Adolescents show differential dysfunctions related to Alcohol and Cannabis Use Disorder severity in emotion and executive attention neuro-circuitries

Abstract

Alcohol and cannabis are two substances that are commonly abused by adolescents in the United States and which, when abused, are associated with negative medical and psychiatric outcomes across the lifespan. These negative psychiatric outcomes may reflect the detrimental impact of substance abuse on neural systems mediating emotion processing and executive attention. However, work indicative of this has mostly been conducted either in animal models or adults with Alcohol and/or Cannabis Use Disorder (AUD/CUD). Little work has been conducted in adolescent patients. In this study, we used the Affective Stroop task to examine the relationship in 82 adolescents between AUD and/or CUD symptom severity and the functional integrity of neural systems mediating emotional processing and executive attention. We found that AUD symptom severity was positively related to amygdala responsiveness to emotional stimuli and negatively related to responsiveness within regions implicated in executive attention and response control (i.e., dorsolateral prefrontal cortex, anterior cingulate cortex, precuneus) as a function of task performance. In contrast, CUD symptom severity was unrelated to amygdala responsiveness but positively related to responsiveness within regions including precuneus, posterior cingulate cortex, and inferior parietal lobule as a function of task performance. These data suggest differential impacts of alcohol and cannabis abuse on the adolescent brain.

Keywords: Adolescent; Alcohol Use Disorder; Amygdala; Cannabis Use Disorder; Prefrontal cortex; fMRI.

Fig. 1

Diagram of aST for a trial with a negatively valenced stimulus. The (A) first row indicates an incongruent trial, the (B) second row indicates a congruent trial, and the (C) third row indicates a view trial.

Fig. 2

AUDIT-by-emotion interaction within the (A) Amygdala ROI (x = 29 mm, y = −7 mm, z = −7 mm). Participants with higher AUDIT scores showed increased responses to emotional relative to neutral stimuli (k = 5 voxels). Values in the bar graph represent the correlation coefficients between AUDIT scores and BOLD responses for each emotion; * indicates significant differences between partial correlation values (Steiger's Z > 1.96, p < 0.05). (B) AUDIT-by-CUDIT interaction effect within the negative view trials (k = 9 voxels). Values in the bar graph represent the beta weights for the effect of AUDIT score on BOLD response within the range of CUDIT scores indicated. * indicates regions of interest significant at p < 0.05 identified via the Johnson-Neyman technique.

Fig. 3

AUDIT-by-task condition interactions within the (A) dlPFC (x = 26 mm, y = 35 mm, z = 44 mm); (B) ACC/dmPFC (x = 2 mm, y = 11 mm, z = 44 mm); and (C) PCC/Precuneus (x = 11 mm, y = −67 mm, z = 29 mm). Participants with higher AUDIT scores showed decreased responses in these brain regions during incongruent trials relative to congruent and view trials. Values in the bar graphs represent the correlation coefficients between AUDIT scores and BOLD response for each task condition within each cluster. CUDIT Score-by-task condition interactions within the (D) PCC (x = 8 mm, y = −52 mm, z = 26 mm). Participants with higher CUDIT scores showed increased responses in these brain regions during incongruent trials relative to congruent and view. Values in the bar graphs represent the correlation coefficients between CUDIT scores and BOLD responses for each task condition within each cluster. * indicates significant differences between partial correlation values (Steiger's Z > 1.96, p < 0.05).

Fig. S1

Main effects of task condition within clusters displayed in Fig. 3. * indicates significant differences at p < 0.05.