PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Alcohol Clin Exp Res. Author manuscript; available in PMC 2013 August 1.
Published in final edited form as:
PMCID: PMC3412943
NIHMSID: NIHMS350147

Resiliency in adolescents at high-risk for substance abuse: flexible adaptation via subthalamic nucleus and linkage to drinking and drug use in early adulthood

Barbara J. Weiland, Ph.D.,1,2 Joel T. Nigg, Ph.D.,3 Robert C. Welsh, Ph.D.,1,4 Wai-Ying Wendy Yau, B.S.,1,5 Jon-Kar Zubieta, M.D., Ph.D.,1,4,5 Robert A. Zucker, Ph.D.,1,2 and Mary M. Heitzeg, Ph.D.1,2

Abstract

Introduction

The personality trait resiliency is the ability to flexibly adapt impulse control relative to contextual demand. Low resiliency has been linked to later alcohol/drug problems. The underlying psychological and neural mechanisms are unknown but neurocomputational models suggested relations between resiliency and working memory. Cortical-striatal connectivity has been proposed to underlie adaptive switches between cautious and risky behaviors.

Methods

Working memory was probed in sixty-seven 18–22 year olds from a larger community study of alcoholism, using the n-back task during functional magnetic resonance imaging. Functional connectivity between task-related regions was investigated with psychophysiological interaction analysis. Resiliency was measured in early teen years and related to early adulthood measures of drinking/drug use, task activation and connectivity. Relationships with risk factors, including family history, age of drinking onset and number of alcohol problems were also investigated.

Results

Higher resiliency was related to lower levels of substance use, fewer alcohol problems and better working memory performance. Whole brain regression revealed resiliency negatively correlated with activation of subthalamic nucleus (STN) and pallidum during the n-back. High and Low resiliency quartile groups (n=17 each) differed in coupling strength between STN and median cingulate cortex, a region of reduced activation during working memory. The High resiliency group had later onset of drinking, fewer alcohol problems, had used fewer illicit drugs and were less likely to smoke cigarettes than their Low resiliency counterparts,

Conclusions

These findings suggest that resiliency in early adolescence may protect against alcohol problems and drug use, though the direction of this effect is currently unknown. This protective factor may relate to executive functioning as supported by the finding of a neural link shared between resiliency and working memory in basal ganglia structures. The STN, a key basal ganglia structure, may adaptively link flexible impulse control with cognitive processing, potentially modulating substance use outcomes.

Key Works: Resiliency, Substance Use, Working Memory, STN, fMRI, PPI

Resiliency is defined as the ability for flexible adaptation of psychological control functions appropriate to local context (Block et al., 1988, Eisenberg and Morris, 2002). For example, someone with high resiliency could be both impulsive at a party and appropriately controlled in a classroom and able to cope with stress by modulating impulse control. Conversely, someone with low resiliency might be over- or under-controlled in all settings, adapting less effectively to stress and unable to modulate behavior.

We previously examined developmental trajectories of resiliency relative to substance use (Wong et al., 2006) as part of the Michigan Longitudinal Study (MLS), an ongoing, prospective community study of families with parental alcoholism and contrast nonalcoholic families (Zucker et al., 2000, Zucker et al., 1996). That study found low resiliency, measured at 3–4 years, was associated with early alcohol use (by age 14) and drunkenness by age 17 (Wong et al., 2006), similar to a general community study which found low resiliency in preschoolers related to drug use in early adolescence (Block et al., 1988). Resiliency, which remained stable from preschool through adolescence (Wong et al., 2006), may represent an important component among the psychological strategies needed to cope with the range of personal, social and cognitive challenges facing today’s youth.

As temperament and executive function have been identified as precursors and mediators of psychopathology (Barkley, 1997, Eisenberg et al., 2000, Nigg, 2000), another MLS study investigated and found consistent relations between resiliency and executive functioning (Martel et al., 2007). Resiliency may share underpinnings with executive abilities, perhaps enhancing its development, or interacting with it, to shape cognition and social adaptation. Indeed, resiliency and executive functioning contributed to adolescent outcomes in an additive, incremental fashion, as opposed to overlapping in their effects (Martel et al., 2007), supporting the hypothesis that they are to some degree separable.

However, executive function is itself a broad, under-specified construct, likely including response suppression, planning, mental set shifting, and working memory (e.g., (Pennington and Ozonoff, 1996). The present study examines neural responses during working memory which has been identified as relevant to impulsive behavior (Finn et al., 1999) and substance use outcomes (Corral et al., 1999, Ozkaragoz et al., 1997). Neuroimaging studies show working memory processes involve activation of a network including the basal ganglia and anterior cingulate, parietal and prefrontal cortices (Smith and Jonides, 1997, Owen et al., 2005, Chang et al., 2007) accompanied by deactivation in the ‘default network’ including posterior cingulate and medial frontal cortex (Greicius and Menon, 2004, Raichle et al., 2001, Spadoni et al., 2008) similar to other executive tasks. It has been suggested that resiliency is related to an anterior attention system involving cingulate and prefrontal cortices and their projections to the basal ganglia and thalamus (Eisenberg et al., 2003, Rothbart et al., 2000).

Neurocomputational models propose the basal ganglia performs dynamic gating of working memory via disinhibition to allow the prefrontal cortex to focus on task demands (O’Reilly and Frank, 2006). Inhibitory control of the basal ganglia is influenced by the subthalamic nucleus (STN) whose activity produces slower more accurate choices (Frank et al., 2007). The STN is believed to be dynamically modulated through “proactive inhibition” of sensorimotor responses by the medial prefrontal cortex, precuneus/posterior cingulate and inferior parietal cortex as well as the dorsal anterior cingulate based on the degree of decision conflict (Frank, 2006, Botvinick et al., 2001, Yeung et al., 2004, Bush et al., 1998, Ballanger et al., 2009). In addition, changes in cortical–striatal connectivity are proposed to underlie adaptive switches between cautious and risky behaviors (Forstmann et al., 2010). We hypothesized, then, that resiliency may influence substance use outcomes via flexible adaptation entailing some of the same neural networks as working memory.

To test this hypothesis, resiliency, measured by observer ratings during early teen years, was examined in relation to brain activation during working memory and with self-reported drinking and drug use in young adulthood. We expected resiliency to be negatively related to both substance use and to inhibitory activity in the cortico-striatal working memory pathway. We also hypothesized resiliency would be linked with measures of vulnerability to substance abuse, including family history of alcoholism (National Institute on Alcohol Abuse and Alcoholism, 2000), early onset of drinking (Hingson et al., 2006, Grant and Dawson, 1997), and higher levels of alcohol-related problems (Bonomo et al., 2004, Dick et al., 2011, Viner and Taylor, 2007). Using psychophysiological interaction analysis (PPI, (Friston et al., 1997), we evaluated functional coupling of regions associated with resiliency, proposing differences by level of resiliency (high versus low quartiles).

METHODS AND MATERIALS

Participants

Participants were 67 right-handed youth (43 males/24 females), aged 18.0 – 22.3 years (mean 20.2±1.2), recruited from the MLS study of families with parental alcoholism (FH+; 2/3 of sample) and contrast nonalcoholic families (FH-; 1/3 of sample). Parental alcoholism was based on DSM-IV criteria. A detailed description regarding MLS recruitment strategy/assessment procedures can be found elsewhere (Zucker et al., 2000, Zucker et al., 1996). Exclusionary criteria were: 1) any neurological, acute, uncorrected or chronic medical illness; 2) any current or recent (within six months) treatment with centrally active medications; 3) a history of psychosis or schizophrenia in first-degree relatives. Presence of most Axis I psychiatric or developmental disorders was also exclusionary. However, externalizing disorders (i.e., conduct disorder, attention deficit/hyperactivity disorder (ADHD), or prior substance use disorder) were not exclusionary as these are on the same developmental spectrum with alcoholism risk (Krueger, 1999). In addition, participants were given a multidrug five-panel urine screen before scanning and those with a positive drug screen were not included in this study. Participants gave written informed consent after explanation of the experimental protocol, as approved by the local Institutional Review Board.

Measures

Resiliency Measure

Resiliency was assessed by observer ratings using the California Child Q-Sort common language version (Block and Block, 1980, Caspi et al., 1992) for participants when they were 12–15 years (mean 13.5±0.9) as part of the ongoing MLS (Martel et al., 2007). The Q-sort was completed by clinically-trained assessors, blinded to family history status of subjects, following a 3–4 hour interview/testing protocol with the child (Shedler and Block, 1990). The Q-Sort consists of 100 cards that must be placed in a forced-choice, nine-category normal distribution. The assessor described the subject by placing descriptive cards in one of the categories, ranging from 1 (least descriptive) to 9 (most descriptive). The resiliency scale was indexed by eleven items suggested by Eisenberg et al. (2003, 1997), e.g., is resourceful in initiating activities; uses and responds to reason. All items scored are listed in Supplemental Table S1. Scores are means of item totals with high scores indicative of more resiliency. Resiliency has been shown relatively stable over time (Hart et al., 1998) which was also true in our sample (3.0–17.9 years, r=0.27, p<0.01, (Martel et al., 2009). We chose an early adolescence time point as predictor to outcome in early adulthood. To maximize variance, Low and High resiliency groups were defined as lower and upper quartiles (n=17 each).

Drinking and Drug Use Measures

The self-report Drinking and Drug History (DDHx, Zucker et al., 1990, Zucker and Fitzgerald, 1994) was completed by participants annually since age 11. Data used were collected mean 0.83±1.04 years from scan and reflect extent of drug involvement since age 11. Total drinking in past 6 months was calculated from DDHx counts of drinking days/month multiplied by drinks usually consumed/drinking day. Alcohol problems (AP) were number of drinking problems (out of possible 37 items) ever reported by the subject since age 11. Number of illicit drugs ever used was quantified from a list of 18 drugs ever reported using over their lifetime. Cigarette smoking was determined from DDHx assessment or the more proximal pre-scan screening question, “Do you smoke?”, and coded: 0 (non-smoker, n=23) and 1 (smoker, n=44). Substance use data were not normally distributed (Kolmogorov-Smirnov (KS) test, Z’s>1.91, p’s<0.032) and were not normalized with standard transformations (square root, inverse; KS Z’s>1.34, p’s<0.035), and were therefore treated as a non-parametric variables. Age of onset was determined from the first annual DDHx on which the target reported first drink age and was normally distributed (KS test, Z=1.123, p=0.160).

fMRI Task

The n-back task (Callicott et al., 1999) required subjects to continually update their mental set while responding to previously seen stimuli. Subjects viewed stimuli consisting of numbers (1–4) shown in random sequence displayed on a diamond-shaped box (see Figure 1). On each trial, subjects press one of four buttons to indicate the appropriate numeral. In the 0-back condition, the correct response is the numeral currently displayed on the screen; for 2-back, the correct response is the numeral seen 2 screens back. The actual task included five conditions (0-, 1-, 2-, 3-back and rest); each of 5 runs consisted of 30-second blocks of each load, pseudorandomly ordered, with 15 stimuli per block. Total task time was approximately 15 minutes with all responses recorded.

Figure 1
Depiction of n-back working memory task showing correct responses for the 0-back and 2-back conditions. The 0-back response requires identifying the numeral currently presented while the 2-back response involves updating, temporarily maintaining and storing ...

MRI Data Acquisition

Whole-brain BOLD fMRI data were acquired on a 3.0 Tesla GE Signa system, Excite2 release, (Milwaukee, WI), standard radio frequency coil. Functional imaging was performed using T2*-weighted single-shot combined spiral in/out acquisition (Glover and Law, 2001): repetition time (TR)=2000 ms, echo time (TE)=30 ms, flip angle=90°, field-of-view (FOV)=0cm, 64×64 matrix, slice thickness=4mm, 29 slices. High-resolution anatomical T1 scans were acquired for spatial normalization. Motion was minimized with foam pads and emphasis on the importance of keeping still.

Data Analysis

Resiliency, Performance and Drinking and Drug Use

Performance was measured by: 1) response time (RT); 2) performance decrement (PD) defined as the difference in correct response rates in 0-back and 2-back conditions (Jansma et al., 2000). The 2-back load was utilized as it maintained cognitive demand without exceeding capacity constraints (Callicott et al., 1999, Sweet et al., 2008) with satisfactory accuracy (77.9%±16.7 here). RTs less than 100 milliseconds were considered anticipatory responding and removed from subsequent analyses. All performance measures were normally distributed (KS test, p’s>0.07) in SPSS version 17 (Chicago, IL).

Bivariate Pearson’s or Spearman’s rank correlations were used to test associations between resiliency and performance and substance use measures. Significance (2-tailed) was established at p<0.025 accounting for multiple comparisons; RT and PD were highly correlated (r= −0.332, p=0.008), as were all substance use/vulnerability measures (ρ’s>0.692, p’s<0.004). Independent sample t-tests were used to investigate differences in resiliency, performance, substance use, and vulnerability by gender, family history and resiliency groups.

fMRI Data Preprocessing

Functional data were reconstructed using iterative image reconstruction (Sutton et al., 2003, Fessler et al., 2005) and motion-corrected using FSL 4.0 (Analysis Group, FMRIB, Oxford, UK). Runs exceeding 2mm translation or 2° rotation were excluded: six subjects had one run; five subjects had two runs removed. Number of subjects with excluded runs did not differ by resiliency or family history (p’s>0.203), but trended by gender (χ2=3.62, p=0.073). Age of onset and substance use did not differ between subjects with/without excluded runs (p’s>0.234). Image processing was completed using statistical parametric mapping SPM2 (Wellcome Institute of Cognitive Neurology, Oxford, UK). Functional images were spatially normalized to standard stereotactic space as defined by the Montreal Neurological Institute. A 6mm full-width-half-maximum Gaussian spatial smoothing kernel was applied.

Individual Task Statistical Maps and Group Correlation Analysis

A general linear model using SPM’s canonical hemodynamic response function (HRF), modeled each condition (rest, 0-, 1-, 2-, and 3-back), and six motion regressors. Linear contrasts compared task load versus 0-back. A second-level one-sample t-test investigated task effect using 2-back versus 0-back contrasts. A second-level linear regression used individual resiliency scores as covariate and the same contrasts as dependent variable. To test our hypotheses, an a priori regions-of-interest mask (frontal-cingulate-parietal-basal ganglia; (Bogacz et al., 2010, O’Reilly and Frank, 2006)) was created using WFU Pickatlas (Maldjian et al., 2003). Regions of significant correlation within this mask were identified using a voxel-wise threshold of p<0.005 uncorrected, combined with cluster size threshold of 61 contiguous voxels. This combined threshold provides protection against type I error (Forman et al., 1995) and was estimated with Monte Carlo simulation using AlphaSim (Douglas Ward, http://afni.nimh.nih.gov/pub/dist/doc/program_help/AlphaSim.html) giving an overall corrected threshold of p<0.05. For identified clusters, activation data were extracted from individual contrast maps for correlation with behavioral measures

Individual and Group Functional Connectivity Analysis

Psychophysiological interaction (PPI) determines regions whose time-series of activation exhibit significant covariance with the seed region as a function of task manipulation, i.e. 2-back versus 0-back. Regressing out the contribution of the seed region time-series and the experimental context, the interaction is the contribution-dependent change in regional responses to the experimental factor (Friston et al., 1997), here working memory load. The clusters identified in the regression analysis (see Results), right pallidum and subthalamic nucleus (STN), were used for PPI seed regions of interest (ROIs). For these ROIs, the time-series data from the primary model was extracted for 5mm-diameter spheres centered at [MNI coordinates: 20, −2, 6] for right pallidum and [10, −14, −8] for STN. Anatomical validation of STN is based on work by Aron and Poldrack (2006). Each ROI time-series was deconvolved with the canonical HRF to create neuronal time-series (Gitelman et al., 2003). The PPI interaction term was the product of the neuronal time-series and a contrast vector coding for main effect of task (1 for 2-back; −1 for 0-back). This term was convolved with the HRF; PPI model regressors consisted of the interaction term, contrast vector and extracted time-series plus motion regressors from the original design (Friston et al., 1997). Single subject contrasts for the first regressor (interaction term) were calculated and used for second-level two sample t-tests evaluating Low and High resiliency groups.

Significant group differences in functional coupling with seed ROIs, masked with the a priori mask described above, were determined using a voxel-wise threshold of p<0.005, cluster threshold of 71 voxels, from an additional Monte Carlo estimation as previously described. For identified clusters, connectivity data were extracted from individual PPI maps for correlation with behavioral measures.

RESULTS

Resiliency, Demographic and Performance Data

Table 1 presents demographic and working memory performance data for the entire group and the Low- and High-quartile resiliency groups separately. Mean resiliency score for the entire sample was 5.5±1.2, (range 3.1–7.2), in line with that of the entire MLS population: 5.8±0.8, (range 3.1–7.6), n=496. Performance and RTs were in agreement with previous work using the same n-back task (Jansma et al., 2000). No family history or gender effects were found for performance (p’s>0.18). Resiliency did not differ by family history (t=0.06, p=0.952) but females showed a trend for higher resiliency (t=1.81, p=0.075); therefore gender was added as a covariate in subsequent analyses.

Table 1
Subject Characteristics

Consistent with previous work from the entire MLS sample (Martel et al., 2007), Pearson’s correlations showed positive correlation between resiliency and IQ (r=0.39, p=0.001); IQ was added as a covariate in subsequent analyses. Resiliency had negative correlations with RT and PD (Table 2) maintaining significance when controlled for gender and IQ (p’s<0.045).

Table 2
Statistics for Relationships between Resiliency with Working Memory Performance, Age of Onset and Drinking and Drug Use for Entire Sample (n=67)

Resiliency, Vulnerability and Substance Use Data

There were no differences in alcohol problems, total drinking, illicit drugs used, or smoking by family history (t’s<1.45, p’s>=0.151) or gender (t’s<1.58, p’s>=0.158). There was a trend for a difference in age of onset by family history (FH-/FH+: 15.6±2.2/14.3±2.7, p=0.056) but not gender (M/F: 14.6±2.0/14.7±2.8, p=0.939). Resiliency negatively correlated with number of alcohol problems and number of illicit drugs used, was significantly lower in cigarette smokers, and showed a positive trend with age of onset (Table 2).

Low and High resiliency groups did not differ by gender or family history, but did differ in reaction times and all substance use measures except total drinking in the past six months, as expected.

Neuroimaging Results

Effect of Task in 2-back Working Memory

Second level analysis for the entire sample (n=67) revealed working memory task activation occurred in right prefrontal and anterior cingulate cortex, inferior parietal lobe and left inferior frontal gyrus similar to previous studies (Callicott et al., 1999, Braver et al., 1997, Jansma et al., 2000, Owen et al., 2005). Task-related reductions in activation were found in posterior cingulate cortex, medial frontal regions and bilateral pre-/post-central gyri similar to previous work (Hampson et al., 2006). See Figure 2 for 2-back versus 0-back contrast mappings and supplemental Table S2.

Figure 2
Group whole-brain contrast maps for 2-back vs. 0-back contrast for working memory task, displayed at a threshold of p<0.05 FDR corrected and minimum cluster size of 10, coordinates in MNI space. Orange regions represent regions more activated ...

Correlation of Resiliency with 2-back Task Related Regions

Resiliency correlated with reduced activation in right pallidum and STN (see Table 3). At a more lenient threshold, p<0.005 uncorrected, voxel extent 50, this effect also held in the left STN and pallidum. Figure 3 shows STN and pallidum ROIs and extracted data plotted against resiliency.

Figure 3
A) STN and pallidum regions of interest with negative correlation with resiliency measures from whole-brain group analysis displayed at a threshold of p<0.005 and minimum cluster size of 50, coordinates in MNI space. B) Extracted bilateral pallidum ...
Table 3
Brain Regions Identified in Whole Brain Analyses during 2-back vs. 0-back Working Memory Task

Post-hoc analysis of extracted ROI data was conducted excluding subjects with any diagnosis of SUD, conduct disorder, or ADHD (n=7: two FH-/five FH+, resulting in n=60). Resiliency remained significantly correlated with both regions at p<0.005, confirming that effects of externalizing disorders were not driving these correlations. Additional post-hoc regression of resiliency with STN and pallidum activation maintained significance controlling for IQ (p’s<0.001) and gender (p’s<0.028).

Relationship between Activation Data and Behavioral Measures

Right STN activation showed a positive trend with 2-back PD (r =0.22, p=0.089) and no correlations with RT. No correlations were found between substance use, alcohol problems, age of onset, and activation across the entire sample (p’s>0.68) or within gender, family history, or resiliency groups (p’s>0.14).

PPI Analysis of Functional Connectivity with Right Pallidum

The test comparing Low and High resiliency quartiles revealed no clusters having connectivity differences meeting significance criteria.

PPI Analysis of Functional Connectivity with Right STN

The Low>High resiliency contrast revealed the Low group had stronger connectivity between right STN and a cluster in right median cingulate (BA 23), an area with reduced activation during 2-back demand, see Table 3. Post-hoc regression maintained significance when controlling for IQ and gender (F=19.03, p<0.001). At a more lenient threshold, p<0.005 uncorrected, voxel extent 25, a cluster in left median cingulate was identified, see Figure 4. The High>Low contrast had no clusters meeting significance criteria.

Figure 4
A) Bilateral median cingulate regions with different connectivity between Low and High resiliency groups (n=17 each), displayed at a threshold of p<0.005, voxel extent = 25, coordinates in MNI space. B) Connectivity strength with right STN for ...

Relationship between Connectivity Data and Behavioral Measures

For the entire sample, right STN-right median cingulate connectivity correlated with 0-back RT (r=0.30, p=0.016), but with no other performance measures (p’s>0.169). There were no other correlations between connectivity and substance use or vulnerability measures (p’s>0.077), across the entire sample or within family history/resiliency groups; therefore, these data do not support the hypothesized relationship between connectivity and substance use despite connectivity differences by resiliency group.

DISCUSSION

Resiliency has been defined as the ability to modulate impulses, affect expression, and behavior to adapt to environment context (Eisenberg et al., 2003) and identified as protective against behavioral and substance use problems (Block et al., 1988). Its neural basis has not been explored but may help identify developmental biomarkers of risk in psychiatric disorders. As earlier behavioral research linked resiliency in early childhood with later drinking and drunkenness onset in mid-adolescence, we anticipated a similar relationship in this study. It was present: resiliency in early adolescence was associated with a later onset of drinking, fewer alcohol problems and less substance use in the transition years. These findings thus extend the continuity of this relationship from early childhood to early adulthood. Resiliency also correlated with faster reaction times and less decrement during increased cognitive demand, supporting a link between this trait and executive function. Therefore, resiliency in adolescence and drinking and drug use behaviors appear to be related though further study is needed to determine the direction of this association. Possibly those who engage in early use of substances have cognitive changes that lead to altered working memory and reduced resiliency, a trait that is highly related to executive functioning. Alternatively, those who are most resilient when entering adolescence may resist the use of substances most efficiently.

The novel finding reported is a neural link shared between resiliency and working memory. Negative relationships were found between resiliency and activity in the interconnected basal ganglia structures, the STN and pallidum, during working memory. Likely because of their size and partial volume averaging with surrounding white matter, these nuclei were not detected in first level task analyses, but were prominently localized in subsequent correlational analyses.

Investigation of task-related functional connectivity found Low and High resiliency groups differed in connectivity strength between the STN and median cingulate. The Low resiliency group had significantly more substance use, earlier onset of drinking, and more alcohol problems than the High group, as expected. However, these use and vulnerability measures were not related to connectivity; therefore it is unclear whether this functional coupling represents a pathway linking resiliency, working memory and substance use as predicted.

Resiliency and Basal Ganglia during Working Memory

Working memory is a limited capacity, constantly updated, system that temporarily maintains and stores information and interfaces thought processing, perception and action (Baddeley, 2003) and suggested as the core cognitive element of higher order regulation (Unsworth et al., 2009). The negative correlation between the regulatory trait resiliency and neural function of the STN and pallidum during working memory suggests an association with flexible adaptation of control during cognitive challenge. The STN, as a key basal ganglia structure, has reciprocal connections with the pallidum as part of the thalamocortical pathway, and influences information processing within the basal ganglia and through projections to the frontal regions (Temel et al., 2005, Aron and Poldrack, 2006). The pallidum is considered the main output structure from the limbic system (Temel et al., 2005, Zhang et al., 2005).

Although the present study found only a trend for an association between STN activation and working memory, the findings are consistent with, and add to, an emerging literature. Experimental and clinical studies have highlighted interactions between higher-order cognitive processing and the STN. Animals with STN lesions show impaired working memory (El Massioui et al., 2007), and with disconnections between the STN and prefrontal cortex show reduced accuracy, increased perseveration and slowed response (Chudasama et al., 2003). In Parkinson’s patients, deep brain stimulation of the STN has improved motor and some executive performance, including working memory, suggesting that STN stimulation was ‘releasing the brake’ on frontal function (Jahanshahi et al., 2000) and supporting a role for the STN in higher order cognitive regulation (Marceglia et al., 2011).

The STN and Substance Use

The STN’s influence may have relevance to substance use through regulation of behavior. Animals with STN lesions exhibit increased impulsive action, decreased impulsive choice (Uslaner and Robinson, 2006) and changes in motivation (Baunez et al., 2005, Winstanley et al., 2005). The lesion-induced changes increased motivation for natural (food) rewards and reduced motivation for drug (cocaine) rewards suggesting influence on incentive salience. Further, in ‘high-drinker’ rats, STN lesions enhanced motivation for alcohol but further decreased it in ‘low-drinker’ rats, suggesting a role involving motivation and individual preference (Lardeux and Baunez, 2007) with the authors proposing the STN as a target for treatment of addiction (Baunez et al., 2005). Again, Parkinson’s patients provide complementary information; STN stimulation, combined with reduction in dopaminergic treatment, has decreased pathological gambling (Ardouin et al., 2006, Bandini et al., 2007) and addiction (Witjas et al., 2005) in patients.

In the present study, there was no association found between STN or pallidum activation and levels of substance use as would be expected if resiliency were influencing these behaviors via this basal ganglia pathway. However, during this transition period into adulthood, drinking and drug use are at their highest (Substance Abuse and Mental Health Services Administration, 2006, Johnston et al., 2004) which may be overshadowing this relationship. Further studies are needed to determine who continues heightened use, who desists, the impact of liability over time, and whether the relationship between resiliency and basal ganglia activation may manifest as a protective factor.

Connectivity of the STN during Working Memory

Using PPI analysis, we found group differences in connectivity between the STN and median cingulate cortex, a default network region (Fransson and Marrelec, 2008), when contrasting a low-level (0-back) and a cognitively demanding (2-back) task. In the High resiliency group, these regions showed greater coupling in 0-back than in 2-back condition, whereas the Low resiliency group showed the opposite pattern. A recent PET study showed that stimulation of the STN decreased metabolism in a large region of the cingulate gyrus, centered in BA24 and encompassing the median cingulate region identified in the present study, substantiating a link between these regions (Le Jeune et al., 2010). Anatomically, in non-human primates, the cingulate regions of BA23 (controlling movement execution) and BA24 (controlling higher cognitive aspects of movement) both send projections to the STN providing integrated motor information to the basal ganglia (Takada et al., 2001). Consistent with this, we found that connectivity between the STN and median cingulate was related to increased processing speed during the low demand 0-back condition. As an influence on inter-regional communication efficiency, this connectivity may represent a pathway linking resiliency and flexible implementation.

Limitations

Some limitations to this study should be noted. First, PPI does not yield information regarding causal relationships. The coupling between the STN and median cingulate infers inter-regional correlations, but does not identify direction or even whether other, unmeasured regions are driving activation at both loci. Additionally, resiliency could be argued to influence motivation, instead of, or in addition to, working memory. However, previous work by our group found consistent relationships between resiliency and other measures of executive functioning including response inhibition, interference control, and planning (Martel et al., 2007), supporting a relationship with performance. In addition, we did not find differences in current levels of substance use based on family history. Some studies have found increased drinking during young adulthood in FH+ subjects (LaBrie et al., 2010, Harford et al., 1992) while others have not (Engs, 1990, Schuckit and Sweeney, 1987). As density of family history has been shown to moderate alcohol outcomes(Conway et al., 2003), the definition of liability one uses may impact findings. Familial risk is defined in this study as at least one parent with alcohol abuse or dependence, in contrast to other definitions (e.g., any biological relative with a ‘significant’ drinking problem (LaBrie et al., 2010) or graduated scales (ie. Engs, 1990, Schuckit and Sweeny, 1987 and Harford et al., 2010). Importantly, another longitudinal study, which used FH criteria similar to this study, found that despite the absences of a family history effect on drinking at baseline (mean 18.5 years), FH+ subjects were less likely to transition out of heavy drinking (Jackson et al., 2001). We expect that continued longitudinal examination of outcomes, which is underway in this sample, will reveal the family history, as well as the resiliency, groups diverging as they exit the current high-usage period. We did find a trend for earlier age of onset of drinking in the FH+ sample, suggesting they are indeed at heightened liability for poor outcome.

Conclusion

Alcohol and drug use are outcomes of decisions supported by both cognitive and behavioral functions; the young adults in our study with early high resiliency were less likely to smoke, had tried fewer illicit drugs, and had fewer alcohol problems than their less resilient counterparts independent of familial liability. The STN may represent a neural link between individual resiliency and cognitive processing, potentially influencing substance abuse risk. Our PPI analysis linking resiliency with connectivity strength between the STN and cingulate regions potentially represents efficiency of communication between salience assigned by limbic regions and flexible adaptation facilitated by cognitive neural circuits. The extensive developmental time span of the resiliency/substance involvement relationship suggests that neural connections may be present considerably earlier than observed here. Other studies will need to examine this possibility.

Supplementary Material

Supp Table S1-S2

Acknowledgments

This work was supported by NIH grants T32 AA07477 to RAZ, K01 DA020088 to MMH, R01 AA12217 and R37 AA07065 to RAZ, and the Phil F. Jenkins Foundation award to JKZ.

Footnotes

Supplementary information is available at the Alcoholism Clinical & Experimental Research website.

DISCLOSURES

The authors declare that, except for income received from our primary employer, no financial support or compensation has been received from any individual or corporate entity over the past three years for research or professional service and there are no personal financial holdings that could be perceived as constituting a potential conflict of interest, with the exceptions below:

Dr. Zubieta, Consultant, Eli Lilly Co, Merck, Johnson & Johnson.

References

  • Ardouin C, Voon V, Worbe Y, Abouazar N, Czernecki V, Hosseini H, Pelissolo A, Moro E, Lhommée E, Lang AE, Agid Y, Benabid A-L, Pollak P, Mallet L, Krack P. Pathological gambling in Parkinson’s disease improves on chronic subthalamic nucleus stimulation. Movement Disorders. 2006;21:1941–1946. [PubMed]
  • Aron AR, Poldrack RA. Cortical and subcortical contributions to Stop signal response inhibition: role of the subthalamic nucleus. J Neurosci. 2006;26:2424–2433. [PubMed]
  • Baddeley A. Working memory: looking back and looking forward. Nat Rev Neurosci. 2003;4:829–839. [PubMed]
  • Ballanger B, Eimeren Tv, Moro E, Lozano AM, Hamani C, Boulinguez P, Pellecchia G, Houle S, Poon YY, Lang AE, Strafella AP. Stimulation of the subthalamic nucleus and impulsivity: Release your horses. Annals of Neurology. 2009;66:817–824. [PMC free article] [PubMed]
  • Bandini F, Primavera A, Pizzorno M, Cocito L. Using STN DBS and medication reduction as a strategy to treat pathological gambling in Parkinson’s disease. Parkinsonism & Related Disorders. 2007;13:369–371. [PubMed]
  • Barkley RA. Behavioral inhibition, sustained attention, and executive functions: constructing a unifying theory of ADHD. Psychol Bull. 1997;121:65–94. [PubMed]
  • Baunez C, Dias C, Cador M, Amalric M. The subthalamic nucleus exerts opposite control on cocaine and ‘natural’ rewards. Nature Neuroscience. 2005;8:484–489. [PubMed]
  • Block J, Block J. Series The California Q-set: orientation and introduction. University of California; Berkeley: 1980. The California Q-set: orientation and introduction.
  • Block J, Block JH, Keyes S. Longitudinally Foretelling Drug Usage in Adolescence: Early Childhood Personality and Environmental Precursors. Child Development. 1988;59:336–355. [PubMed]
  • Bogacz R, Wagenmakers E-J, Forstmann BU, Nieuwenhuis S. The neural basis of the speed-accuracy tradeoff. Trends in Neurosciences. 2010;33:10–16. [PubMed]
  • Bonomo YA, Bowes G, Coffey C, Carlin JB, Patton GC. Teenage drinking and the onset of alcohol dependence: a cohort study over seven years. Addiction. 2004;99:1520–1528. [PubMed]
  • Botvinick MM, Braver TS, Barch DM, Carter CS, Cohen JD. Conflict monitoring and cognitive control. Psychological Review. 2001;108:624–652. [PubMed]
  • Braver TS, Cohen JD, Nystrom LE, Jonides J, Smith EE, Noll DC. A Parametric Study of Prefrontal Cortex Involvement in Human Working Memory. NeuroImage. 1997;5:49–62. [PubMed]
  • Bush G, Whalen PJ, Rosen BR, Jenike MA, McInerney SC, Rauch SL. The counting Stroop: an interference task specialized for functional neuroimaging--validation study with functional MRI. Hum Brain Mapp. 1998;6:270–282. [PubMed]
  • Callicott JH, Mattay VS, Bertolino A, Finn K, Coppola R, Frank JA, Goldberg TE, Weinberger DR. Physiological characteristics of capacity constraints in working memory as revealed by functional MRI. Cereb Cortex. 1999;9:20–26. [PubMed]
  • Caspi A, Block J, Block JH, Klopp B, Lynam D, Moffitt TE, Stouthamer-Loeber M. A “common-language” version of the California Child Q-Set for personality assessment. Psychological Assessment. 1992;4:512–523.
  • Chang C, Crottaz-Herbette S, Menon V. Temporal dynamics of basal ganglia response and connectivity during verbal working memory. NeuroImage. 2007;34:1253–1269. [PubMed]
  • Chudasama Y, Baunez C, Robbins TW. Functional Disconnection of the Medial Prefrontal Cortex and Subthalamic Nucleus in Attentional Performance: Evidence for Corticosubthalamic Interaction. J Neurosci. 2003;23:5477–5485. [PubMed]
  • Conway KP, Swendsen JD, Merikangas KR. Alcohol expectancies, alcohol consumption, and problem drinking: The moderating role of family history. Addictive Behaviors. 2003;28:823–836. [PubMed]
  • Corral M, Holguin S, Cadaveira F. Neuropsychological characteristics in children of alcoholics: familial density. Journal of Studies on Alcohol and Drugs. 1999;60:509–513. [PubMed]
  • Dick DM, Aliev F, Viken R, Kaprio J, Rose RJ. Rutgers Alcohol Problem Index Scores at Age 18 Predict Alcohol Dependence Diagnoses 7 Years Later. Alcoholism: Clinical and Experimental Research. 2011;35:1011–1014. [PMC free article] [PubMed]
  • Eisenberg N, Guthrie IK, Fabes RA, Reiser M, Murphy BC, Holgren R, Maszk P, Losoya S. The Relations of Regulation and Emotionality to Resiliency and Competent Social Functioning in Elementary School Children. Child Development. 1997;68:295–311. [PubMed]
  • Eisenberg N, Guthrie IK, Fabes RA, Shepard S, Losoya S, Murphy B, Jones S, Poulin R, Reiser M. Prediction of Elementary School Children’s Externalizing Problem Behaviors from Attentional and Behavioral Regulation and Negative Emotionality. Child Development. 2000;71:1367–1382. [PubMed]
  • Eisenberg N, Morris AS. Children’s emotion-related regulation. Adv Child Dev Behav. 2002;30:189–229. [PubMed]
  • Eisenberg N, Valient C, Fabes RA, Smith CL, Reiser M, Shepard SA, Losoya SH, Guthrie IK, Murphy BC, Cumberland AJ. The relations of effortful control and ego control to children’s resiliency and social functioning. Dev Psychol. 2003;39:761–776. [PubMed]
  • El Massioui N, Chéruel F, Faure A, Conde F. Learning and memory dissociation in rats with lesions to the subthalamic nucleus or to the dorsal striatum. Neuroscience. 2007;147:906–918. [PubMed]
  • Engs RC. Family background of alcohol abuse and its relationship to alcohol consumption among college students: an unexpected finding. J Stud Alcohol. 1990;51:542–547. [PubMed]
  • Fessler J, Lee S, Olafsson V, Shi H, Noll D. Toeplitz-based iterative image reconstruction for MRI with correction for magnetic field inhomogeneity. IEEE Transactions on Signal Processing. 2005;53:3393–3402.
  • Finn PR, Justus A, Mazas C, Steinmetz JE. Working memory, executive processes and the effects of alcohol on Go/No-Go learning: testing a model of behavioral regulation and impulsivity. Psychopharmacology (Berl) 1999;146:465–472. [PubMed]
  • Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC. Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magn Reson Med. 1995;33:636–647. [PubMed]
  • Forstmann BU, Anwander A, Schäfer A, Neumann J, Brown S, Wagenmakers E-J, Bogacz R, Turner R. Cortico-striatal connections predict control over speed and accuracy in perceptual decision making. Proceedings of the National Academy of Sciences. 2010;107:15916–15920. [PubMed]
  • Frank MJ. Hold your horses: A dynamic computational role for the subthalamic nucleus in decision making. Neural Networks. 2006;19:1120–1136. [PubMed]
  • Frank MJ, Scheres A, Sherman SJ. Understanding decision-making deficits in neurological conditions: insights from models of natural action selection. Philosophical Transactions of the Royal Society B: Biological Sciences. 2007;362:1641–1654. [PMC free article] [PubMed]
  • Fransson P, Marrelec G. The precuneus/posterior cingulate cortex plays a pivotal role in the default mode network: Evidence from a partial correlation network analysis. NeuroImage. 2008;42:1178–1184. [PubMed]
  • Friston KJ, Buechel C, Fink GR, Morris J, Rolls E, Dolan RJ. Psychophysiological and modulatory interactions in neuroimaging. Neuroimage. 1997;6:218–229. [PubMed]
  • Gitelman DR, Penny WD, Ashburner J, Friston KJ. Modeling regional and psychophysiologic interactions in fMRI: the importance of hemodynamic deconvolution. Neuroimage. 2003;19:200–207. [PubMed]
  • Glover GH, Law CS. Spiral-in/out BOLD fMRI for increased SNR and reduced susceptibility artifacts. Magn Reson Med. 2001;46:515–522. [PubMed]
  • Grant BF, Dawson DA. Age at onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: results from the national longitudinal alcohol epidemiologic survey. Journal of Substance Abuse. 1997;9:103–110. [PubMed]
  • Greicius MD, Menon V. Default-mode activity during a passive sensory task: uncoupled from deactivation but impacting activation. J Cogn Neurosci. 2004;16:1484–1492. [PubMed]
  • Hampson M, Driesen NR, Skudlarski P, Gore JC, Constable RT. Brain connectivity related to working memory performance. J Neurosci. 2006;26:13338–13343. [PMC free article] [PubMed]
  • Harford TC, Parker DA, Grant BF. Family History, Alcohol Use and Dependence Symptoms among Young Adults in the United States. Alcoholism: Clinical and Experimental Research. 1992;16:1042–1046. [PubMed]
  • Hart D, Keller M, Edelstein W, Hofmann V. Childhood personality influences on social†“cognitive development: A longitudinal study. Journal of Personality and Social Psychology. 1998;74:1278–1289. [PubMed]
  • Hingson RW, Heeren T, Winter MR. Age at Drinking Onset and Alcohol Dependence: Age at Onset, Duration, and Severity. Arch Pediatr Adolesc Med. 2006;160:739–746. [PubMed]
  • Jackson KM, Sher KJ, Gotham HJ, Wood PK. Transitioning into and out of large-effect drinking in young adulthood. Journal of Abnormal Psychology. 2001;110:378–391. [PubMed]
  • Jahanshahi M, Ardouin CMA, Brown RG, Rothwell JC, Obeso J, Albanese A, Rodriguez-Oroz MC, Moro E, Benabid AL, Pollak P, Limousin-Dowsey P. The impact of deep brain stimulation on executive function in Parkinson’s disease. Brain. 2000;123:1142–1154. [PubMed]
  • Jansma JM, Ramsey NF, Coppola R, Kahn RS. Specific versus nonspecific brain activity in a parametric N-back task. Neuroimage. 2000;12:688–697. [PubMed]
  • Johnston L, O’Malley P, Bachman J, Schulenberg JE. Series Monitoring the Future national survey results on drug use, 1975–2003. I. National Institute on Drug Abuse; Bethesda, MD: 2004. Monitoring the Future national survey results on drug use, 1975–2003. Volume I: Secondary school students. Secondary school students.
  • Krueger RF. Personality traits in late adolescence predict mental disorders in early adulthood: a prospective-epidemiological study. J Pers. 1999;67:39–65. [PubMed]
  • LaBrie JW, Migliuri S, Kenney SR, Lac A. Family history of alcohol abuse associated with problematic drinking among college students. Addictive Behaviors. 2010;35:721–725. [PMC free article] [PubMed]
  • Lardeux S, Baunez C. Alcohol Preference Influences the Subthalamic Nucleus Control on Motivation for Alcohol in Rats. Neuropsychopharmacology. 2007;33:634–642. [PubMed]
  • Le Jeune F, Péron J, Grandjean D, Drapier S, Haegelen C, Garin E, Millet B, Vérin M. Subthalamic nucleus stimulation affects limbic and associative circuits: a PET study. European Journal of Nuclear Medicine and Molecular Imaging. 2010;37:1512–1520. [PubMed]
  • Maldjian JA, Laurienti PJ, Kraft RA, Burdette JH. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. NeuroImage. 2003;19:1233–1239. [PubMed]
  • Marceglia S, Fumagalli M, Priori A. What neurophysiological recordings tell us about cognitive and behavioral functions of the human subthalamic nucleus. Expert Review of Neurotherapeutics. 2011;11:139–149. [PubMed]
  • Martel MM, Nigg JT, Wong MM, Fitzgerald HE, Jester JM, Puttler LI, Glass JM, Adams KM, Zucker RA. Childhood and adolescent resiliency, regulation, and executive functioning in relation to adolescent problems and competence in a high-risk sample. Dev Psychopathol. 2007;19:541–563. [PubMed]
  • Martel MM, Pierce L, Nigg JT, Jester JM, Adams K, Puttler LI, Buu A, Fitzgerald H, Zucker RA. Temperament pathways to childhood disruptive behavior and adolescent substance abuse: testing a cascade model. J Abnorm Child Psychol. 2009;37:363–373. [PMC free article] [PubMed]
  • National Institute on Alcohol Abuse and Alcoholism. NIAAA, Tenth Special Report to the US Congress on Alcohol and Health: Highlights from current research, NIAAA, Tenth Special Report to the US Congress on Alcohol and Health: Highlights from current research. U.S. Department of Health and Human Services; Bethesda, MD: 2000. Alcohol involvement over the life course; pp. 28–53.
  • Nigg JT. On inhibition/disinhibition in developmental psychopathology: views from cognitive and personality psychology and a working inhibition taxonomy. Psychol Bull. 2000;126:220–246. [PubMed]
  • O’Reilly RC, Frank MJ. Making Working Memory Work: A Computational Model of Learning in the Prefrontal Cortex and Basal Ganglia. Neural Computation. 2006;18:283–328. [PubMed]
  • Owen AM, McMillan KM, Laird AR, Bullmore E. N-back working memory paradigm: A meta-analysis of normative functional neuroimaging studies. Human Brain Mapping. 2005;25:46–59. [PubMed]
  • Ozkaragoz T, Satz P, Noble EP. Neuropsychological functioning in sons of active alcoholic, recovering alcoholic, and social drinking fathers. Alcohol. 1997;14:31–37. [PubMed]
  • Pennington BF, Ozonoff S. Executive functions and developmental psychopathology. J Child Psychol Psychiatry. 1996;37:51–87. [PubMed]
  • Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL. A default mode of brain function. Proceedings of the National Academy of Sciences of the United States of America. 2001;98:676–682. [PubMed]
  • Rothbart MK, Ahadi SA, Evans DE. Temperament and personality: origins and outcomes. J Pers Soc Psychol. 2000;78:122–135. [PubMed]
  • Schuckit MA, Sweeney S. Substance use and mental health problems among sons of alcoholics and controls. J Stud Alcohol. 1987;48:528–534. [PubMed]
  • Shedler J, Block J. Adolescent drug use and psychological health: A longitudinal inquiry. American Psychologist. 1990;45:612–630. [PubMed]
  • Smith EE, Jonides J. Working memory: a view from neuroimaging. Cognit Psychol. 1997;33:5– 42. [PubMed]
  • Spadoni AD, Norman AL, Schweinsburg AD, Tapert SF. Effects of family history of alcohol use disorders on spatial working memory BOLD response in adolescents. Alcohol Clin Exp Res. 2008;32:1135–1145. [PMC free article] [PubMed]
  • Substance Abuse and Mental Health Services Administration. Series Results From the 2005 National Survey on Drug Use and Health: National Findings. NSDUH Series H-30, DHHS Pub. No. SMA 06–4194. SAMHSA, Office of Applied Studies; Rockville MD: 2006. Results From the 2005 National Survey on Drug Use and Health: National Findings. NSDUH Series H-30, DHHS Pub. No. SMA 06–4194.
  • Sutton BP, Noll DC, Fessler JA. Fast, iterative image reconstruction for MRI in the presence of field inhomogeneities. IEEE Trans Med Imaging. 2003;22:178–188. [PubMed]
  • Sweet LH, Paskavitz JF, Haley AP, Gunstad JJ, Mulligan RC, Nyalakanti PK, Cohen RA. Imaging phonological similarity effects on verbal working memory. Neuropsychologia. 2008;46:1114–1123. [PubMed]
  • Takada M, Tokuno H, Hamada I, Inase M, Ito Y, Imanishi M, Hasegawa N, Akazawa T, Hatanaka N, Nambu A. Organization of inputs from cingulate motor areas to basal ganglia in macaque monkey. European Journal of Neuroscience. 2001;14:1633–1650. [PubMed]
  • Temel Y, Blokland A, Steinbusch HWM, Visser-Vandewalle V. The functional role of the subthalamic nucleus in cognitive and limbic circuits. Progress in Neurobiology. 2005;76:393–413. [PubMed]
  • Unsworth N, Redick TS, Heitz RP, Broadway JM, Engle RW. Complex working memory span tasks and higher-order cognition: A latent-variable analysis of the relationship between processing and storage. Memory. 2009;17:635– 654. [PubMed]
  • Uslaner JM, Robinson TE. Subthalamic nucleus lesions increase impulsive action and decrease impulsive choice − mediation by enhanced incentive motivation? European Journal of Neuroscience. 2006;24:2345–2354. [PubMed]
  • Viner RM, Taylor B. Adult outcomes of binge drinking in adolescence: findings from a UK national birth cohort. Journal of Epidemiology and Community Health. 2007;61:902–907. [PMC free article] [PubMed]
  • Winstanley CA, Baunez C, Theobald DE, Robbins TW. Lesions to the subthalamic nucleus decrease impulsive choice but impair autoshaping in rats: the importance of the basal ganglia in Pavlovian conditioning and impulse control. Eur J Neurosci. 2005;21:3107–3116. [PubMed]
  • Witjas T, Baunez C, Henry JM, Delfini M, Regis J, Cherif AA, Peragut JC, Azulay JP. Addiction in Parkinson’s disease: Impact of subthalamic nucleus deep brain stimulation. Movement Disorders. 2005;20:1052–1055. [PubMed]
  • Wong MM, Nigg JT, Zucker RA, Puttler LI, Fitzgerald HE, Jester JM, Glass JM, Adams K. Behavioral Control and Resiliency in the Onset of Alcohol and Illicit Drug Use: A Prospective Study From Preschool to Adolescence. Child Development. 2006;77:1016–1033. [PMC free article] [PubMed]
  • Yeung N, Botvinick MM, Cohen JD. The Neural Basis of Error Detection: Conflict Monitoring and the Error-Related Negativity. Psychological Review. 2004;111:931–959. [PubMed]
  • Zhang Y, Bailey KR, Toupin MM, Mair RG. Involvement of ventral pallidum in prefrontal cortex-dependent aspects of spatial working memory. Behav Neurosci. 2005;119:399–409. [PubMed]
  • Zucker RA, Ellis DA, Fitzgerald HE, Bingham CR, Sanford K. Other evidence for at least two alcoholisms II: Life course variation in antisociality and heterogeneity of alcoholic outcome. Development and Psychopathology. 1996;8:831–848.
  • Zucker RA, Fitzgerald HE, Refior SK, Puttler LI, Pallas DM, Ellis DA. The clinical and social ecology of childhood for children of alcoholics: Description of a study and implications for a differentiated social policy. In: Fitzgerald HE, Lester BM, Zuckerman BS, editors. Children of addiction: Research, Health and Policy Issues., Children of addiction: Research, Health and Policy Issues. Routledge Falmer; New York: 2000. pp. 109–141.