“The effect of alcohol use on brain networks of cognitive control and salience attribution in young adulthood: causal inferences from a cotwin control study”
Jeremy Harper, University of Minnesota, Department of Psychology
Impairments in inhibitory control, salience attribution, and their underlying brain networks, contribute to alcohol misuse. Prior research often assumes a neurotoxic effect of youth alcohol use on the brain. This assumption remains largely untested as other factors (e.g., familial risk) may confound any causal relationship. This dissertation leveraged a genetically-informative population-based sample of 24-year-old twins (N = 673) and a quasi-experimental research design to evaluate causal alcohol exposure effects on multimodal structural/functional measures of inhibitory control/salience cortical networks. Cotwin control analyses suggested that deviations in specific aspects of the control/salience networks (i.e., lateral prefrontal/frontal medial cortex thickness; midfrontal EEG activity during an inhibitory control) were due to both a potential neurotoxic exposure effect (twin differences in alcohol use) and the familial vulnerability for alcohol misuse. This dissertation provides novel evidence that alcohol-related reductions in the structure and function of inhibitory control and salience attribution brain networks likely represent both the neurotoxic effects of young adult alcohol exposure and the premorbid brain-based characteristics of the familial predisposition to misuse alcohol. The dual effects of these two alcohol-related causal influences on the brain have equally important and complementary public health implications in regard to policy and prevention efforts to curb youth drinking.
Alcohol use is a leading public health concern for young adults. According to a 2018 United States national survey1, approximately 55% (18.8 million) of individuals aged 18 to 25 were regular users and 17% (3.1 million) of those were heavy users. Emerging adulthood (age 18-25) is an important developmental period during which continued changes in neurocognitive development2,3 (particularly the prefrontal cortex [PFC]) occur concomitantly alongside peak lifetime levels of alcohol (mis)use4. This may create a vulnerable period where the structure/ function of the still-developing young adult brain is especially sensitive to the neurotoxic effects of alcohol.
It is well established that impairments in inhibitory control and salience attribution processes, and their proposed brain-based substrates assessed with MRI/EEG, reflect core attributes of alcohol misuse5–8. Research using structural MRI reports grey matter deficits in regions including the dorsolateral PFC, anterior cingulate/frontal medial cortex, insula, and temporal cortex, in relation to alcohol use9,10, all of which are areas implicated in prominent neurocognitive models of inhibitory control/salience11–14. Complementary work using EEG oscillations, which can index network-level neuronal mechanisms related to cognition at millisecond precision15, suggests that drinking is associated with reduced frontal medial theta-band power during inhibitory control tasks16–18, itself a compelling candidate electrophysiological mechanism of control/salience processes19.
Research has often assumed a causal neurotoxic effect of alcohol on control/salience brain networks20. However, those who misuse alcohol typically differ from those who do not on important characteristics that may influence both drinking and brain outcomes; rather than reflecting a causal exposure effect, alcohol-related brain anomalies may instead be due to familial risk (e.g., genetic risk, rearing environment) influences. Despite significant public health implications regarding these two potential causal sources (e.g., policy decisions, targeted prevention efforts), there remains a lack of evidence regarding potential causal influences of youth alcohol use on the brain because the overwhelming majority of prior research is correlational. There is a pressing need for novel research designs approximating true experiments21 to make stronger causal inferences regarding effects of alcohol on the young adult brain.
This dissertation was designed to address this important gap by testing the causal relationship between alcohol (mis)use in emerging adulthood and structural (MRI cortical thickness) and functional (EEG theta activity) measures of inhibitory control and salience attribution cortical networks in a large population-based, etiologically-informative sample of 24-year-old twins.
We hypothesized that greater alcohol use would be associated with reduced cortical thickness in control/salience networks and frontal medial theta power during a go/nogo inhibitory control task. We tested whether observed alcohol effects differed between women and men given suggestive evidence for increased liability to substance-related effects in women and the need for more well-powered neuroimaging studies evaluating substance-related sex differences8,22,23. Of primary interest was the cotwin control analysis24, a “natural” quasi-experiment that uses twins as ideal genetic/shared environmental controls, to more appropriately/stringently evaluate for causal neurotoxic alcohol effects (unconfounded by all familial influences) than is possible with cross-sectional or longitudinal studies of genetically unrelated individuals21,24.
Participants were 673 same-sex twins (55% women; 60% monozygotic [MZ]) with EEG and MRI data assessed at age-24 (mean[SD] age = 24.4[0.8] years) from the population-based Minnesota Twin Family Study enrichment sample25. By design and standard MRI safety exclusions, fewer individuals underwent structural MRI scans (n = 436) than EEG (n = 638) recording (n = 401 had both).
Substance use history was assessed using an expanded version of the Substance Abuse Module of the Composite International Diagnostic Interview26. The average of four dimensional alcohol use items (spanning no to heavy use) designed to assess use/exposure across emerging adulthood comprised the drinking index27: frequency of drinking; typical number of drinks per occasion; maximum number of drinks consumed in 24hrs; number of intoxications (Cronbach’s α = 0.74; possible range: 0.00–5.75).
Structural MRI was collected on 3T Siemens Trio (n = 100) and Prisma (n = 336) MRI scanners and processed using the standard Freesurfer pipeline as described in our previous report28. Cortical thickness of 112 areas covering the dorsal and lateral prefrontal, frontal medial, cingulate, and insula regions (among others) within the control and salience networks29 from the Schaefer et al.30 400-area atlas were calculated for analysis.
Continuous 61-channel EEG data were recorded during a go/nogo inhibitory control task and processed according to our previous report17. Theta was calculated as the average power between 2.5-7.0 Hz and 250-500 ms post-stimulus at midfrontal electrodes FCz/Cz.
Linear mixed models (LMMs)31,32 were fit in R33 with random intercepts adjusting for within-twin-pair correlations in dependent measures. First, we tested the individual-level associations (not adjusting for familial confounding) between alcohol use and brain measures (cortical thickness; theta), then evaluated sex-related differences for measures with significant associations with a sex × drink index interaction.
Follow-up cotwin control (CTC) analyses24 evaluated causal alcohol exposure effects by utilizing twins as ideal genetic and shared environmental controls to adjust for all measured and unmeasured sources of familial confounding. Outcomes (e.g., cortical thickness) are compared between members of a twin pair; if a twin drank more than their cotwin, the outcome of the lesser-using twin provides a close approximation of the expected outcome (unobserved counterfactual) for the heavier-using twin had she/he drank less. LMMs were fit with the brain outcome as the dependent variable and the within-pair (an individual’s deviation from his/her twin-pair mean score) and between-pair (twin-pair mean) effects as the independent variables. A significant within-pair effect is consistent with the potential causal effect of alcohol exposure (unconfounded by all familial factors that influence alcohol use34) on an outcome (e.g., heavier-drinking twins exhibiting decreased thickness relative to lesser-drinking cotwins). A significant between-pair effect is consistent with a familial risk effect influencing both drinking and the brain outcome. We also compared the magnitude of within-pair effects between monozygotic (MZ; 100% genetic control) and dizygotic (DZ; 50% genetic control) twin pairs with a zygosity × within-pair interaction; statistically comparable MZ/DZ effects are strongly consistent with an exposure effect24.
The mean drinking index score was 2.81 (SD = 1.02; range: 0.00–5.75). For those with scores ≥ 2.81, their average use across emerging adulthood corresponded to drinking 1–4 times per week, 4–6 drinks each occasion, as many as 11–20 at one time, and having been intoxicated 51–149 times. The MZ twin correlation for the drink index was more than twice the DZ correlation (MZ = 0.59; DZ = 0.21; ps ≤ 0.005), and both were well below unity, suggesting both a moderately heritable influence on drinking and substantial within-pair differences in alcohol exposure.
MRI: Greater drinking index scores were significantly associated with reduced thickness in ten control and salience network cortical areas (βs = -0.26 to -0.16, false discovery rate adjusted ps ≤ 0.037). Areas included the control network left lateral and right dorsolateral PFC, right precuneus, and right temporal lobe, and salience network bilateral frontal medial (three sub-areas), parietal medial (two sub-areas), and right medial posterior prefrontal cortex. No effects statistically differed as a function of sex (ps ≥ 0.126). These areas, particularly the six (pre)frontal areas, have been implicated in both the normative expression14 and alcohol-related impairments8 of inhibitory control/salience attribution processes.
In CTC analyses, the between-pair effect was significant for all ten areas (βs = -0.24 to -0.14, ps ≤ 0.029), and the within-pair effect was significant for the control network left lateral prefrontal and right temporal cortices, and salience network bilateral frontal medial and right parietal medial cortices (βs = -0.12 to -0.08, ps ≤ 0.045). Consistent with expectations for a within-pair effect24, the MZ and DZ effects were equivalent (ps ≥ 0.509).
EEG: The results of source localization analysis35 indicated estimated cortical sources of theta-band power during inhibitory control in the anterior midcingulate cortex and associated frontal medial areas.
Greater drink index scores were associated with decreased theta power (β = -0.08, p = 0.043), suggesting diminished frontal medial brain activity during inhibitory control in relation to alcohol use. This effect differed by sex (β = -0.28, p = 0.025) such that it was significant for women (β = -0.16, p = 0.004) but not men (β = 0.03, p = 0.606).
The CTC indicated a significant between-pair (β = -0.20, p = 0.002) and within-pair (β = -0.10, p = 0.045) effect on control-related frontal medial theta power in women. Again consistent with expectations for a within-pair effect, the MZ and DZ effects were equivalent (p = 0.837).
Overall, the pattern of CTC results suggest that the observed effects likely reflect both the brain manifestation of the familial predisposition to misuse alcohol, and specifically for lateral PFC, temporal, and frontal/parietal medial thickness and medial frontal theta power, a causal alcohol exposure effect. That is, after adjusting for all measured and unmeasured sources of potential familial risk (i.e., genetics, shared environment) that affect these brain outcomes, heavier-drinking twins showed decreased thickness of the lateral PFC, temporal, and frontal/parietal medial areas and frontal medial theta power relative to their lesser-drinking cotwins.
This dissertation represents a significant advancement in the neurobiology of alcohol misuse and provides important novel evidence regarding the causal effects of alcohol exposure on the young adult brain. We used a quasi-experimental cotwin control analysis to test the causal effects of alcohol on multimodal MRI/EEG correlates of control/salience attribution brain network in a field where true experimentation is often unfeasible. A major strength of this study is its potential generalizability to the community at large by use of a population-based representative sample of 24-year-old young adults whose level/range of drinking is comparable to that observed in the general population1.
Evidence from the cotwin control analysis suggested a causal effect of alcohol exposure on the thickness of the lateral PFC, temporal, and frontal/parietal medial areas of the control/salience networks, and frontal medial theta-band power during an inhibitory control task. That is, reductions in these brain measures were observed in twins who drank more relative to their lesser-drinking cotwin, strongly implying that alcohol exposure during emerging adulthood confers a detrimental causal effect to inhibitory control and salience attribution brain networks. Effects on frontal medial theta-band power were only observed in women, possibly suggesting that they may be at higher risk for certain alcohol-related causal effects; more work is necessary to understand sex differences in alcohol-related brain outcomes. Exposure effects appear to develop as early as age 24; emerging adulthood may represent a ‘window of vulnerability’ for the neurotoxic consequences of alcohol (mis)use. If confirmed through further research, the finding that even normative levels of alcohol use during emerging adulthood produce causal effects on key brain networks of inhibitory control and salience attribution has potentially significant public health implications.
In addition, we also observed alcohol effects that are likely not due to drinking and instead reflect the brain-based expression of the vulnerability to misuse alcohol. The same familial risk factors (e.g., genetic liability) that influence drinking also appear to influence the structure and function of the control/salience brain measures associated with drinking. These control/salience brain measures are likely premorbid characteristics that would be observed prior to, and confer risk for, drinking escalation. Because alcohol misuse is comorbid with the externalizing spectrum36, individuals with these brain-based predispositions are also at heightened risk for other negative outcomes including addiction, illicit drug use, and antisocial behavior.
Results suggest that premorbid dysregulation of the inhibitory control/salience attribution brain networks may predispose an individual toward alcohol misuse, which then confers an additional neurotoxic effect on certain aspects of the systems. Findings can be used to inform etiological/developmental models of alcohol misuse and public health decisions. Both the neurotoxic and familial predisposition effects reported in this dissertation have important public health implications in terms of targeted preventions. A potential consequence of these dual alcohol-related causal effects is that efforts should be focused on both reducing young adult drinking to prevent neurotoxic exposure effects, and targeting individuals with the brain-based predisposition (endophenotypes37) toward alcohol misuse to help identify and intervene with high-risk youth before they begin drinking.
Emerging adulthood is an important developmental period characterized by peak lifetime levels of alcohol use and misuse that may impact the structural or functional outcomes of the still-developing young adult brain. This dissertation used a genetically-informative twin-based research design to provide novel evidence that reductions in cortical thickness and functional activation of inhibitory control and salience attribution cortical networks are likely due to both the familial predisposition to use alcohol and (for certain measures) the additional deleterious effects of alcohol misuse. The dual effects of familial risk and alcohol exposure on the brain have equally important and complementary public health implications. In terms of prevention, efforts to reduce young adult drinking would reduce the likelihood of individuals developing these neurotoxic effects as early as emerging adulthood, and efforts to target high-risk youth (e.g., identifying those with these brain-based predispositions) before alcohol initiation could decrease their risk of developing problematic drinking and other related negative outcomes, such as addiction and antisocial behavior, later in life.
- Substance Abuse and Mental Health Services Administration. Key substance use and mental health indicators in the United States: Results from the 2018 National Survey on Drug Use and Health (HHS Publication No. PEP19-5068, NSDUH Series H-54). Rockville, MD: Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration (2019).
- Shaw, P. et al. Neurodevelopmental trajectories of the human cerebral cortex. J. Neurosci. 28, 3586–3594 (2008).
- Sowell, E. R. et al. Mapping cortical change across the human life span. Nat. Neurosci. 6, 309–315 (2003).
- Grant, B. F. et al. Prevalence of 12-Month Alcohol Use, High-Risk Drinking, and DSM-IV Alcohol Use Disorder in the United States, 2001-2002 to 2012-2013. JAMA Psychiatry vol. 74 911 (2017).
- Goldstein, R. Z. & Volkow, N. D. Dysfunction of the prefrontal cortex in addiction: neuroimaging findings and clinical implications. Nat. Rev. Neurosci. 12, 652–669 (2011).
- Koob, G. F. & Volkow, N. D. Neurocircuitry of Addiction. Neuropsychopharmacology 35, 1051–1051 (2010).
- Everitt, B. J. & Robbins, T. W. Drug Addiction: Updating Actions to Habits to Compulsions Ten Years On. Annu. Rev. Psychol. 67, 23–50 (2016).
- Zilverstand, A., Huang, A. S., Alia-Klein, N. & Goldstein, R. Z. Neuroimaging Impaired Response Inhibition and Salience Attribution in Human Drug Addiction: A Systematic Review. Neuron 98, 886–903 (2018).
- Welch, K. A., Carson, A. & Lawrie, S. M. Brain Structure in Adolescents and Young Adults with Alcohol Problems: Systematic Review of Imaging Studies. Alcohol Alcohol 48, 433–444 (2013).
- Feil, J. et al. Addiction, compulsive drug seeking, and the role of frontostriatal mechanisms in regulating inhibitory control. Neurosci. Biobehav. Rev. 35, 248–275 (2010).
- Ridderinkhof, K. R., van den Wildenberg, W. P. M., Segalowitz, S. J. & Carter, C. S. Neurocognitive mechanisms of cognitive control: The role of prefrontal cortex in action selection, response inhibition, performance monitoring, and reward-based learning. Brain Cogn. 56, 129–140 (2004).
- Botvinick, M., Braver, T. S., Barch, D. M., Carter, C. S. & Cohen, J. D. Conflict monitoring and cognitive control. Psychol. Rev. 108, 624–652 (2001).
- Botvinick, M. & Braver, T. Motivation and cognitive control: from behavior to neural mechanism. Annu. Rev. Psychol. 66, 83–113 (2015).
- Ullsperger, M., Danielmeier, C. & Jocham, G. Neurophysiology of Performance Monitoring and Adaptive Behavior. Physiol. Rev. 94, 35–79 (2014).
- Siegel, M., Donner, T. H. & Engel, A. K. Spectral fingerprints of large-scale neuronal interactions. Nat. Rev. Neurosci. 13, 121–134 (2012).
- Kamarajan, C. et al. The role of brain oscillations as functional correlates of cognitive systems: a study of frontal inhibitory control in alcoholism. Int. J. Psychophysiol. 51, 155–180 (2004).
- Harper, J., Malone, S. M. & Iacono, W. G. Impact of alcohol use on EEG dynamics of response inhibition: a cotwin control analysis. Addict. Biol. 23, 256–267 (2018).
- Pandey, A. K. et al. Delta, theta, and alpha event-related oscillations in alcoholics during Go/NoGo task: Neurocognitive deficits in execution, inhibition, and attention processing. Prog. Neuropsychopharmacol. Biol. Psychiatry 65, (2016).
- Cavanagh, J. F. & Frank, M. J. Frontal theta as a mechanism for cognitive control. Trends Cogn. Sci. 18, 414–421 (2014).
- Jacobus, J. & Tapert, S. F. Neurotoxic Effects of Alcohol in Adolescence. Annual Review of Clinical Psychology 9, 703–721 (2013).
- Thapar, A. & Rutter, M. Do natural experiments have an important future in the study of mental disorders? Psychol. Med. 49, 1079–1088 (2019).
- Becker, J. B. & Koob, G. F. Sex Differences in Animal Models: Focus on Addiction. Pharmacol. Rev. 68, 242–263 (2016).
- Erol, A. & Karpyak, V. M. Sex and gender-related differences in alcohol use and its consequences: Contemporary knowledge and future research considerations. Drug Alcohol Depend. 156, 1–13 (2015).
- McGue, M., Osler, M. & Christensen, K. Causal Inference and Observational Research: The Utility of Twins. Perspect. Psychol. Sci. 5, 546–556 (2010).
- Keyes, M. A. et al. The Enrichment Study of the Minnesota Twin Family Study: Increasing the Yield of Twin Families at High Risk for Externalizing Psychopathology. Twin Res. Hum. Genet. 12, 489–501 (2009).
- Robins, L. N., Babor, T. F. & Cottler, L. B. Composite international diagnostic interview: expanded substance abuse module. St. Louis: Authors (1987).
- McGue, M., Malone, S., Keyes, M. & Iacono, W. G. Parent-Offspring Similarity for Drinking: A Longitudinal Adoption Study. Behav. Genet. 44, 620–628 (2014).
- Wilson, S., Malone, S. M., Hunt, R. H., Thomas, K. M. & Iacono, W. G. Problematic alcohol use and hippocampal volume in a female sample: disentangling cause from consequence using a co-twin control study design. Psychol. Med. 1–12 (2017).
- Yeo, B. T. T. et al. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165 (2011).
- Schaefer, A. et al. Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cereb. Cortex 28, 3095–3114 (2018).
- Bates, D., Machler, M., Bolker, B. M. & Walker, S. C. Fitting Linear Mixed-Effects Models Using lme4. J. Stat. Softw. 67, 1–48 (2015).
- Kuznetsova, A., Brockhoff, P. B. & Christensen, R. H. B. lmerTest package: Tests in linear mixed effects models. J. Stat. Softw. 82, 1–26 (2017).
- R Core Team. R: A language and environment for statistical computing. (R Foundation for Statistical Computing, 2019).
- Begg, M. D. & Parides, M. K. Separation of individual-level and cluster-level covariate effects in regression analysis of correlated data. Stat. Med. 22, 2591–2602 (2003).
- Oostenveld, R., Fries, P., Maris, E. & Schoffelen, J.-M. FieldTrip: open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data. Comput. Intell. Neurosci. 2011, 1 (2011).
- Iacono, W. G., Malone, S. M. & McGue, M. Behavioral disinhibition and the development of early-onset addiction: common and specific influences. Annu. Rev. Clin. Psychol. 4, 325–348 (2008).
- Iacono, W. G., Malone, S. M. & Vrieze, S. I. Endophenotype best practices. Int. J. Psychophysiol. 111, 115–144 (2017).