Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Psychiatry Research: Neuroimaging journal homepage: www.elsevier.com/locate/psychresns

Adolescent risk-taking and resting state functional connectivity Samuel J. DeWitt a, Sina Aslan a,b, Francesca M. Filbey a,n a b

Center for BrainHealth, School of Behavioral and Brain Sciences, University of Texas at Dallas, Dallas, TX, USA Advance MRI, LLC, Frisco, TX, USA

art ic l e i nf o

a b s t r a c t

Article history: Received 30 May 2013 Received in revised form 25 February 2014 Accepted 26 March 2014

The existing literature on the role of emotion regulation circuits (amygdala-prefrontal cortex) in the adolescent brain yields mixed results, particularly on the role of these regions in the context of reward sensitivity and risk-taking behavior sensitivity and risk-taking behavior. Here, we examined functional connectivity in the resting state in 18 risk-taking (RT) adolescents compared with 18 non-risk-taking (NRT) adolescents as defined by the Youth Risk Behavior Surveillance Survey. Separate seed-based correlations with bilateral amygdala and bilateral nucleus accumbens used as the seed were performed to determine functional connectivity using functional magnetic resonance imaging (fMRI). The results showed greater connectivity between the amygdala (seed region) and the right middle frontal gyrus, left cingulate gyrus, left precuneus and right inferior parietal lobule in RT adolescents than in NRT adolescents. Likewise, there was greater connectivity between the nucleus accumbens (seed region) and the right middle frontal gyrus in RT adolescents compared with NRT adolescents. These findings suggest that risk-taking behavior in adolescents is associated with hyperconnectivity during the resting state in networks associated with emotion regulation, reward sensitivity, executive control, and the default mode. & 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: Amygdala Emotion regulation Prefrontal cortex Resting state functional connectivity

1. Introduction Risk-taking behavior is a growing concern in today's adolescent population. The Youth Risk Behavior Surveillance Survey of high school students in 2011 reported that in the 30 days preceding the survey, 38.7% drank alcohol, and 23.1% used marijuana (Eaton et al., 2012). The survey further revealed that 15.3% of adolescents had four or more sexual partners in their lifetime, and 32.8% had been in a physical fight in the last 12 months (Eaton et al., 2012). Emerging evidence in the neuroimaging literature suggests that a combination of developing brain regions and aberrant network connectivity among these regions may be moderating adolescent decision-making and risk-taking (Wetherill et al., 2012). One such brain network is the emotion-regulation network (Perlman et al., 2012). Emotion regulation is commonly thought to consist of a regulatory feedback loop whereby limbic regions (such as the amygdala) provide input to prefrontal cortex (PFC) regions (including the dorsolateral PFC in adults (Staudinger et al., 2011) and the ventrolateral PFC in children and young adults (McRae et al., 2012)), which in turn provide reciprocal input back to limbic regions allowing for regulation of emotional reactivity, i.e., emotion regulation (Levesque et al., 2004). Disruptions in brain regions

n Correspondence to: 2200 West Mockingbird Lane, Dallas, TX 75235, USA. Tel.: þ 1 972 883 3311. E-mail address: Francesca.fi[email protected] (F.M. Filbey).

within the regulatory loop can lead to antisocial and risk-seeking behaviors. For example, Joseph et al. (2009) showed that high sensation-seeking adults exhibited greater neural activity in PFC regions related to processing emotion including in the insula and medial orbital frontal cortex (OFC) as compared with participants defined as low sensation-seekers. Emotion regulation as it relates to risk-taking behavior is shown in adults to consist of cognitive control over emotional responses with the net effect of a reduction in reward-circuitry activation (Martin and Delgado, 2011). The behavioral findings on the association between emotion regulation and risk-taking in adolescents suggest that the ability to make effective decisions in the context of risk may be affected by states of heightened arousal (Rivers et al., 2008; Spear, 2009). Studies that have examined emotional response and regulation in adolescents reveal behavioral and neural patterns distinct from those of children and adults in limbic, PFC, and reward-circuitry regions (Ernst et al., 2005; Galvan et al., 2006; Guyer et al., 2008). Many researchers have used functional connectivity analysis to investigate the relationship among regions involved in emotion regulation. Functional connectivity is defined as a temporal correlation of a neurophysiological index measured in brain areas, such as the blood-oxygen-level-dependent (BOLD) signal (Biswal et al., 1997). The relationship between neuronal activation patterns of differing brain regions is described by the level of functional connectivity between the regions (van den Heuvel and Hulshoff Pol, 2010). Functional networks among brain regions observed

http://dx.doi.org/10.1016/j.pscychresns.2014.03.009 0925-4927/& 2014 Elsevier Ireland Ltd. All rights reserved.

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

2

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

during rest (i.e., the resting state) are shown to correspond closely to the networks of those regions when a task is carried out (Smith et al., 2010; Kannurpatti et al., 2012). Such observations suggest that knowledge of the intrinsic organization of brain regions informs neuronal activation patterns when the brain is engaged in a task. Studies that look at connectivity and activation patterns in regions related to emotion regulation in adolescents have not denoted a clearly defined pattern of connectivity. Resting-state connectivity in regions related to emotion regulation including the amygdala and ventromedial PFC (vmPFC) has been shown to be significantly weaker in children compared with adults (Qin et al., 2012). However, in a recent study in which participants were shown emotional faces (happy vs. scared), functional connectivity observed from childhood through early adulthood showed a switch from positive amygdala-PFC connectivity to negative connectivity around the age of 10 (Gee et al., 2013). Thus far, most existing studies that have examined functional connectivity in adolescents have focused on task-dependent responses, with few evaluating its relationship to risky behavior. Some findings have shown a reduction in connectivity in areas of emotion regulation related to risky behavior for adolescents engaged in emotion-related tasks (Hare et al., 2008), particularly in adolescents with clinical disorders associated with risky behavior. One such study examined functional and structural amygdala-PFC connectivity in youth with conduct disorder and psychopathic traits (Marsh et al., 2011). Findings from this study showed reduced functional connectivity between the amygdala and the orbitofrontal cortex (OFC) during a task where participants were asked to make moral judgments. Interestingly, a similar amygdala-OFC disconnection has been observed in adults with borderline personality disorder and is suggested to be related to an inability of the OFC to downregulate amygdalar response to aversive stimuli (New et al., 2007). Reduced functional connectivity between the amygdala and the rostral anterior cingulate during a learning task was also observed in youth with conduct disorder (Finger et al., 2012). Other studies, however, found increased activation in fronto-limbic regions to be associated with risk. For example, Silveri et al. (2011) found enhanced activation in subjects performing a visual Stroop Task in the ventral cingulate cortex (connected to the amygdala) and the middle frontal gyrus in adolescents who had a family history of substance abuse compared with adolescents who did not. The authors suggested that enhanced activation may be the result of neuronal inefficiency, such as a disproportionate amount of activation for a given task in these regions. Taken together, these studies suggest that aberrant functional connectivity may be a hallmark of behavioral problems/disorders, particularly in adolescents. However, a clear pattern of functional connectivity associated with risk-taking behavior needs to be established. In the present study, we used resting state functional connectivity to assess the regions responsible for emotion regulation in a risk-taking adolescent population. Given the role of reward-circuitry/reward sensitivity in adolescent risk-taking behavior, we examined these regions as well. The focus of our investigation was on intrinsic organization related to emotion regulation and reward sensitivity. Thus, we determined resting state functional connectivity with the amygdale (i.e., as the seed region) given its role in emotion. Previous studies in healthy adults have suggested that the amygdala (including its subdivisions) has distinct functional connectivity patterns throughout the brain, which highlight its diffused role in emotional input (Roy et al., 2009). Investigations of resting state functional connectivity between the amygdala and regions of the PFC have revealed aberrant connectivity patterns in adult with posttraumatic stress disorder (Brown et al., 2014) and major depressive disorder (Yue et al., 2013), as well as in adolescents

suffering from depression (Connolly et al., 2013). Studies have shown a reduction in task-based functional connectivity between the amygdala and the PFC for adolescents with clinical diagnoses such as conduct disorder. However, the focus of our study is in developing adolescents who engage in risk-taking behavior. Specifically, those who do not yet meet criteria for any disorders such as substance use problems and have no history of psychopathology/behavioral disorders. This allows for the possibility to determine potential pre-morbid processes that may reflect vulnerability to later pathology. Silveri et al. (2011) showed increased frontolimbic activation in those with an increased risk for substance abuse (i.e., family history). Similarly, we also expect that resting state functional connectivity between the amygdala and PFC areas in risk-taking (RT) adolescents to be greater compared with that in non-risk-taking (NRT) adolescents. This would suggest that altered resting state functional connectivity associated with early risktaking is a vulnerability factor that preceded problematic behavior such as substance dependence. To investigate the role of reward sensitivity/reward-seeking behavior, we determined resting state functional connectivity with the nucleus accumbens, a key region of the ventral striatum involved in the brain's reward circuitry. Given the extant literature, which describes a hyperactive rewardcircuitry as a hallmark of the adolescent brain, we likewise expect resting state functional connectivity between the nucleus accumbens and PFC areas in RT adolescents to be greater than that in NRT adolescents. Elucidating these neural mechanisms as they relate to adolescent risk could lead to a better understanding of the neurocognitive profile and mechanisms of risk-taking, which can inform risk-prevention strategies for adolescents.

2. Methods Written parental informed consent and assent were obtained from all subjects in accordance with the Institutional Review Board (IRB) of our academic institutions, The University of Texas at Dallas and The University of Texas Southwestern Medical Center at Dallas.

2.1. Participants Eighteen RT participants were age- and sex-matched to 18 NRT participants (12–17 years old). Risk-taking behavior was assessed using the Youth Risk Behavior Surveillance Survey (YRBSS) and included items related to sexual activity, substance use, or violent behavior (Eaton et al., 2012). Those who endorsed any risktaking item were categorized in the RT group. Those in the NRT group did not endorse any risk-taking item. The eligibility requirements included the following: aged 12–17 years, right-hand-dominant, no history of brain injury or brain-related illness, no MRI contraindications (e.g., metal implants, claustrophobia, pregnancy), no Axis I disorders, and no current use of psychoactive medications. Adolescents who reported current illicit drug use other than marijuana in the last 60 days were excluded to control for any observable effects of illicit drugs on brain activation. Two members of the RT group reported regular alcohol use at a rate of once per week for no more than a year. One member of the RT group reported regular marijuana at this same level of frequency/duration. Participants were screened for psychiatric disorders using the Kiddie Schedule for Affective Disorders and Schizophrenia-Present and Lifetime (KSADS-PL) diagnostic (Kaufman et al., 1997) interview, and substance use history was obtained via a substance use history questionnaire. Both groups were matched on age (mean age per group: 14 years) and gender (8 males per group). The two groups were significantly different in mean income levels, with the NRT group having significantly higher income (p o 0.05) (Table 1). Table 1 Subject characteristics and household income, mean (SD).

NRT Group RT Group

N

Age

Gender (M/F)

Household income

18 18

14.0 (1.7) 14.0 (1.6)

10/8 10/8

$94,600 ($39,300)n $66,235 ($29,637)n

n Significant difference at p o 0.05, missing data from 3 in NRT group, 1 in RT group.

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 2.2. MRI acquisitions MRI scans were collected on a 3T MR system (Philips Medical Systems, Best, The Netherlands) using body coil transmission and head coil reception. Resting state fMRI scans were collected using the following sequence parameters: field of view¼ 220  220, matrix ¼ 64  64, slice thickness¼3.5 mm, voxel size ¼ 3.44  3.44  4 mm3, 39 axial slices, repetition time (TR)/echo time (TE) ¼2000/29 ms, flip angle ¼ 751, 150 image volumes, and scan duration¼5 min. A high resolution T1-weighted image was also acquired as an anatomical reference with the following parameters: Magnetization Prepared Rapid Acquisition of Gradient Echo (MPRAGE) sequence, TR/TE ¼ 8.3/3.8 ms, shot interval ¼ 2100 ms, inversion time ¼ 1100 ms, flip angle ¼121, 160 sagittal slices, voxel size¼ 1  1  1 mm3, field of view¼ 256  256  160 mm3, and duration¼ 4 min.

2.3. MRI analysis Resting state fMRI was analyzed using Analysis of Functional NeuroImages (AFNI) (NIMH Scientific and Statistical Computing Core, Bethesda, MD, USA). The images were preprocessed with motion correction, slice timing correction, removal of the linear trend, spatial normalization to standard Talairach space (matrix ¼ 61  73  61, resolution ¼3  3  3 mm3), and spatially smoothed with a full-width-halfmaximum (FWHM) of 6 mm. Next a band-pass filtering (0.01–0.1 Hz) was applied to the preprocessed signal time course on a voxel-by-voxel basis to keep only lowfrequency fluctuations. Also, white matter and cerebrospinal fluid (CSF) signals were regressed out using averaged signals from the white matter and the ventricles from each voxel time series. Functional connectivity was measured using a seed-based approach by choosing bilateral amygdala. The amygdala seed was defined as a cubical region with the center voxel at 7 23,  6,  16 per Talairach coordinates bilaterally. Each cube consisted of 27 voxels (3  3  3) with a total volume of 729 mm3. Functional connectivity analysis was also measured using bilateral nucleus accumbens as the seed. The seed was a cubical region with the center voxel at 712,  8,  8 (Talairach x,y,z) for the left and right regions. The cross-correlation coefficients (cc) between the seed regions and all other voxels were calculated to generate a correlational map. Then, a Fisher z-transform was performed to convert the cc maps to z-maps to perform voxel-based analysis (VBA). For VBA, the effective smoothness was calculated to identify the cluster threshold. For cluster extent inference, we applied a family-wise error correction (FWE-corrected) by using 3dClustSim in AFNI (NIMH Scientific and Statistical Computing Core, Bethesda, MD, USA), which estimates the probability of false positive activation clusters throughout the whole brain based on two criteria: smoothness of the voxel map and cluster-defining threshold. We estimated the smoothness to be 10-mm FWHM (inherent smoothness plus additional smoothness applied) and set the clusterdefining threshold to the 99.5th percentile of t-statistic distribution. A FWEcorrected significance level of 0.05 yielded a required minimum cluster size of 79 voxels (2133 mm3) (Cox, 1996). In order to investigate the relationship between resting state functional connectivity and behavioral measures, Pearson's correlations were performed per

3

group between regional z-scores of connectivity and the reward dependence, harm avoidance and novelty seeking subscales of the Junior Temperament and Character Inventory (JTCI) (Luby et al., 1999), and the Impulsivity and Sensation Seeking (ImpSS) scale (Zuckerman, 1996) as well as the YRBSS.

3. Results 3.1. MRI measurements 3.1.1. Main effects Table 2 summarizes the one-sample t-test of resting state functional connectivity between the amygdala seed and several regions including the following: positive correlations for the NRT group with the right parahippocampal gyrus, left middle frontal gyrus and left inferior frontal gyrus, and negative correlations with the right middle frontal gyrus/superior frontal gyrus. For the RT group, significant positive clusters included the right parahippocampal gyrus, left precuneus and bilateral middle frontal gyrus (p o0.05, FWE-corrected, cluster Z459 mm3). Table 3 shows significant regions using the nucleus accumbens as the seed region that include bilateral middle temporal gyrus and left inferior frontal gyrus for the NRT group and several regions for the RT group, including the left middle frontal gyrus and bilateral superior frontal gyrus (po 0.05, FWE-corrected, cluster Z459 mm3).

3.1.2. Group differences Voxel based comparisons between the two groups revealed greater connectivity in the RT group compared with the NRT group between the amygdala and the following regions: left cingulate gyrus, left precuneus, right middle frontal gyrus (MFG) and right inferior parietal lobule (po0.05, FWE-corrected, cluster Z2133 mm3) (Fig. 1). The reverse contrast (i.e., NRT4RT) did not yield any significant results. Voxel based comparisons between the two groups using the nucleus accumbens as the seed region revealed greater connectivity in the RT group compared with the NRT group in the right MFG (po0.05, FWE-corrected, cluster Z2133 mm3) (Fig. 2). These findings demonstrate heightened resting state functional connectivity in several key regions related to emotion regulation and reward sensitivity for RT adolescents compared with NRT adolescents (see Table 4).

Table 2 Clusters that showed significant positive and negative correlations with the amygdala in RT and NRT groups. Brain regions

NRT group Positive correlation R. Parahippocampal gyrus/R. Amygdala L. Middle frontal gyrus/L. Precentral gyrus R. Cingulate gyrus R. Superior parietal lobule L. Middle frontal gyrus/L. Inferior frontal gyrus L. Postcentral gyrus/L. Precentral gyrus L. Superior frontal gyrus/L. Medial frontal gyrus R. Inferior temporal gyrus/R. Fusiform gyrus Negative correlation R. Middle frontal gyrus/R. Superior frontal gyrus R. Middle frontal gyrus RT group Positive correlation R. Parahippocampal gyrus/R. Amygdala L. Precuneus L. Middle frontal gyrus/Superior frontal gyrus R. Middle frontal gyrus/R. Inferior frontal gyrus

Cluster (mm3)

BA

TLRC X

Y

Z

z-Score

437,811 1,053 1,026 1,026 756 729 648 459

34 9 32 7 46 3 8 20

24  48 9 24  48  42  12 57

6 6 9  63 33  15 27  57

 15 39 39 60 18 57 48  15

35.64 5.07 4.80 4.83 4.66 4.82 5.00 4.39

648 486

9 9

45 45

42 30

33 33

 4.75  4.52

547,479 1,296 837 459

34 7 6 9

24 3 48 60

6  63  63  15

 15 60 39 36

38.77 4.82 4.88 4.67

po 0.05 (FWE-corrected) with cluster Z 459 mm3. TLRC—Talairach coordinates, BA—Brodmann area, L/R—Left/Right.

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4

Table 3 Clusters that showed significant positive correlations with the nucleus accumbens in RT and NRT groups. Brain regions

NRT Group Positive correlation L. Subcallosal gyrus/L. Lentiform nucleus R. Middle temporal gyrus L. Middle temporal gyrus L. Inferior frontal gyrus/L. Middle frontal gyrus L. Middle frontal gyrus/L. Inferior frontal gyrus RT Group Positive correlation L. Subcallosal gyrus/L. Lentiform nucleus L. Paracentral lobule R. Superior frontal gyrus L. Inferior temporal gyrus L. Middle frontal gyrus/L. Superior frontal gyrus L. Middle temporal gyrus/L. Inferior temporal gyrus L. Lingual gyrus R. Superior frontal gyrus L. Precentral gyrus/L. Inferior frontal gyrus L. Middle frontal gyrus/L. Superior frontal gyrus

Cluster (mm3)

BA

TLRC X

Y

Z

z-Score

207,657 2,862 1,458 702 756

34 21 39 47 46

 12 72  42  42  48

9  39  69 36 33

9  12 27 0 18

35.51 5.47 4.92 7.29 4.66

233,172 7,290 5,292 2,268 1,080 1,053 837 810 729 513

34 4 6 20 8 21 18 6 44 10

 15 3 3  57  21  63 3 6  57  33

6  36 3  42 36  48  93 30 12 48

 12 72 69  21 39 6  18 57 6 18

28.27 7.19 6.41 5.68 5.80 7.21 6.17 5.90 4.74 4.75

po 0.05 (FWE-corrected) with cluster Z 459 mm3. TLRC—Talairach coordinates, BA—Brodmann area, L/R—Left/Right.

Fig. 1. Results of VBA analysis of functional connectivity with amygdala as seed is superimposed on a T1 structural image. (A) The RT group showed higher connectivity to the amygdala seed in left cingulate gyrus, left precuneus, right middle frontal gyrus (MFG) and right inferior parietal lobule (IPL) compared to the NRT group, p o 0.05 (FWEcorrected) with cluster Z2133 mm3. (B) Average z-scores per group are displayed in the bar plots for each region.

Because of possible micro-movement effects on these findings, we compared translational and rotational movements between groups and found no significant differences (p40.05). Additionally, we used the “scrubbing” methods introduced by Power et al. (2012). With the same framewise displacement (FD) threshold of 0.50 mm and the same DVARS threshold of 50 a.u., none of our participants needed to be removed. We ran our resting state functional connectivity analysis again with the frames above threshold removed for each participant. The results using the amygdala seed showed that three out of the original four regions (right inferior parietal lobule, left precuneus and right middle frontal gyrus) were significant at po0.005 for RT4NRT. The reverse contrast NRT4RT, as well as both contrasts using the nucleus accumbens as a seed region, did not

reveal any significant results. The remaining analyses and values shown in images and tables reflect results obtained with all 150 frames per participant included. 3.2. Behavioral measures Pearson's correlations per group between connectivity and behavioral measures did not yield any significant results for either the amygdala seed or the nucleus accumbens seed. When both groups were examined together, however, a positive correlation was found between right MFG connectivity to nucleus accumbens and scores on the novelty-seeking subscale of the JTCI (r¼0.43, p¼0.008), indicating higher connectivity of right MFG to the ventral

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

5

Fig. 2. Results of VBA analysis of functional connectivity with the nucleus accumbens as the seed region are superimposed on a T1 structural image. (A) The RT group showed higher connectivity to the nucleus accumbens region in the right middle frontal gyrus (MFG) compared with the NRT group, p o 0.05 (FWE-corrected) with cluster Z 2133 mm3. (B) Average z-scores per group are displayed in the bar plot for right MFG.

striatum was associated with higher scores of novelty-seeking. Post hoc t-tests revealed that the ImpSS scores were significantly higher for the RT group (mean¼9.33) compared with the NRT group (mean¼ 6.39) (t(34)¼2.45, p¼0.02). The novelty-seeking subscale score from the JTCI was significantly higher for the RT group (mean¼ 8.89) compared with the NRT group (mean¼3.11) (t(34)¼ 5.11, po0.001). No significant difference between the two groups was found for the reward-seeking or harm-avoidance subscales of the JTCI. 3.3. Post hoc analysis Demographic results revealed a significant group difference based on income, with NRT adolescents having significantly higher income compared with RT adolescents. However, income data were not available for three members of the NRT group and one member of the RT group. For the subset of participants for which we had income data (NRT (n ¼ 14), RT (n ¼17)), we added income as a covariate to our original functional connectivity analyses. The results using the amygdala seed region showed three of the original four regions from the group contrast RT4 NRT remained, with right middle frontal gyrus being the only one that was no longer significant. However, by lowering the cluster threshold slightly, this region becomes visible in the contrast. The reverse contrast, NRT4RT still yielded no significant results. The results using the nucleus accumbens seed region did not change with the addition of the income covariate. In order to determine how connectivity was associated with age, we correlated the results of our connectivity analyses with age for each group. The only significant correlations with age were seen in the NRT group. z-Scores of connectivity to the nucleus accumbens yielded positive correlations for left middle temporal gyrus (r ¼ 0.52, p¼ 0.027) and left inferior frontal gyrus/middle frontal gyrus (r ¼0.52, p¼ 0.027). For our RT group, we ran Pearson correlations on age and number of risk items endorsed on the YRBSS, sum on the IMPSS, and sum on the three subsets of the JTCI

Table 4 Clusters of positive correlations with amygdala and nucleus accumbens (seed regions) where RT 4NRT. Brain regions

Amygdala seed R. inferior parietal lobule L. Precuneus L. Cingulate gyrus R. Middle frontal gyrus Nucleus accumbens seed R. Middle frontal gyrus

Cluster (mm3)

BA

TLRC X

Y

Z

z-Score

6561 3591 2700 2214

40 7 31 10

48 3 6 24

 54  60  30 66

39 42 33 6

4.79 4.56 4.38 4.75

2268

10

39

39

12

4.44

p o0.05 (FWE-Corrected) with cluster Z 459 mm3. TLRC—Talairach coordinates, BA —Brodmann area, L/R ¼Left/Right.

(novelty seeking, reward dependence and harm avoidance). The results showed a significant positive correlation of age to items endorsed on the YRBSS only (r ¼0.68, p ¼0.002).

4. Discussion While Silveri et al. (2011) showed similar activation patterns for adolescents at increased risk for substance abuse based on family history, to our knowledge, the present findings are the first to show enhanced resting state functional connectivity patterns (vs. task- induced connectivity patterns) in adolescents who engage in risk-taking behavior (vs. having familial risk). Although, there are few previous studies of intrinsic functional connectivity in adolescents, our findings are concordant with previous research in adult substance users (with a mean age of 31, SD of 7) showing increased intrinsic amygdala connectivity, specifically with the precuneus (Xie et al., 2011). Heightened resting state functional connectivity of frontoparietal regions has also been observed in

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

6

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

adult cigarette smokers relative to healthy controls (Janes et al., 2010). As our data focused on connectivity during the resting state, our results demonstrate that areas underlying emotion regulation and reward sensitivity are in a state of heightened connectivity in the absence of any goal-directed behavior in RT adolescents compared with NRT adolescents. It may be that heightened resting state functional connectivity is associated with hyperactivation in these regions when adolescents are engaged in a task. Hyperactivation of brain regions, specifically the striatum, has been considered a major contributor to adolescent risk-taking behavior (for review, see Galvan, 2010). Right MFG showed increased functional connectivity with both the amygdala and nucleus accumbens in our study. Increased MFG activation is shown in adolescents who meet criteria for video game addiction when they are shown visual stimuli of video games (Han et al., 2012). Furthermore, hyperactivation of regions involved in emotion regulation has also been implicated in clinical disorders seen in adolescents. Increased connectivity of the amygdala with the left cingulate gyrus, right precuneus, and MFG is seen in adolescents with pediatric bipolar disorder (Pavuluri et al., 2009). Pavuluri et al. (2009) suggested that hyperactivity of this amygdala network is related to “intense automatic emotional reactivity” and overall heightened limbic activation in adolescents with bipolar disorder. While there is support for the idea of concordance in both resting and task-induced functional connectivity (i.e., heightened resting state functional connectivity ¼heightened task-induced activation), others suggest an opposing association between the two types of functional connectivity. For example, the strength model of self-regulation (Baumeister and Heatherton, 1993) states that there are limited neural resources for regulation and that active regulation in one area may deplete regulation in another area. This idea is supported in the self-regulation literature (Gailliot et al., 2007; Hagger et al., 2010). In the context of the present findings, it is possible that the heightened connectivity between emotion regulation and reward-sensitivity regions in risk-taking adolescents may deplete the resources of these regions. A recent study in which adult participants engaged in a modified monetary incentive delay task, which included an emotionregulation component, showed that an increase in dorsolateral PFC (DLPFC) activation attenuated reward encoding in the left putamen (Staudinger et al., 2011). If regions of the PFC are engaged in heightened emotion-regulatory processes at rest due to higher amygdala connectivity as seen in our study, the strength model would suggest that this investment of neural processing may lead to depletion in other processes such as the region's ability to effectively regulate reward processes in risk-taking adolescents. Neuronal processing in these regions has been observed to be unique among adolescents as compared with children and adults (Steinberg, 2004). Ernst's triadic model of the neurobiology of motivated behavior in adolescents implicates heightened rewardseeking via the nucleus accumbens, and a reduction in harm avoidance via the amygdala and a supervisory role via the PFC (Ernst et al., 2006). In accordance with this idea, we found that RT adolescents had more widespread connectivity between the ventral striatum (reward-seeking hub) and the PFC whereas NRT adolescents exhibited more widespread connectivity between the amygdala (harm-avoidant hub) and the PFC (specifically the right MFG). However, the strength of the positive correlations in these regions to both seeds was stronger for RT adolescents in group contrasts. In the context of Baumeister's strength model, higher resting state functional connectivity for the RT group between the supervisory hub (PFC) and the reward-seeking and harm-avoidance hubs simultaneously may lead to a depletion of the PFC's ability to optimally “supervise” an emotional response in the context of motivational behavior (Ernst et al., 2006). Furthermore, Gee et al. (2013) suggested

that the positive to negative switch observed around age 10 suggests a switch from bottom-up (amygdala-controlled) to top-down (PFCcontrolled) processing. Our findings of sustained, positively increased amygdala-PFC connectivity in RT adolescents suggest their patterns are more akin to those of younger children, which may explain their sub-optimal emotional processing. Hypoactivation of emotion-regulation regions, specifically the amygdala, is seen in male adolescents with antisocial substance disorder who are asked to engage in a risk-taking monetary task (Crowley et al., 2010). Similarly, decreases in activation of the MFG in response to negative affective pictures are seen in adolescent males with a history of violent behavior compared with controls (Qiao et al., 2012). Higher connectivity of these emotion-regulation regions under resting conditions in risk-taking adolescents may be the source of neural depletion, resulting in hypoactivation in these regions during task-related behaviors, especially in the context of risk-taking. Our findings also showed increased connectivity between the amygdala and the right inferior parietal lobe (IPL). The IPL and the DLPFC are key regions in the executive network. As outlined by Petersen and Posner (Posner and Petersen, 1990; Petersen and Posner, 2012) and supported in task paradigms and computational modeling (Dosenbach et al., 2008, 2007, 2006), a dual network view of the executive network exists. In this view, cinguloopercular control plays a role in task maintenance, whereas the frontoparietal system seems to be implicated in task initiation/ switching. Importantly, these computational models show that connectivity between these regions is distinct and maintained at rest. (Dosenbach et al., 2008). Development of this executive network occurs throughout childhood and into early adulthood (Posner and Rothbart, 2007; Rothbart et al., 2011) and is associated with self-regulation. Furthermore, a higher level of executive control in adolescents is associated with fewer antisocial behaviors (Rothbart et al. 2011). In keeping with the strength model of selfregulation, it is possible that the increased connectivity with the amygdala we observed in the PFC and the IPL may result in depletion of neural resources in these regions. This depletion may adversely affect executive control, resulting in more frequent antisocial behaviors (i.e. risk-taking). We also found increased functional connectivity between the amygdala and the precuneus, a key area within the default mode network, in the RT group. The precuneus is involved in internal processing that is detached from the outside world, such as incorporating personal experiences into thoughts about future events, as well as monitoring the outside world (Buckner et al., 2008). Thus, increased emotional input from the amygdala may affect the ability of the precuneus in making accurate distinctions about future events and information from the outside world, including the way adolescents perceive and think about risktaking. Similar to previous reports, we found that RT adolescents had greater impulsivity and sensation seeking as measured by the ImpSS and novelty seeking as measured by the JTCI (Brejard et al., 2012; Schepis et al., 2008). While the direct correlations per group between these behavioral measures and resting state functional connectivity did not reach significant levels, we did observe a significant positive correlation between right MFG-accumbens connectivity and novelty-seeking scores when we looked at the entire sample. The strong correlation between age and risk items endorsed on the YRBSS that we observed in the RT group is consistent with the literature, which shows an increase in risktaking trajectories through the adolescent years. As is the case with all investigations of resting state functional connectivity, the extent to which we are able to relate our current findings to task-related activation of these regions is limited. While the differences in emotion-regulation circuits observed at

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎

rest provide insight into the intrinsic organization of the adolescent brain, future research is necessary to understand how these differences might affect behavior while engaged in a task that involves emotion regulation. Future studies should focus on the explicit relationship between resting state functional connectivity and task-specific activation of these regions in the context of risktaking in order to elucidate the disparate findings related to hyper/ hypoactivation. Our findings are also constrained by our definition of risk-taking behavior since the definition was based upon endorsement of items on the YRBSS only. Future studies should investigate these functional connectivity patterns in the context of specified risk-taking behavior such as adolescent substance abuse in an effort to better inform targeted treatments. Finally, although reported alcohol and marijuana use was minimal and only reported in a few members of the RT group, it is possible that repeated substance use in these individuals, may result in neurotoxicity that is driving the observed effects. Future studies should take this possibility into account. To conclude, the goal of the present study was to investigate the role of intrinsic activity of regions involved in emotion regulation and reward sensitivity for risk-taking adolescents. We found that our RT adolescents exhibited increased functional connectivity in amygdala-PFC (areas involved in emotion regulation), ventral striatum-PFC, as well as amygdala-PFC/IPL (areas involved in executive control and default mode network) compared with our NRT adolescents. This increased functional connectivity may serve as a vulnerability factor that pre- disposes adolescents towards future, pathological risk-taking behavior. These findings suggest the importance of intrinsic functional connectivity of the amygdala and the PFC in regulating emotions in adolescents who may engage in risk-taking behavior and could advance prevention and treatment development for these atrisk youth.

Acknowledgements This project was funded through a seed grant provided by the Center for BrainHealth, School of Behavioral and Brain Sciences, UT Dallas. This research was supported by Texas Legislature appropriated ARRA Funding for the Middle School Brain Years Program (2009–2011). We would like to thank Vicki Germer, Nneka Chukwueke, Aishia Brown and Elizabeth Huber for their assistance during data collection.

References Baumeister, R.F., Heatherton, T.F., 1993. Self-regulation failure: an overview. Psychological Inquiry 2 (1), 1–15. Biswal, B.B., Van Kylen, J., Hyde, J.S., 1997. Simultaneous assessment of flow and BOLD signals in resting-state functional connectivity maps. NMR in Biomedicine 10, 165–170. Brejard, V., Bonnet, A., Pedinielli, J.L, 2012. The role of temperament and emotional awareness in risk taking in adolescents. L'Encephale 38, 1–9. Buckner, R.L., Andrews-Hanna, J.R., Schacter, D.L., 2008. The brain's default network: anatomy, function, and relevance to disease. Annals of the New York Academy of Sciences 1124, 1–38. Brown, V.M., Labar, K.S., Haswell, C.C., Gold, A.L, Mid-Atlantic MIRECC Workgroup, Beall, S.K., Van Voorhees, E., Marx, C.E., Calhoun, P.S., Fairbank, J.A., Green, K.T., Tupler, L.A., Weiner, R.D., Beckham, J.C., Brancu, M., Hoerle, J.M., Pender, M., Kudler, H, Swinkels, C.M., Nieuwsma, J.A., Runnals, J.J., Youssef, N.A., McDonald, S.D, Davison, R., Yoash-Gantz, R., Taber, K.H., Hurley, R., McCarthy, G., Morey, R.A., 2014. Altered resting-state functional connectivity of basolateral and centromedial amygdala complexes in posttraumatic stress disorder. Neuropsychopharmacology 39 (2), 351–359. Connolly, C.G., Wu, J., Ho, T.C., Hoeft, F., Wolkowitz, O., Eisendrath, S., Frank, G., Hendren, R., Max, J.E., Paulus, M.P., Tapert, S.F., Banerjee, D., Simmons, A.N., Yang, T.T., 2013. Resting-state functional connectivity of subgenual anterior cingulate cortex in depressed adolescents. Biological Psychiatry 74 (12), 898–907.

7

Cox, R.W., 1996. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical Research, An International Journal 29, 162–173. Crowley, T.J., Dalwani, M.S., Mikulich-Gilbertson, S.K., Du, Y.P., Lejuez, C.W., Raymond, K.M., Banich, M.T., 2010. Risky decisions and their consequences: neural processing by boys with antisocial substance disorder. PLoS One 5, 9. Dosenbach, N.U., Fair, D.A., Cohen, A.L., Schlaggar, B.L., Petersen, S.E., 2008. A dualnetworks architecture of top-down control. Trends in Cognitive Sciences 12, 99–105. Dosenbach, N.U., Fair, D.A., Miezin, F.M., Cohen, A.L., Wenger, K.K., Dosenbach, R.A., Fox, M.D., Snyder, A.Z., Vincent, J.L., Raichle, M.E., Schlaggar, B.L., Petersen, S.E., 2007. Distinct brain networks for adaptive and stable task control in humans. Proceedings of the National Academy of Sciences of the United States of America 104, 11073–11078. Dosenbach, N.U., Visscher, K.M., Palmer, E.D., Miezin, F.M., Wenger, K.K., Kang, H.C., Burgund, E.D., Grimes, A.L., Schlaggar, B.L., Petersen, S.E., 2006. A core system for the implementation of task sets. Neuron 50, 799–812. Eaton, D.K., Kann, L., Kinchen, S., Shanklin, S., Flint, K.H., Hawkins, J., Harris, W.A., Lowry, R., McManus, T., Chyen, D., Whittle, L., Lim, C., Wechsler, H., 2012. Youth risk behavior surveillance—United States, 2011. Morbidity and Mortality Weekly Report Surveillance Summaries (Washington, D.C.: 2002) 61, 1–162. Ernst, M., Nelson, E.E., Jazbec, S., McClure, E.B., Monk, C.S., Leibenluft, E., Blair, J., Pine, D.S., 2005. Amygdala and nucleus accumbens in responses to receipt and omission of gains in adults and adolescents. NeuroImage 25, 1279–1291. Ernst, M., Pine, D.S., Hardin, M., 2006. Triadic model of the neurobiology of motivated behavior in adolescence. Psychological Medicine 36, 299–312. Finger, E.C., Marsh, A., Blair, K.S., Majestic, C., Evangelou, I., Gupta, K., Schneider, M.R., Sims, C., Pope, K., Fowler, K., Sinclair, S., Tovar-Moll, F., Pine, D., Blair, R.J., 2012. Impaired functional but preserved structural connectivity in limbic white matter tracts in youth with conduct disorder or oppositional defiant disorder plus psychopathic traits. Psychiatry Research: Neuroimaging 202, 239–244. Gailliot, M.T., Plant, E.A., Butz, D.A., Baumeister, R.F., 2007. Increasing selfregulatory strength can reduce the depleting effect of suppressing stereotypes. Personality & Social Psychology Bulletin 33, 281–294. Galvan, A., Hare, T.A., Parra, C.E., Penn, J., Voss, H., Glover, G., Casey, B.J., 2006. Earlier development of the accumbens relative to orbitofrontal cortex might underlie risk-taking behavior in adolescents. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience 26, 6885–6892. Galvan, A., 2010. Adolescent development of the reward system. Frontiers in Human Neuroscience 46. Gee, D.G., Humphreys, K.L, Flannery, J., Goff, B., Telzer, E.H., Shapiro, M., Hare, T.A., Bookheimer, S.Y., Tottenham, N., 2013. A developmental shift from positive to negative connectivity in human amygdala-prefrontal circuitry. The Journal of Neuroscience: The Official Journal of the Society of Neuroscience 33 (10), 4584–4593. Guyer, A.E., Monk, C.S., McClure-Tone, E.B., Nelson, E.E., Roberson-Nay, R., Adler, A.D., Fromm, S.J., Leibenluft, E., Pine, D.S., Ernst, M., 2008. A developmental examination of amygdala response to facial expressions. Journal of Cognitive Neuroscience 20, 1565–1582. Hagger, M.S., Wood, C., Stiff, C., Chatzisarantis, N.L., 2010. Ego depletion and the strength model of self-control: a meta-analysis. Psychological Bulletin 136, 495–525. Han, D.H., Kim, S.M., Lee, Y.S., Renshaw, P.F., 2012. The effect of family therapy on the changes in the severity of on-line game play and brain activity in adolescents with on-line game addiction. Psychiatry Research: Neuroimaging 202 (2), 126–131. Hare, T.A., Tottenham, N., Galvan, A., Voss, H.U., Glover, G.H., Casey, B.J., 2008. Biological substrates of emotional reactivity and regulation in adolescence during an emotional go- nogo task. Biological Psychiatry 63, 927–934. Janes, A.C., Nickerson, L.D., Frederick Bde, B., Kaufman, M.J., 2010. Prefrontal and limbic resting state brain network functional connectivity differs between nicotine-dependent smokers and non-smoking controls. Drug and Alcohol Dependence 125 (3), 252–259. Joseph, J.E., Liu, X., Jiang, Y., Lynam, D., Kelly, T.H., 2009. Neural correlates of emotional reactivity in sensation seeking. Psychological Science 20, 215–223. Kannurpatti, S.S., Rypma, B., Biswal, B.B., 2012. Prediction of task-related BOLD fMRI with amplitude signatures of resting-state fMRI. Frontiers in Systems Neuroscience 6, 7. Kaufman, J., Birmaher, B., Brent, D., Rao, U., Flynn, C., Moreci, P., Williamson, D., Ryan, N., 1997. Schedule for Affective Disorders and Schizophrenia for SchoolAge Children- Present and Lifetime Version (K-SADS-PL): initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry 36, 980–988. Levesque, J., Joanette, Y., Mensour, B., Beaudoin, G., Leroux, J.M., Bourgouin, P., Beauregard, M., 2004. Neural basis of emotional self-regulation in childhood. Neuroscience 129, 361–369. Luby, J.L., Svrakic, D.M., McCallum, K., Przybeck, T.R., Cloninger, C.R., 1999. The Junior Temperament and Character Inventory: preliminary validation of a child self-report measure. Psychological Reports 84, 1127–1138. Marsh, A.A., Finger, E.C., Fowler, K.A., Jurkowitz, I.T., Schechter, J.C., Yu, H.H., Pine, D.S., Blair, R.J., 2011. Reduced amygdala-orbitofrontal connectivity during moral judgments in youths with disruptive behavior disorders and psychopathic traits. Psychiatry Research: Neuroimaging 194, 279–286. Martin, L.N., Delgado, M.R., 2011. The influence of emotion regulation on decisionmaking under risk. Journal of Cognitive Neuroscience. 9, 2569–2581. McRae, K., Gross, J.J., Weber, J., Robertson, E.R., Sokol-Hessner, P., Ray, R.D., Gabrieli, J.D., Ochsner, K.N., 2012. The development of emotion regulation: an fMRI study of

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i

8

S.J. DeWitt et al. / Psychiatry Research: Neuroimaging ∎ (∎∎∎∎) ∎∎∎–∎∎∎ cognitive reappraisal in children, adolescents and young adults. Social Cognitive

and Affective Neuroscience 7, 11–22. New, A.S., Hazlett, E.A., Buchsbaum, M.S., Goodman, M., Mitelman, S.A., Newmark, R., Trisdorfer, R., Haznedar, M.M., Koenigsberg, H.W., Flory, J., Siever, L.J., 2007. Amygdala-prefrontal disconnection in borderline personality disorder. Neuropsychopharmacology 32, 1629–1640. Pavuluri, M.N., Passarotti, A.M., Harral, E.M., Sweeney, J.A., 2009. An fMRI study of the neural correlates of incidental versus directed emotion processing in pediatric bipolar disorder. Journal of the American Academy of Child and Adolescent Psychiatry 48, 308–319. Perlman, G., Simmons, A.N., Wu, J., Hahn, K.S., Tapert, S.F., Max, J.E., Paulus, M.P., Brown, G.G., Frank, G.K., Campbell-Sills, L., Yang, T.T., 2012. Amygdala response and functional connectivity during emotion regulation: a study of 14 depressed adolescents. Journal of Affective Disorders 139, 75–84. Petersen, S.E., Posner, M.I., 2012. The attention system of the human brain: 20 years after. Annual Review of Neuroscience 35, 73–89. Posner, M.I., Petersen, S.E., 1990. The attention system of the human brain. Annual Review of Neuroscience 13, 25–42. Posner, M.I., Rothbart, M.K., 2007. Research on attention networks as a model for the integration of psychological science. Annual Review of Psychology 58, 1–23. Power, J.D., Barnes, K.A., Snyder, A.Z., Schlaggar, B.L., Petersen, S.E., 2012. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. NeuroImage 63 (2), 999. Qiao, Y., Xie, B., Du, X., 2012. Abnormal response to emotional stimulus in male adolescents with violent behavior in China. European Child and Adolescent Psychiatry 21 (4), 193–198. Qin, S, Young, C.B., Supekar, K, Uddin, L.Q., Menon, V., 2012. Immature integration and segregation of emotion-related brain circuitry in young children. Proceedings of the National Academy of Sciences of the United States of America 109 (20), 7941–7946. Rivers, S.E., Reyna, V.F., Mills, B., 2008. Risk taking under the influence: a fuzzytrace theory of emotion in adolescence. Developmental Review 28, 107–144. Rothbart, M.K., Sheese, B.E., Rueda, M.R., Posner, M.I., 2011. Developing mechanisms of self- regulation in early life. Emotion Review: Journal of the International Society for Research on Emotion 3, 207–213.

Roy, A.K., Shehzad, Z., Margulies, D.S., Kelly, A.M., Uddin, L.Q., Gotimer, K., Biswal, B.B., Castellanos, F.X., Milham, M.P., 2009. Functional connectivity of the human amygdala using resting state fMRI. NeuroImage 45, 614–626. Schepis, T.S., Desai, R.A., Smith, A.E., Cavallo, D.A., Liss, T.B., McFetridge, A., Potenza, M.N., Krishnan-Sarin, S., 2008. Impulsive sensation seeking, parental history of alcohol problems, and current alcohol and tobacco use in adolescents. Journal of Addiction Medicine 2, 185–193. Silveri, M.M., Rogowska, J., McCaffrey, A., Yurgelun-Todd, D.A., 2011. Adolescents at risk for alcohol abuse demonstrate altered frontal lobe activation during Stroop performance. Alcoholism, Clinical and Experimental Research 35, 218–228. Smith, J.F., Pillai, A., Chen, K., Horwitz, B., 2010. Identification and validation of effective connectivity networks in functional magnetic resonance imaging using switching linear dynamic systems. NeuroImage 52, 1027–1040. Spear, L.P., 2009. Heightened stress responsivity and emotional reactivity during pubertal maturation: implications for psychopathology. Development and Psychopathology 21, 87–97. Staudinger, M.R., Erk, S., Walter, H., 2011. Dorsolateral prefrontal cortex modulates striatal reward encoding during reappraisal of reward anticipation. Cerebral Cortex 21, 2578–2588. Steinberg, L., 2004. Risk taking in adolescence: what changes, and why?. Annals of the New York Academy of Sciences, 1021; , pp. 51–58. van den Heuvel, M.P., Hulshoff Pol, H.E., 2010. Exploring the brain network: a review on resting-state fMRI functional connectivity. European Neuropsychopharmacology: The Journal of the European College of Neuropsychopharmacology 20, 519–534. Wetherill, R.R., Bava, S., Thompson, W.K., Boucquey, V., Pulido, C., Yang, T.T., Tapert, S.F., 2012. Frontoparietal connectivity in substance-naive youth with and without a family history of alcoholism. Brain Research 1432, 66–73. Xie, C., Li, S.J., Shao, Y., Fu, L., Goveas, J., Ye, E., Li, W., Cohen, A.D., Chen, G., Zhang, Z., Yang, Z., 2011. Identification of hyperactive intrinsic amygdala network connectivity associated with impulsivity in abstinent heroin addicts. Behavioural Brain Research 216, 639–646. Yue, Y., Yuan, Y., Hou, Z., Jiang, W., Bai, F., Zhang, Z., 2013. Abnormal functional connectivity of amygdala in late-onset depression was associated with cognitive deficits. PLoS One 8 (9), 10. Zuckerman, M., 1996. The psychobiological model for impulsive unsocialized sensation seeking: a comparative approach. Neuropsychobiology 34, 125–129.

Please cite this article as: DeWitt, S.J., et al., Adolescent risk-taking and resting state functional connectivity. Psychiatry Research: Neuroimaging (2014), http://dx.doi.org/10.1016/j.pscychresns.2014.03.009i