with larger amygdala GMV. Recovery of GMV may reflect metabolic increases in choline containing compounds and N-acetylaspartate as observed in alcohol abstainers; changes that were also associated with improved cognition (Bendszus et al. 2001; Parks et al. 2002). Alternatively, it is plausible that increased cerebellar and/or amygdala GMV in CUD is associated with a greater likelihood of achieving abstinence, which may underpin intact cognitive and emotional control (Connolly et al. 2012; Wrase et al. 2008). Nevertheless, in our study abstinence length did not mediate the relationship between GMV and emotion recognition, potentially because of the cross-sectional nature of this study.

Exploratory social functioning analyses revealed a significant difference in the interviewer severity score driven by greater dysfunction in both the CUD+ and CUD-ST groups compared to healthy controls; there was also a trend for the CUD-ST group to spend more time alone. This result is consistent with accumulating evidence highlighting social difficulties in individuals with current substance use disorders (Kornreich et al. 2002; Poudel and Gautam 2017; Volkow et al. 2011), specifically cocaine (Bedi et al. 2019; Pachado et al. 2018). Notably, emotion recognition did not mediate the relationship between GMV of the social brain network (specifically, the cerebellum) and social function in our sample. Similarly, and despite its associations with cerebellar and amygdala GMV across all individuals with CUD, abstinence length did not mediate the relationship between GMV and social function.

Our findings should be interpreted in light of several limitations. First, the ERT was limited to the presentation of static emotions. Future research should use dynamic stimuli that may more closely mirror real-life facial expressions (Ambadar et al. 2005) and incorporate more positive emotions (e.g., amusement, excitement) to determine whether cocaine’s effects generalize to other positive emotions. Second, the inclusion of a more detailed social functioning scale, with greater sensitivity or more objective real-life assessment, may allow for better characterization of the relationship between social function, emotion recognition, and GMV. Third, while the number of cigarettes did not differ between groups among current smokers, we did not account for differences in the distribution of smokers and its potential influence on outcomes. Fourth, clinical variables not assessed in this study, such as stress or anxiety, may have contributed to emotion recognition deficits and/or lower GMV in CUD+. In addition, the inclusion of a more comprehensive sleep questionnaire as well as objective indices of sleep (e.g., actigraphy) may contribute to a better understanding of the impact of sleep on social cognition and GMV during abstinence in individuals with CUD. Given the prevalence of sleep difficulties in CUD, future research should examine the role of the glymphatic system and potential relationships with GMV, especially in the cerebellum where, in humans, large changes in sleep-driven diffusion have been observed (Demiral et al. 2019). Fifth, ERT administration and MRI were not conducted on the same study visit (however, there was no significant relationships between the difference in days from ERT to MRI and ERT outcomes, suggesting this variable did not contribute significantly to results). Sixth, results remain to be replicated in a larger sample (particularly with respect to the GMV findings) and using longitudinal (rather than cross-sectional) designs. Of note, the latter will help to determine if more intact emotion recognition performance and GMV play a role in the ability to achieve and maintain abstinence (survivorship bias). Lastly, future studies should incorporate functional imaging techniques, in combination with social cognitive testing and VBM, to assess the association between neural activity in the social brain network and GMV. Our findings highlight the need for longitudinal, within-subject, studies to investigate the temporal relationship between duration of cocaine abstinence, emotion recognition, social function, and GMV in CUD.

4.2. Conclusions

Our findings demonstrate emotion recognition impairments (happiness, sadness, and fear) in currently using individuals with CUD and suggest that selective deficits (in fear) may persist with up to 6-months of abstinence. Reduced cerebellar GMV may underlie deficits in positive emotion recognition in CUD. Encouraging long-term abstinence may directly impact GMV recovery and rescue social-cognitive impairments. A deeper understanding of emotion recognition, social function, and their neural correlates may help to elucidate novel therapeutic targets to enhance current treatment approaches for individuals suffering from CUD.

Supplementary Material

1762896_Supinfo

Acknowledgements:

This work was supported by a T32 DA007135 to RA Rabin and a K01DA043615 to MA Parvaz; RZ Goldstein and N Alia-Klein were supported by grants from the National Institutes of Health (R01DA041528, R01DA047851). MA Parvaz, N Alia-Klein and RZ Goldstein were also supported by internal seed funds from the Icahn School of Medicine at Mount Sinai. The authors have nothing to disclose.

References

  1. Adolphs R (2009) The social brain: neural basis of social knowledge. Annu Rev Psychol 60: 693–716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Adolphs R, Tranel D, Damasio H, Damasio A (1994) Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala. Nature 372: 669–72. [DOI] [PubMed] [Google Scholar]
  3. Alia-Klein N, Parvaz MA, Woicik PA, Konova AB, Maloney T, Shumay E, Wang R, Telang F, Biegon A, Wang GJ, Fowler JS, Tomasi D, Volkow ND, Goldstein RZ (2011) Gene x disease interaction on orbitofrontal gray matter in cocaine addiction. Arch Gen Psychiatry 68: 283–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ambadar Z, Schooler JW, Cohn JF (2005) Deciphering the enigmatic face: the importance of facial dynamics in interpreting subtle facial expressions. Psychol Sci 16: 403–10. [DOI] [PubMed] [Google Scholar]
  5. American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders (5th edition) Author, Washington, DC [Google Scholar]
  6. Bachi K, Parvaz MA, Moeller SJ, Gan G, Zilverstand A, Goldstein RZ, Alia-Klein N (2018) Reduced Orbitofrontal Gray Matter Concentration as a Marker of Premorbid Childhood Trauma in Cocaine Use Disorder. Front Hum Neurosci 12: 51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Baillieux H, De Smet HJ, Paquier PF, De Deyn PP, Marien P (2008) Cerebellar neurocognition: insights into the bottom of the brain. Clin Neurol Neurosurg 110: 763–73. [DOI] [PubMed] [Google Scholar]
  8. Barnes J, Ridgway GR, Bartlett J, Henley SM, Lehmann M, Hobbs N, Clarkson MJ, MacManus DG, Ourselin S, Fox NC (2010) Head size, age and gender adjustment in MRI studies: a necessary nuisance? Neuroimage 53: 1244–55. [DOI] [PubMed] [Google Scholar]
  9. Basterfield C, Hester R, Bowden SC (2019) A meta-analysis of the relationship between abstinence and neuropsychological functioning in methamphetamine use disorder. Neuropsychology 33: 739–753. [DOI] [PubMed] [Google Scholar]
  10. Bathish R, Best D, Savic M, Beckwith M, Mackenzie J, Lubman DI (2017) “Is it me or should my friends take the credit?” The role of social networks and social identity in recovery from addiction. Journal of Applied Social Psychology 47: 35–46. [Google Scholar]
  11. Beck ATS RA; Brown GK (1996) Manual for the Beck Depression Inventory-II. Psychological Corporation, San Antonio, TX [Google Scholar]
  12. Bedi G, Hao X, Van Dam NT, Cooper ZD, Rubin E, Vadhan NP, Marino L, Haney M (2019) Social motivational processing and interpersonal function in aging cocaine smokers. Addict Biol 24: 1044–1055. [DOI] [PubMed] [Google Scholar]
  13. Bendszus M, Weijers HG, Wiesbeck G, Warmuth-Metz M, Bartsch AJ, Engels S, Boning J, Solymosi L (2001) Sequential MR imaging and proton MR spectroscopy in patients who underwent recent detoxification for chronic alcoholism: correlation with clinical and neuropsychological data. AJNR Am J Neuroradiol 22: 1926–32. [PMC free article] [PubMed] [Google Scholar]
  14. Bickart KC, Wright CI, Dautoff RJ, Dickerson BC, Barrett LF (2011) Amygdala volume and social network size in humans. Nat Neurosci 14: 163–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Campbell AN, Nunes EV, Miele GM, Matthews A, Polsky D, Ghitza UE, Turrigiano E, Bailey GL, VanVeldhuisen P, Chapdelaine R, Froias A, Stitzer ML, Carroll KM, Winhusen T, Clingerman S, Perez L, McClure E, Goldman B, Crowell AR (2012) Design and methodological considerations of an effectiveness trial of a computer-assisted intervention: an example from the NIDA Clinical Trials Network. Contemp Clin Trials 33: 386–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Castellano F, Bartoli F, Crocamo C, Gamba G, Tremolada M, Santambrogio J, Clerici M, Carra G (2015) Facial emotion recognition in alcohol and substance use disorders: A meta-analysis. Neurosci Biobehav Rev 59: 147–54. [DOI] [PubMed] [Google Scholar]
  17. Choudhury S, Blakemore SJ, Charman T (2006) Social cognitive development during adolescence. Soc Cogn Affect Neurosci 1: 165–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Connolly CG, Bell RP, Foxe JJ, Garavan H (2013) Dissociated grey matter changes with prolonged addiction and extended abstinence in cocaine users. PLoS One 8: e59645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Connolly CG, Foxe JJ, Nierenberg J, Shpaner M, Garavan H (2012) The neurobiology of cognitive control in successful cocaine abstinence. Drug Alcohol Depend 121: 45–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Cooper GA, Kronstrand R, Kintz P, Society of Hair T (2012) Society of Hair Testing guidelines for drug testing in hair. Forensic Sci Int 218: 20–4. [DOI] [PubMed] [Google Scholar]
  21. D’Agata F, Caroppo P, Baudino B, Caglio M, Croce M, Bergui M, Tamietto M, Mortara P, Orsi L (2011) The recognition of facial emotions in spinocerebellar ataxia patients. Cerebellum 10: 600–10. [DOI] [PubMed] [Google Scholar]
  22. Dackis CA, Gold MS (1985) New concepts in cocaine addiction: the dopamine depletion hypothesis. Neurosci Biobehav Rev 9: 469–77. [DOI] [PubMed] [Google Scholar]
  23. Dal Monte O, Krueger F, Solomon JM, Schintu S, Knutson KM, Strenziok M, Pardini M, Leopold A, Raymont V, Grafman J (2013) A voxel-based lesion study on facial emotion recognition after penetrating brain injury. Soc Cogn Affect Neurosci 8: 632–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Demiral SB, Tomasi D, Sarlls J, Lee H, Wiers CE, Zehra A, Srivastava T, Ke K, Shokri-Kojori E, Freeman CR, Lindgren E, Ramirez V, Miller G, Bandettini P, Horovitz S, Wang GJ, Benveniste H, Volkow ND (2019) Apparent diffusion coefficient changes in human brain during sleep - Does it inform on the existence of a glymphatic system? Neuroimage 185: 263–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Ekman P (1972) Universals and cultural differences in facial expressions of emotion, Lincoln: University of Nebraska Press [Google Scholar]
  26. Ekman P, Friesen WV, O’Sullivan M, Chan A, Diacoyanni-Tarlatzis I, Heider K, Krause R, LeCompte WA, Pitcairn T, Ricci-Bitti PE, et al. (1987) Universals and cultural differences in the judgments of facial expressions of emotion. J Pers Soc Psychol 53: 712–7. [DOI] [PubMed] [Google Scholar]
  27. Ersche KD, Barnes A, Jones PS, Morein-Zamir S, Robbins TW, Bullmore ET (2011) Abnormal structure of frontostriatal brain systems is associated with aspects of impulsivity and compulsivity in cocaine dependence. Brain 134: 2013–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Ersche KD, Hagan CC, Smith DG, Jones PS, Calder AJ, Williams GB (2015) In the face of threat: neural and endocrine correlates of impaired facial emotion recognition in cocaine dependence. Transl Psychiatry 5: e570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ferrari C, Ciricugno A, Urgesi C, Cattaneo Z (2019) Cerebellar contribution to emotional body language perception: a TMS study. Soc Cogn Affect Neurosci. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Fox HC, Bergquist KL, Casey J, Hong KA, Sinha R (2011) Selective cocaine-related difficulties in emotional intelligence: relationship to stress and impulse control. Am J Addict 20: 151–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Gan G, Zilverstand A, Parvaz MA, Preston-Campbell RN, d’Oleire Uquillas F, Moeller SJ, Tomasi D, Goldstein RZ, Alia-Klein N (2019) Habenula-prefrontal resting-state connectivity in reactive aggressive men - A pilot study. Neuropharmacology 156: 107396. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Gawin FH (1991) Cocaine addiction: psychology and neurophysiology. Science 251: 1580–6. [DOI] [PubMed] [Google Scholar]
  33. Hanlon CA, Dufault DL, Wesley MJ, Porrino LJ (2011) Elevated gray and white matter densities in cocaine abstainers compared to current users. Psychopharmacology (Berl) 218: 681–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hirsiger S, Hanggi J, Germann J, Vonmoos M, Preller KH, Engeli EJE, Kirschner M, Reinhard C, Hulka LM, Baumgartner MR, Chakravarty MM, Seifritz E, Herdener M, Quednow BB (2019) Longitudinal changes in cocaine intake and cognition are linked to cortical thickness adaptations in cocaine users. Neuroimage Clin 21: 101652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Hoff AL, Riordan H, Morris L, Cestaro V, Wieneke M, Alpert R, Wang GJ, Volkow N (1996) Effects of crack cocaine on neurocognitive function. Psychiatry Res 60: 167–76. [DOI] [PubMed] [Google Scholar]
  36. Johnson MH, Griffin R, Csibra G, Halit H, Farroni T, de Haan M, Tucker LA, Baron-Cohen S, Richards J (2005) The emergence of the social brain network: evidence from typical and atypical development. Dev Psychopathol 17: 599–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Jovanovski D, Erb S, Zakzanis KK (2005) Neurocognitive deficits in cocaine users: a quantitative review of the evidence. J Clin Exp Neuropsychol 27: 189–204. [DOI] [PubMed] [Google Scholar]
  38. Kaag AM, Schulte MHJ, Jansen JM, van Wingen G, Homberg J, van den Brink W, Wiers RW, Schmaal L, Goudriaan AE, Reneman L (2018) The relation between gray matter volume and the use of alcohol, tobacco, cocaine and cannabis in male polysubstance users. Drug Alcohol Depend 187: 186–194. [DOI] [PubMed] [Google Scholar]
  39. Kalivas PW (2007) Neurobiology of cocaine addiction: implications for new pharmacotherapy. Am J Addict 16: 71–8. [DOI] [PubMed] [Google Scholar]
  40. Kampman KM, Volpicelli JR, McGinnis DE, Alterman AI, Weinrieb RM, D’Angelo L, Epperson LE (1998) Reliability and validity of the Cocaine Selective Severity Assessment. Addict Behav 23: 449–61. [DOI] [PubMed] [Google Scholar]
  41. Kemmis L, Hall JK, Kingston R, Morgan MJ (2007) Impaired fear recognition in regular recreational cocaine users. Psychopharmacology (Berl) 194: 151–9. [DOI] [PubMed] [Google Scholar]
  42. Kim YT, Kwon DH, Chang Y (2011) Impairments of facial emotion recognition and theory of mind in methamphetamine abusers. Psychiatry Res 186: 80–4. [DOI] [PubMed] [Google Scholar]
  43. Kipps CM, Duggins AJ, McCusker EA, Calder AJ (2007) Disgust and happiness recognition correlate with anteroventral insula and amygdala volume respectively in preclinical Huntington’s disease. J Cogn Neurosci 19: 1206–17. [DOI] [PubMed] [Google Scholar]
  44. Kornreich C, Philippot P, Foisy ML, Blairy S, Raynaud E, Dan B, Hess U, Noel X, Pelc I, Verbanck P (2002) Impaired emotional facial expression recognition is associated with interpersonal problems in alcoholism. Alcohol Alcohol 37: 394–400. [DOI] [PubMed] [Google Scholar]
  45. Kuypers KP, Steenbergen L, Theunissen EL, Toennes SW, Ramaekers JG (2015) Emotion recognition during cocaine intoxication. Eur Neuropsychopharmacol 25: 1914–21. [DOI] [PubMed] [Google Scholar]
  46. LeDoux JE (2000) Emotion circuits in the brain. Annu Rev Neurosci 23: 155–84. [DOI] [PubMed] [Google Scholar]
  47. Lotze M, Domin M, Gerlach FH, Gaser C, Lueders E, Schmidt CO, Neumann N (2019) Novel findings from 2,838 Adult Brains on Sex Differences in Gray Matter Brain Volume. Sci Rep 9: 1671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Makris N, Gasic GP, Seidman LJ, Goldstein JM, Gastfriend DR, Elman I, Albaugh MD, Hodge SM, Ziegler DA, Sheahan FS, Caviness VS Jr., Tsuang MT, Kennedy DN, Hyman SE, Rosen BR, Breiter HC (2004) Decreased absolute amygdala volume in cocaine addicts. Neuron 44: 729–40. [DOI] [PubMed] [Google Scholar]
  49. McLellan AT, Kushner H, Metzger D, Peters R, Smith I, Grissom G, Pettinati H, Argeriou M (1992) The Fifth Edition of the Addiction Severity Index. J Subst Abuse Treat 9: 199–213. [DOI] [PubMed] [Google Scholar]
  50. Moeller SJ, Kundu P, Bachi K, Maloney T, Malaker P, Parvaz MA, Alia-Klein N, London ED, Goldstein RZ (2020) Self-awareness of problematic drug use: Preliminary validation of a new fMRI task to assess underlying neurocircuitry. Drug Alcohol Depend 209: 107930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Moeller SJ, Zilverstand A, Konova AB, Kundu P, Parvaz MA, Preston-Campbell R, Bachi K, Alia-Klein N, Goldstein RZ (2018) Neural Correlates of Drug-Biased Choice in Currently Using and Abstinent Individuals With Cocaine Use Disorder. Biol Psychiatry Cogn Neurosci Neuroimaging 3: 485–494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Morgan MJ, Marshall JP (2013) Deficient fear recognition in regular cocaine users is not attributable to elevated impulsivity or conduct disorder prior to cocaine use. J Psychopharmacol 27: 526–32. [DOI] [PubMed] [Google Scholar]
  53. Noonan MP, Mars RB, Sallet J, Dunbar RIM, Fellows LK (2018) The structural and functional brain networks that support human social networks. Behav Brain Res 355: 12–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. O’Neill J, Cardenas VA, Meyerhoff DJ (2001) Separate and interactive effects of cocaine and alcohol dependence on brain structures and metabolites: quantitative MRI and proton MR spectroscopic imaging. Addict Biol 6: 347–361. [DOI] [PubMed] [Google Scholar]
  55. Pace-Schott EF, Morgan PT, Malison RT, Hart CL, Edgar C, Walker M, Stickgold R (2008) Cocaine users differ from normals on cognitive tasks which show poorer performance during drug abstinence. Am J Drug Alcohol Abuse 34: 109–21. [DOI] [PubMed] [Google Scholar]
  56. Pachado MP, Scherer JN, Guimaraes LSP, von Diemen L, Pechansky F, Kessler FHP, de Almeida RMM (2018) Markers for Severity of Problems in Interpersonal Relationships of Crack Cocaine Users from a Brazilian Multicenter Study. Psychiatr Q 89: 923–936. [DOI] [PubMed] [Google Scholar]
  57. Parks MH, Dawant BM, Riddle WR, Hartmann SL, Dietrich MS, Nickel MK, Price RR, Martin PR (2002) Longitudinal brain metabolic characterization of chronic alcoholics with proton magnetic resonance spectroscopy. Alcohol Clin Exp Res 26: 1368–80. [DOI] [PubMed] [Google Scholar]
  58. Parvaz MA, Konova AB, Proudfit GH, Dunning JP, Malaker P, Moeller SJ, Maloney T, Alia-Klein N, Goldstein RZ (2015) Impaired neural response to negative prediction errors in cocaine addiction. J Neurosci 35: 1872–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Parvaz MA, Moeller SJ, d’Oleire Uquillas F, Pflumm A, Maloney T, Alia-Klein N, Goldstein RZ (2017) Prefrontal gray matter volume recovery in treatment-seeking cocaine-addicted individuals: a longitudinal study. Addict Biol 22: 1391–1401. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Parvaz MA, Moeller SJ, Goldstein RZ (2016) Incubation of Cue-Induced Craving in Adults Addicted to Cocaine Measured by Electroencephalography. JAMA Psychiatry 73: 1127–1134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Payer DE, Lieberman MD, Monterosso JR, Xu J, Fong TW, London ED (2008) Differences in cortical activity between methamphetamine-dependent and healthy individuals performing a facial affect matching task. Drug Alcohol Depend 93: 93–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Pell GS, Briellmann RS, Chan CH, Pardoe H, Abbott DF, Jackson GD (2008) Selection of the control group for VBM analysis: influence of covariates, matching and sample size. Neuroimage 41: 1324–35. [DOI] [PubMed] [Google Scholar]
  63. Pera-Guardiola V, Contreras-Rodriguez O, Batalla I, Kosson D, Menchon JM, Pifarre J, Bosque J, Cardoner N, Soriano-Mas C (2016) Brain Structural Correlates of Emotion Recognition in Psychopaths. PLoS One 11: e0149807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Pernet CR, Wilcox R, Rousselet GA (2012) Robust correlation analyses: false positive and power validation using a new open source matlab toolbox. Front Psychol 3: 606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Phillips ML, Drevets WC, Rauch SL, Lane R (2003) Neurobiology of emotion perception I: The neural basis of normal emotion perception. Biol Psychiatry 54: 504–14. [DOI] [PubMed] [Google Scholar]
  66. Potvin S, Stavro K, Rizkallah E, Pelletier J (2014) Cocaine and cognition: a systematic quantitative review. J Addict Med 8: 368–76. [DOI] [PubMed] [Google Scholar]
  67. Poudel A, Gautam S (2017) Age of onset of substance use and psychosocial problems among individuals with substance use disorders. BMC Psychiatry 17: 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Preacher KJ, Hayes AF (2008) Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 40: 879–91. [DOI] [PubMed] [Google Scholar]
  69. Preller KH, Hulka LM, Vonmoos M, Jenni D, Baumgartner MR, Seifritz E, Dziobek I, Quednow BB (2014) Impaired emotional empathy and related social network deficits in cocaine users. Addict Biol 19: 452–66. [DOI] [PubMed] [Google Scholar]
  70. Rando K, Tuit K, Hannestad J, Guarnaccia J, Sinha R (2013) Sex differences in decreased limbic and cortical grey matter volume in cocaine dependence: a voxel-based morphometric study. Addict Biol 18: 147–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Schmahmann JD, Weilburg JB, Sherman JC (2007) The neuropsychiatry of the cerebellum - insights from the clinic. Cerebellum 6: 254–67. [DOI] [PubMed] [Google Scholar]
  72. Siemsen BM, Giannotti G, McFaddin JA, Scofield MD, McGinty JF (2019) Biphasic effect of abstinence duration following cocaine self-administration on spine morphology and plasticity-related proteins in prelimbic cortical neurons projecting to the nucleus accumbens core. Brain Struct Funct 224: 741–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Sim ME, Lyoo IK, Streeter CC, Covell J, Sarid-Segal O, Ciraulo DA, Kim MJ, Kaufman MJ, Yurgelun-Todd DA, Renshaw PF (2007) Cerebellar gray matter volume correlates with duration of cocaine use in cocaine-dependent subjects. Neuropsychopharmacology 32: 2229–37. [DOI] [PubMed] [Google Scholar]
  74. Stoodley CJ, Schmahmann JD (2010) Evidence for topographic organization in the cerebellum of motor control versus cognitive and affective processing. Cortex 46: 831–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Tiffany ST, Singleton E, Haertzen CA, Henningfield JE (1993) The development of a cocaine craving questionnaire. Drug Alcohol Depend 34: 19–28. [DOI] [PubMed] [Google Scholar]
  76. Uono S, Sato W, Kochiyama T, Sawada R, Kubota Y, Yoshimura S, Toichi M (2017) Neural substrates of the ability to recognize facial expressions: a voxel-based morphometry study. Soc Cogn Affect Neurosci 12: 487–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Van Overwalle F, Baetens K, Marien P, Vandekerckhove M (2014) Social cognition and the cerebellum: a meta-analysis of over 350 fMRI studies. Neuroimage 86: 554–72. [DOI] [PubMed] [Google Scholar]
  78. Verdejo-Garcia A, Rivas-Perez C, Vilar-Lopez R, Perez-Garcia M (2007) Strategic self-regulation, decision-making and emotion processing in poly-substance abusers in their first year of abstinence. Drug Alcohol Depend 86: 139–46. [DOI] [PubMed] [Google Scholar]
  79. Volkow ND, Baler RD, Goldstein RZ (2011) Addiction: pulling at the neural threads of social behaviors. Neuron 69: 599–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Wardle MC, Garner MJ, Munafo MR, de Wit H (2012) Amphetamine as a social drug: effects of d-amphetamine on social processing and behavior. Psychopharmacology (Berl) 223: 199–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
  81. Wilkinson G (1993) WRAT-3 : Wide Range Achievement Test Administration Manual. Wide Range Inc., Wilmington, DE [Google Scholar]
  82. Woicik PA, Moeller SJ, Alia-Klein N, Maloney T, Lukasik TM, Yeliosof O, Wang GJ, Volkow ND, Goldstein RZ (2009) The neuropsychology of cocaine addiction: recent cocaine use masks impairment. Neuropsychopharmacology 34: 1112–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  83. Wrase J, Makris N, Braus DF, Mann K, Smolka MN, Kennedy DN, Caviness VS, Hodge SM, Tang L, Albaugh M, Ziegler DA, Davis OC, Kissling C, Schumann G, Breiter HC, Heinz A (2008) Amygdala volume associated with alcohol abuse relapse and craving. Am J Psychiatry 165: 1179–84. [DOI] [PubMed] [Google Scholar]
  84. Zahr NM, Mayer D, Rohlfing T, Hasak MP, Hsu O, Vinco S, Orduna J, Luong R, Sullivan EV, Pfefferbaum A (2010) Brain injury and recovery following binge ethanol: evidence from in vivo magnetic resonance spectroscopy. Biol Psychiatry 67: 846–54. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1762896_Supinfo
NIHMS1762896-supplement-1762896_Supinfo.docx (27.2KB, docx)

RESOURCES