It remains unknown whether neural abnormalities stimulated by early heavy alcohol use persist throughout the lifespan or revert to normative patterns after periods of short or long-term abstinence. fMRI studies provide a framework for understanding the regions of the brain associated with risky decision-making. The process of making a decision involves three distinct stages: 1) the assessment and formation of preference among options; 2) the selection and execution of an action; and 3) the experience or evaluation of an outcome . Areas that reliably activate in fMRI tasks of risk-taking and reward-based decision-making include both cognitive and affective brain areas, such as the orbitofrontal/ventromedial, and dorsolateral prefrontal cortices, anterior cingulate, ventral striatum , amygdala, and insula . However, some fundamental differences in BOLD activation patterns are observed in adolescent populations completing these tasks, compared to adults and children. Specifically, adolescents display an increased BOLD response in the nucleus accumbens in response to rewarding outcomes compared to both adults and children . There is also some evidence of reduced BOLD activation in the amygdala when experiencing decision outcomes, relative to adults . This finding is of interest because the amygdala is involved in responding to negative stimuli and can function as a “behavioral brake”, signaling individuals to retreat from potentially dangerous situations . Adolescents also engage prefrontal regulatory structures to a lesser extent than do adults when making risky decisions ,vertical grow system and these findings can be interpreted as underdeveloped cognitive control systems, relative to socioemotional networks .
Neural functioning during risky decision-making has also been linked to adolescent self-reports of their risky behaviors in real life as well as task performance on cognitive measures of risk-taking. Specifically, in the Galvan et al. study , adolescents’ BOLD response in the nucleus accumbens during reward anticipation was correlated with an increased level of self-reported risky behavior, and Eshel and colleagues found that risk-taking performance during a decision-making task was negatively associated with BOLD response in regions associated with cognitive control, including the ventrolateral prefrontal and anterior cingulate regions. One known study to date has examined neural correlates of affective decision making in adolescent alcohol users. Specifically, Xiao and colleagues used a modified version of the Iowa Gambling Task with adolescents in the early stages of binge drinking as well as age-matched controls. They found that compared to controls, the binge drinkers showed increased BOLD response in the left amygdala and bilateral insulae during decision-making; in addition, BOLD response in the insulae was positively correlated with drinking severity, while BOLD response in the orbitofrontal cortex was negatively correlated with drinking severity. Of note, binge drinkers also performed more poorly on the task compared to non-drinkers . This finding is important as differences in task performance often contribute to observed differences in neural functioning. The Balloon Analogue Risk Task is one laboratory based measure of risky decision-making where participants are required to “pump up” balloons that can ultimately explode; the larger a participant inflates the balloon, the greater the risk he/she is willing to take. Behavioral studies using this measure with adults and adolescents have shown that risky decision-making on the BART correlates with scores on self-report of risky behavior as well as scores on personality constructs related to risk-taking.
Rao, Korczykowski, Pluta, Hoang, and Detre first examined use of the BART within an fMRI context. Participants were healthy adults and the task was manipulated to measure both active decision-making and passive/involuntary decision-making . In both conditions, increasing risk correlated with neural activation in bilateral visual pathways, including occipital, fusiform, and parietal cortices, which likely reflects increased attention to or processing of visual information as the balloon inflated to larger sizes. In the active decision-making condition only, increasing risk correlated with neural activation in the ventral tegmental area, striatum, anterior insula, anterior cingulate cortex/medial prefrontal cortex, and dorsolateral prefrontal cortex, suggesting that these regions are directly implicated in risky decision-making. While no known studies have yet used the BART in an fMRI context with adolescents, let alone heavy drinking adolescents, one study has used the task in a sample of young adults, aged 18-23, with a range of alcohol consumption behaviors . Results indicated that during inflation, as the probability of a balloon exploding increased , participants showed decreasing BOLD response in the medial prefrontal cortex and anterior cingulate. In addition, during successful inflation outcomes BOLD response in the medial prefrontal cortex/anterior cingulate was greater when the probability of that balloon exploding had been higher. This may indicate that outcomes were viewed as more rewarding when it took riskier behavior to achieve the outcome. Finally, decreasing medial prefrontal cortex/anterior cingulate BOLD response during riskier inflation and increasing BOLD response in the same regions during successful riskier inflation outcomes was correlated with higher alcohol use and trait disinhibition, and lower IQ .
Little, if any longitudinal information is currently available on the effects of short or long-term abstinence on rates or patterns of change in cognitive or neural functioning in heavy-drinking adolescents. However, it is well established that in adults, abstinence from alcohol leads to some recovery of neurocognitive functioning, with the rate and pattern of recovery depending on several factors, such as age, baseline cognitive ability, length of abstinence, and peak level of alcohol use. During the acute detoxification period, broad cognitive impairment is common, with memory, executive functioning, gait and balance coordination, and visuospatial skills among those domains most affected . With several weeks to months of sobriety, impairments in attention and concentration, reaction time, gait and balance, and selected executive, visuospatial, and memory abilities tend to improve, but mild to moderate difficulties in components of working memory, processing speed, non-verbal skills, and postural stability can remain . After several years of abstinence, adults appear to recover most neuropsychological abilities, although spatial processing may remain affected . In addition to some lasting cognitive impairment, abstinent adults may demonstrate externalizing behavioral symptoms, social deviance, and reward-related decision-making deficits, associated with prior peak alcohol use . Structural brain dysmorphology and altered neural functioning has also been observed in adults with persisting alcohol dependence, even among those individuals free of alcohol-related brain syndromes . In terms of dysmorphology, the brain structures most consistently affected include the corpus callosum , pons and the cerebellar hemispheres and vermis . In addition, both cortical gray matter and white matter sustain widespread volume loss, especially in the prefrontal cortex . Functional neuroimaging studies provide evidence of neural processing inefficiencies in adults with AUDs, which complement associated cognitive and structural findings. Pfefferbaum and colleagues found that alcohol dependent adults showed less BOLD activation in prefrontal regions compared to controls and more activation in posterior and inferior regions while performing a SWM task. In another study of verbal working memory, alcohol dependent adults demonstrated task performance equivalent to that of matched controls, but required increased BOLD activation in the left prefrontal cortex and right superior cerebellum to reach an effective level of performance . In a study of proactive interference ,cannabis grow equipment alcohol dependent adults engaged frontal systems to a higher degree than did controls; while controls activated lower-level subcortical systems to carry out the same task . Collectively, these fMRI results suggest that alcohol dependent adults recruit cerebellar activity to carry out functions that otherwise would be frontal lobe tasks, and frontal lobe activity to carry out functions that otherwise would be lower-level tasks. One interpretation is that the additional activation is “compensatory”, enabling alcohol dependent adults to achieve normal levels of performance, despite cerebellar and/or frontal dysmorphology . As the cerebellar and prefrontal regions are among those most structurally affected in alcoholism, these observations point to a disruption in frontocerebellar circuitry as one primary mechanism of the behavioral impairments characteristic of chronic alcohol use .
Although longitudinal findings of the effects of long-term abstinence on the brains of adults with chronic alcohol dependence are not yet available, some recent longitudinal studies of short-term abstinence have provided promising results. These studies suggest that structural brain damage is at least partially reversible, with increases in gray matter observed after 1 month of abstinence from alcohol ; white matter has also been noted as amenable to recovery with abstinence . Of interest to the study of risk-taking and reward, one study showed significantly greater volumes in the nucleus accumbens of adults with alcohol dependence who had achieved one month of sobriety, compared to alcohol dependent adults who continued drinking. In addition, the volume loss observed in the actively drinking group was greater than that typically observed in Korsakoff’s patients . In adolescents, preliminary results from three studies using the same sample as is proposed in this dissertation provide promising evidence that a similar pattern of cognitive and neural recovery may occur for heavy drinking adolescents after a short period of abstinence. Pulido and colleagues conducted an fMRI study of alcohol cue reactivity in heavy drinking adolescents and controls, assessed across three time points over a five-week period of abstinence. They found that compared to controls, heavy drinking adolescents showed greater baseline BOLD response to pictures of alcohol-related stimuli in six brain regions, including the right superior frontal gyrus, the left medial frontal/striatum, bilateral cerebellum, left cingulate, left pre and post-central gyrus, and the left middle temporal gyrus. After two to three weeks of monitored abstinence, the heavy drinkers evidenced BOLD responding patterns similar to that of controls in five out of the six brain regions that previously showed differences, showing increased BOLD response to alcohol cues only in the right superior frontal gyrus. After four to five weeks of monitored abstinence, no significant group differences in BOLD response related to alcohol cue reactivity were seen in any brain region. Another study assessed risk-taking behavior using the BART task . At the baseline assessment, heavy drinking adolescents had fewer “wins” compared to controls, but when assessed after five weeks of abstinence, the two groups were equivalent on this measure. In addition, heavy drinkers had a higher number of balloon pumps on the last half of the task than did controls, and again, this difference disappeared after drinkers had been abstinent for five weeks. Finally, the number of balloon pumps during the second half of the task was positively correlated with self-report of recent and lifetime substance use. In the third study, Winward and colleagues used a modified computer version of the Paced Auditory Serial Addition Test to measure distress tolerance levels during the task. Findings indicated that although the control adolescents outperformed heavy drinkers across the three time points on the PASAT task itself, there was a significant group by time interaction on self-reported levels of happiness and frustration before and after PASAT administration. Specifically, heavy drinkers showed greater increases in frustration and greater decreases in happiness compared to control adolescents at baseline and after two to three weeks of abstinence; however, this difference disappeared at the third time point . The results of these three studies, although preliminary, provide support for the idea that heavy alcohol use in adolescence may lead to cognitive changes and concurrently, altered neural processing. In addition, these abnormalities may normalize with sustained abstinence, which in turn, could reduce risk-taking behavior and associated negative consequences. The current study aimed to supplement these findings by examining neural functioning during a risky decision-making fMRI task in heavy drinking adolescents and matched controls across a five-week period of abstinence. The goals were to identify brain regions where activation to a risky decision-making task differed between heavy drinking adolescents and controls, examine whether neural activity associated with risky decision-making changed across a five-week period of abstinence and whether trajectories of change over time differed for heavy drinkers vs. controls, and determine whether neural activation in regions showing baseline group differences during risky decision-making predicts differences in neuropsychological functioning , risk-taking performance, or self-report of risk-taking behavior and impulsivity. Understanding the impact of heavy alcohol use on adolescents’ brain functioning during risky decision-making is an important step in clarifying factors which may further risk-taking behavior in adolescence. In addition, measuring longitudinal changes in brain functioning over a period of sustained abstinence from alcohol can elucidate whether observed alterations in neural functioning associated with heavy alcohol use normalize with abstinence or persist after cessation of use.