Abstract

The tendency toward impulsive behavior under emotional duress (negative and positive urgency) predicts a wide range of maladaptive risk-taking and behavioral disorders. However, it remains unclear how urgency relates to limbic system activity as induced from emotional provocation. This study used functional magnetic resonance imaging to examine the relationship between brain responses to visual emotional stimuli and urgency traits. Twenty-seven social drinkers (mean age = 25.2, 14 males) viewed negative (Neg), neutral (Neu), and positive (Pos) images during 6 fMRI scans. Brain activation was extracted from a priori limbic regions previously identified in studies of emotional provocation. The right posterior orbitofrontal cortex (OFC) and left amygdala were activated in the [Neg>Neu] contrast, whereas the left posterior OFC was activated in the [Pos>Neu] contrast. Negative urgency was related to the right lateral OFC (r = 0.43, P = 0.03) and the left amygdala (r = 0.39, P = 0.04) [Neg>Neu] activation. Negative urgency also mediated the relationship between [Neg>Neu] activation and general risk-taking (regression weights = 3.42 for right OFC and 2.75 for the left amygdala). Emotional cue-induced activation in right lateral OFC and left amygdala might relate to emotion-based risk-taking through negative urgency.

Introduction

Negative and positive urgency are personality traits characterized by tendencies to act rashly in the face of extreme negative or positive emotions, respectively (Cyders and Smith 2008a). These traits have been shown to be common risk factors for a wide range of risk-taking behaviors, including problematic alcohol use (see a review by Coskunpinar et al. 2013), binge eating (see a review by Fischer et al. 2008), gambling (e.g., Cyders and Smith 2008b; Fischer and Smith 2008; Michalczuk et al. 2011), sexual risk-taking (e.g., Zapolski et al. 2009; Dir et al. 2013), and drug use (e.g., Verdejo-Garcia et al. 2007; Zapolski et al. 2009). In particular, urgency has been shown to be the most important impulsivity-related trait for maladaptive and problematic levels of risk-taking (Cyders and Smith 2008a) and has been implicated in multiple clinical disorders (see Zapolski et al. 2010 for a review) such as borderline personality disorder (e.g., Peters et al. 2013), alcohol and substance dependence (e.g., Cyders et al. 2007; Verdejo-Garcia et al. 2007), bulimia (e.g., Fischer et al. 2007; Anestis et al. 2009), and antisocial personality disorder (Whiteside et al. 2005).

Despite its clinical relevance, the brain systems related to urgency traits are not yet well understood. Using functional magnetic resonance imaging (fMRI), the current study examines how urgency is related to brain responses induced by emotionally provocative stimuli, and, in turn, how both urgency and these brain responses relate to general risk-taking. Understanding how urgency and limbic activation from emotional stimuli are inter-related sheds light on the underlying mechanisms leading toward emotion-based risk-taking behaviors.

A growing body of evidence examines how urgency is related to and influenced by the frequency and intensity of emotional experiences. First, Cyders and Smith (2007) found that negative and positive urgency loaded onto the Neuroticism domain of the NEO Personality Inventory-Revised (NEO-PI-R; Costa and McCrae 1992). Although clearly overlapping with negative affectivity, much variance in urgency nevertheless remained unexplained by the additive or interactive effects of Neuroticism and lack of planning (72–85% variance remained for positive urgency and 45–61% variance remained for negative urgency). Second, we (Cyders et al. 2010) reported that, although positive urgency predicted increased alcohol consumption in response to positive mood induction, positive urgency was unrelated to baseline mood, and to mood state changes following manipulation. Third, urgency appears to be a unique predictor of risk-taking behavior, independently of emotion frequency/intensity, and it predicts risk-taking over and above the additive and interactive effects of lack of planning and the frequency/intensity of reported emotions (Cyders and Coskunpinar 2010). These cumulative data suggest that urgency shares only a small amount of variance with the experience of emotional responses, per se, and that its relationship to risk-taking persists after controlling self-reported emotional fluctuations or changes in stimulus-induced emotional states.

However, thus far the relationship between urgency and the experience of emotional responses has been based primarily on subjectively rated emotional states that are likely affected by self-report and expectancy biases. Therefore, the current study sought to examine how urgency relates to brain activation from emotionally provocative images, as derived from the International Affective Picture System (IAPS; Lang et al. 1997). Such emotionally charged scenes (positive or negative) induce activation within the visual system (occipital cortex, primarily Brodmann's Areas 18 and 19, the calcarine cortex, the fusiform gyrus [Lane et al. 1997; Reiman et al. 1997; Lang et al. 1998; Phan et al. 2002]) and in limbic regions such as the amygdala, orbitofrontal cortex (OFC), and cingulate cortex (see meta-analyses by Sabatinelli et al. 2011).

Urgency itself has been theorized to be related to multiple brain regions, including the medial, ventromedial, dorsolateral, and orbitofrontal regions (Levesque et al. 2003, 2004; Ochsner et al. 2004; Kim and Hamann 2007; Mak et al. 2009), the anterior cingulate cortex (Levesque et al. 2004), and the amygdala (Ochsner et al. 2004) (see a review by Karyadi et al. 2012). Emerging empirical data have been somewhat inconsistent. Joseph et al. (2009) reported reduced blood oxygen level-dependent (BOLD) responses to a mixture of positively and negatively valenced high-arousal stimuli in the anterior and medial OFC and anterior cingulate. On the other hand, Albein-Urios et al. (2013) found that negative urgency was positively related to amygdala activation during negative emotion maintenance and reappraisal. We recently reported that negative urgency is related to increased ventromedial (limbic) prefrontal cortex activation from alcoholic drink aromas and that the relationship between this activation and both alcohol cravings and problematic alcohol use is mediated by negative urgency (Cyders et al. 2013). This suggests that neural responses to alcohol cues might increase cravings and problematic alcohol use by way of tendencies to engage in risky behaviors in response to negative emotion.

As this prior report focused on the appetitive stimuli of alcoholic drink aromas, we re-analyzed data from Cyders et al. (2013) to instead examine brain responses to the emotionally charged visual stimuli used to induce mood (and in theory, states of urgent behavior). In particular, we examined how positive and negative urgency relate to BOLD responses to visual emotional stimuli in the OFC and the amygdala. We hypothesized that urgency would be related to activation in these areas to either positive or negative emotional stimuli. Finally, we hypothesized that urgency would mediate the relationship between activation in these regions and self-reported risk-taking behaviors.

Materials and Methods

Participants

Thirty-eight right-handed healthy social drinkers (drinking at least 1–3 drinks per week and 1 incidence of drunkenness over the previous month, but excluding alcohol dependence and maternal alcoholism) between the ages of 21 and 35 were recruited, and 30 completed all study procedures. The participants who completed the study did not differ from non-completers on any demographic variables (P-values ranged from 0.19 to 0.86), number of drinks per week (t = −0.58, P = 0.57), or family history of alcohol problems (χ2 = 0.40, df = 1, P = 0.53). Three subjects whose head motion during functional imaging exceeded peak-to-peak translations of 2 mm, and rotations of 2° were excluded from further analyses, resulting in a final sample of n = 27. Participants were in good self-reported health, had no history of neurological disorders of cerebral origin, treatment for psychiatric disorders, had a normal sense of smell as assessed with the smell identification test, and were not taking any medications (licit or illicit, including tobacco). All participants voluntarily signed informed consent statements approved by the Indiana University institutional review board and received $150 for completion of the study.

Self-Report Measures

“The Affect Grid” (Russell et al. 1989) is a 9 × 9 grid with affect descriptors in each corner. Participants indicate the appropriate cell of the grid that represents current emotions. The affect grid results in separate valence (pleasantness vs. unpleasantness) and arousal (high arousal vs. low arousal) ratings and has good inter-rater reliability (0.98 for valence and 0.97 for activation) and convergent validity (with the Positive and Negative Affect Scale; PANAS; Russell et al. 1989). Higher valence ratings indicate increased self-reported pleasant valence, whereas higher arousal ratings indicate increased self-reported arousal.

“The UPPS Impulsive Behavior Scale” (UPPS-P; [Lynam D, Smith GT, Cyders MA, Fischer S, Whiteside SA. 2007. The UPPS-P: A multidimensional measure of risk for impulsive behavior. Unpublished manuscript]) is a 59-item self-report scale, with responses ranging from 1 (agree strongly) to 4 (disagree strongly). The UPPS-P is designed to measure 5 separate dispositions to rash action (see Lynam D, Smith GT, Cyders MA, Fischer S, Whiteside SA. 2007. The UPPS-P: A multidimensional measure of risk for impulsive behavior. Unpublished manuscript); only the positive urgency (14 items; α = 0.91, M [SD] = 1.68 [0.44] in the current sample) and the negative urgency (11 items; α = 0.85; M [SD] = 1.94 [0.42] in the current sample) subscales were utilized for this study. Items were recoded so that higher mean scores on the subscales represent higher levels of urgency.

“The Risky Behavior Scale” (RBS; Fischer and Smith 2004) is an 83-item scale (M [SD] = 169.67 [19.74] in the current sample) that assesses the frequency with which individuals endorse participating in a wide range of risk-taking behaviors. Items are coded on a 5-point Likert-type scale, with 1 indicating never participating in the behavior and 5 indicating often participating in the behavior. Items were summed to produce a score in which larger values indicate more frequent risk-taking behavior.

Procedure

Study Sessions

All procedures were approved by the Indiana University Institutional Review Board. Participants completed 2 sessions, a screening session and an imaging session. In screening sessions, participants completed a series of self-report questionnaires, as well as a urine drug and pregnancy tests to assess eligibility for the imaging session. Upon meeting inclusion criteria, participants were scheduled for imaging (average of 32 days post-screening session) and asked to refrain from alcohol consumption for 3 days prior to imaging. On the imaging day, participants reported to the Indiana University Clinical Research Center between 8 and 10 a.m. and were provided with a light, standardized breakfast. Vitals and repeat drug and pregnancy urine screens were conducted.

Participants were then escorted to the imaging suite, where they rated their current mood and were exposed to the odorants and 3 sample images (1 from each mood condition—Positive, Negative, and Neutral) that they would encounter during imaging (see below). Odorants included the participant's most frequently consumed alcoholic beverage, grape juice, and sham odorant controls. The details and effects of the odorant presentations are reported elsewhere (Cyders et al. 2013) and will not be reviewed here.

Mood images from the IAPS (Lang et al. 1999) were chosen based on the developmental ratings of valence and arousal provided by Lang et al. 1997. Three image groups were formed: Neutral (mean valence = 5.11, mean arousal = 3.28), Negative (mean valence = 2.55, mean arousal = 5.55), and Positive (mean valence = 6.94, mean arousal = 5.78; see Supplementary Table for images used and developmental ratings). In verifying that the images selected were sufficiently different in their intended characteristics, a 3 (Positive, Negative, and Neutral mood condition) × 2 (valence and arousal developmental ratings) repeated measures' analysis of variance on the developmental ratings indicated a main effect of mood condition (F = 502.60, P < 0.001), and an interaction between mood condition and rating (F = 448.63, P < 0.001). Follow-up paired t-test comparisons confirmed that positive image valence and arousal ratings were significantly higher than ratings of the neutral and negative images (t = 23.81, P < 0.001 and t = 42.26, P < 0.001, respectively); negative valence ratings were lower than the neutral valence ratings (t = −32.24, P < 0.001), and negative arousal ratings were higher than the neutral arousal ratings (t = 19.60, P < 0.001). Positive arousal ratings were significantly higher than neutral valence ratings (t = 23.03, P < 0.001), but there was no significant difference between positive and negative image arousal ratings (t = 1.79, P = 0.08).

Scan Characteristics

A total of 6 functional imaging scans of combined olfactory/visual stimuli were performed, with a short (∼5–10 min) break after the first 3 scans. Two scans were completed for each mood (2 positive mood scans consisting of all positive images from the IAPS [Pos], 2 negative mood scans consisting of all negative images from the IAPS [Neg], and 2 neutral mood scans consisting of all neutral images from the IAPS [Neu]; see also Supplementary Table). Mood condition order was randomly assigned across scans, but randomization was restricted so that all 3 moods were presented prior to, and again following, the break. Participants first saw 2 2-s IAPS images, followed by a 2-s odorant trial (alcohol, appetitive control, or sham odorant; see full description in Cyders et al. 2013), and an interstimulus interval ranging from 3 to 10 s. Participants then saw another 2-s IAPS image, followed by another 3- to 10-s interstimulus interval. The presentation of odorant and image repeated a total of 30 times throughout the scan, for a total of 32 images and 30 odorant presentations (10 in each class: alcohol, appetitive control, and sham; see Fig. 1 from Cyders et al. 2013). Between scans, participants again rated current mood.

Figure 1.

Brain regions showing significant [Neg>Neu] and [Pos>Neu] BOLD activation to IAPS images, with a priori regions of interest denoted by circles (blue: left amygdala, yellow and red: left and right OFC, respectively). Display height P = 0.001, k = 5. For more details see Table 1. No activation was found for [Neu>Neg] and [Neu>Pos] contrasts. Color bar scale represents the t-statistic.

Figure 1.

Brain regions showing significant [Neg>Neu] and [Pos>Neu] BOLD activation to IAPS images, with a priori regions of interest denoted by circles (blue: left amygdala, yellow and red: left and right OFC, respectively). Display height P = 0.001, k = 5. For more details see Table 1. No activation was found for [Neu>Neg] and [Neu>Pos] contrasts. Color bar scale represents the t-statistic.

Image Acquisition

BOLD contrast sensitive functional imaging scans were conducted with a 12-channel head coil array on a Siemens 3T Magnetom Trio-Tim scanner (Erlangen). A whole-brain high-resolution anatomic image volume (1.0 × 1.0 × 1.2 mm3 voxels) was first collected using a 3D magnetization prepared rapid gradient echo sequence for anatomic registration of the functional images. For each functional scan, 189 BOLD volumes were acquired in 6:54 min using an echo-planar imaging pulse sequence (gradient echo; 2250/29 ms repetition/echo time, 78° flip angle, 88 × 88 matrix, 2.5 × 2.5 × 3.0 mm3 voxels, 39 slices, GRAPPA acceleration factor 2) that incorporated a 3D prospective acquisition correction to minimize effects of the head motion (Thesen et al. 2000).

Data Analysis

Data were analyzed using SPM8 (Wellcome Department of Imaging Neuroscience, University College, London) and PASW Statistics v. 18.0 (SPSS, Inc.). Functional volumes were corrected for differences in slice acquisition timing and rigid-body realigned to the initial volume of the first functional imaging scan to account for residual movement after prospective motion correction. Each subject's high-resolution anatomic image was co-registered to the reference functional volume (first volume of the first functional scan) and segmented into tissue classes. Nonlinear spatial transformation from native to Montreal Neurological Institute (MNI) space was performed during the segmentation stage, and then those transformation parameters were applied to all BOLD volumes, which were subsequently re-sampled to isotropic (2 mm per side) voxels and smoothed by a 6-mm full-width at half-maximum isotropic Gaussian Kernel. Within-subjects general linear model used SPM's canonical hemodynamic response function and included both image and odorant presentations with events modeled to have 2- and 3-s durations, respectively. However, odorant presentations were treated as events of no interest. Movement parameters from realignment were included as regressors whereas a high-pass filter with a cutoff of 1/128 Hz was applied to each voxel's time series to remove low-frequency noise.

As prior research suggests the activation of particular limbic brain regions in response to emotionally provocative images (Sabatinelli et al. 2011), we targeted 4 a priori defined regions of interest (ROIs; see Supplementary Figure) in the left and right amygdala and the right and left lateral OFC. ROIs were constructed as 8-mm-radius spheres with MarsBaR utility (Brett et al. 2002; http://marsbar.sourceforge.net/) using OFC (right: [40, 32, −12], left: [−40, 32, −14]) and amygdala (right: [22, −4, −18], left: [−20, −6, −18]) coordinates derived from a meta-analysis by Sabatinelli et al. (2011). The mean BOLD contrasts of [Pos>Neu] and [Neg>Neu] images were first assessed for significant peak effects using voxel-wise analyses, and correcting for family-wise error (FWE) based on the ROI volume. For those ROIs in which peak effects surpassed pFWE < 0.05, the mean BOLD contrast values were extracted and then analyzed with SPSS for their relationships with urgency traits, as well as for how urgency might mediate the relationship between brain response and risk-taking.

Results

Self-Report Mood Responses to Images

There were no significant differences between mood ratings across scans of the same mood condition. Valence ratings following the first and the second scan of the same mood condition did not differ from each other and were averaged to create 1 rating for each mood condition. As analyzed in a Mood Condition (3: positive, negative, neutral) × Response Type (2: self-reported valence and arousal) linear-mixed-effects model, there was a significant main effect for Mood Condition (P < 0.001) and Response Type (P = 0.03), as well as a significant interaction between these 2 factors (P < 0.001; also reported in Cyders et al. 2013). Valence ratings following the negative mood condition were significantly lower than following either the neutral (t = −7.14, P < 0.001) or positive (t = −7.46, P < 0.001) mood conditions; valence ratings following the positive mood condition were also significantly higher than those following the neutral mood condition (t = 2.53, P = 0.02). Arousal ratings following both negative and positive mood conditions were higher than those for the neutral mood condition (t = 2.85, P = 0.01 and t = 6.44, P < 0.001, respectively). Contrary to expectations, arousal ratings following the positive mood condition were significantly higher than arousal ratings following the negative mood condition (t = 2.60, P = 0.02).

Negative urgency was significantly related to valence following only the negative mood image condition (r = −0.42, P = 0.03); negative and positive urgency were unrelated to the remaining valence and arousal ratings (all ps > 0.05). As in previous studies, negative and positive urgency were highly related (r = 0.56, P = 0.002).

BOLD Responses to Emotional Images

Using the a priori ROIs to perform small volume corrections (pFWE < 0.05), the [Neg>Neu] contrast elicited peak voxel effects in the right lateral OFC ([40, 30, −16], pFWE = 0.001) and left amygdala ([−20, −6, −14], pFWE = 0.022) (multiple comparison correction across all ROIs was performed by including all voxels within our 4 8-mm-radius spherical ROIs. Peak voxel effects remained significant for the right lateral OFC for the [Neg>Neu] contrast (pFWE = 0.004), whereas effects in the left amygdala showed a trend toward significance (pFWE = 0.087). Effects in the right (pFWE = 0.068) and left lateral OFC (pFWE = 0.069) also showed a trend toward significance) (see Fig. 1), whereas left OFC and the right amygdala failed to reach corrected significance. For the [Pos>Neu] contrast, significant peak effects were present in right ([38, 28, −18], pFWE = 0.017) and left OFC ([−36, 26, −12], pFWE = 0.018). Neither left nor right amygdala nor another potentially interesting limbic region, the anterior cingulate cortex, (a priori hypothesis, we also examined the anterior cingulate as a limbic region of potential interest (Karyadi et al. 2012; Sabatinelli et al. 2013). There were no significant peak effects (P < 0.001, k = 5) in this region in [Neg>Neu] or [Pos>Neu] contrasts [using an 8-mm-radius sphere constructed around the MNI coordinate [2, 19, 25], as suggested by Sabatinelli et al. 2013]) reached corrected peak significance. Brain activations are depicted in Figure 1 and listed in Table 1.

Table 1

BOLD activation to positive and negative images contrasted to the neutral image baseline

Region MNI coordinates x, y, z (mm) Peak significance (unc.) Cluster size k Peak voxel Z 
[Neg>Neu] contrast 
 R Middle occipital gyrus 50, −74, 4 <0.001 10 831 6.05 
 R precentral gyrus 44, 6, 30 <0.001 336 4.50 
 R OFC 40, 30, −16 <0.001 146 4.41* 
 R calcarine gyrus 22, −56, 6 <0.001 36 4.36 
 L parahippocampal gyrus −22, −26, −6 <0.001 71 4.32 
 R ventromedial prefrontal cortex 8, 50, −14 <0.001 4.32 
 R middle temporal gyrus 50, −10, −16 <0.001 21 4.19 
 R parahippocampal gyrus 18, −32, −10 <0.001 18 4.08 
 R superior frontal gyrus 12, 6, 68 <0.001 4.06 
 R inferior frontal gyrus 54, 34, −2 <0.001 32 3.96 
 L postcentral gyrus −42, −16, 56 <0.001 50 3.81 
 L precentral gyrus −40, 6, 30 <0.001 19 3.55 
 L middle temporal gyrus −62, −4, −14 <0.001 16 3.45 
 L amygdala −20, −6, −14 <0.001 3.45 
 R amygdala 16, −8, −12 <0.001 3.43 
 L cuneus/lingual gyrus −2, 68, 4 <0.001 3.39 
 R middle frontal gyrus 38, −2, 58 0.001 10 3.25 
[Pos>Neu] contrast 
 R middle occipital gyrus 48, −74, 4 <0.001 14 544 6.49 
 R inferior frontal gyrus 46, 6, 26 <0.001 222 4.52 
 R ventromedial prefrontal cortex 8, 48, −14 <0.001 33 4.51 
 L inferior parietal lobule −28, −56, 52 <0.001 108 4.50 
 R OFC 34, 26, −18 <0.001 55 4.29* 
 L frontal operculum/insula −34, 24, −12 <0.001 79 4.14** 
 L anterior putamen −30, 6, −6 <0.001 3.93 
 R ventromedial prefrontal cortex 10, 60, −8 <0.001 10 3.88 
 R middle temporal gyrus 50, −14, −18 <0.001 18 3.73 
 L middle temporal gyrus −64, −12, −16 <0.001 32 3.69 
 R ventral anterior thalamus 14, −6, −2 <0.001 10 3.69 
 R precentral gyrus 40, 0, 40 <0.001 27 3.65 
 L superior frontal gyrus −34, −2, 42 <0.001 32 3.60 
 R inferior parietal lobule 60, −30, 28 <0.001 13 3.52 
 R precuneus 10, −40, 48 <0.001 28 3.51 
 L medial prefrontal cortex −6, 62, 0 <0.001 11 3.49 
 L ventromedial prefrontal cortex −4, 54, −16 <0.001 3.43 
 R medial frontal Gyrus/SMA 8, 14, 58 <0.001 3.34 
Region MNI coordinates x, y, z (mm) Peak significance (unc.) Cluster size k Peak voxel Z 
[Neg>Neu] contrast 
 R Middle occipital gyrus 50, −74, 4 <0.001 10 831 6.05 
 R precentral gyrus 44, 6, 30 <0.001 336 4.50 
 R OFC 40, 30, −16 <0.001 146 4.41* 
 R calcarine gyrus 22, −56, 6 <0.001 36 4.36 
 L parahippocampal gyrus −22, −26, −6 <0.001 71 4.32 
 R ventromedial prefrontal cortex 8, 50, −14 <0.001 4.32 
 R middle temporal gyrus 50, −10, −16 <0.001 21 4.19 
 R parahippocampal gyrus 18, −32, −10 <0.001 18 4.08 
 R superior frontal gyrus 12, 6, 68 <0.001 4.06 
 R inferior frontal gyrus 54, 34, −2 <0.001 32 3.96 
 L postcentral gyrus −42, −16, 56 <0.001 50 3.81 
 L precentral gyrus −40, 6, 30 <0.001 19 3.55 
 L middle temporal gyrus −62, −4, −14 <0.001 16 3.45 
 L amygdala −20, −6, −14 <0.001 3.45 
 R amygdala 16, −8, −12 <0.001 3.43 
 L cuneus/lingual gyrus −2, 68, 4 <0.001 3.39 
 R middle frontal gyrus 38, −2, 58 0.001 10 3.25 
[Pos>Neu] contrast 
 R middle occipital gyrus 48, −74, 4 <0.001 14 544 6.49 
 R inferior frontal gyrus 46, 6, 26 <0.001 222 4.52 
 R ventromedial prefrontal cortex 8, 48, −14 <0.001 33 4.51 
 L inferior parietal lobule −28, −56, 52 <0.001 108 4.50 
 R OFC 34, 26, −18 <0.001 55 4.29* 
 L frontal operculum/insula −34, 24, −12 <0.001 79 4.14** 
 L anterior putamen −30, 6, −6 <0.001 3.93 
 R ventromedial prefrontal cortex 10, 60, −8 <0.001 10 3.88 
 R middle temporal gyrus 50, −14, −18 <0.001 18 3.73 
 L middle temporal gyrus −64, −12, −16 <0.001 32 3.69 
 R ventral anterior thalamus 14, −6, −2 <0.001 10 3.69 
 R precentral gyrus 40, 0, 40 <0.001 27 3.65 
 L superior frontal gyrus −34, −2, 42 <0.001 32 3.60 
 R inferior parietal lobule 60, −30, 28 <0.001 13 3.52 
 R precuneus 10, −40, 48 <0.001 28 3.51 
 L medial prefrontal cortex −6, 62, 0 <0.001 11 3.49 
 L ventromedial prefrontal cortex −4, 54, −16 <0.001 3.43 
 R medial frontal Gyrus/SMA 8, 14, 58 <0.001 3.34 

Note: unc: Uncorrected. Brain regions showing increased BOLD response to negative [Neg>Neu] and positive [Pos>Neu] emotional images when compared with baseline neutral images (P = 0.001, k = 5). Family-wise error-corrected peak significance (pFWE < 0.05) within assessed volume of a priori regions of interest is reached in the OFC (right*, left**) and left amygdala (). For the [Pos>Neu] contrast, left OFC activation is part of the left frontal operculum/insula cluster. SMA, supplementary motor area; MNI, Montreal Neurological Institute; L, left; R, right.

Relationship of Urgency and BOLD Activation

Mean BOLD contrast in the a priori regions was unrelated to either self-reported valence/arousal ratings or to positive urgency; however, negative urgency was significantly related to the mean [Neg>Neu] BOLD contrast in the right OFC (r = 0.39, P = 0.04, Fig. 2) and left amygdala (r = 0.43, P = 0.03; Fig. 3).

Figure 2.

Positive correlation between negative urgency and [Neg>Neu] BOLD activation in 8-mm-radius spherical region centered ([40, 32, −12]) in the right lateral OFC. For reference, voxel-wise map of the correlation between [Neg>Neu] BOLD contrast and negative urgency is illustrated by inset (display threshold P = 0.01, k = 30, masked for [Neg>Neu] BOLD activation).

Figure 2.

Positive correlation between negative urgency and [Neg>Neu] BOLD activation in 8-mm-radius spherical region centered ([40, 32, −12]) in the right lateral OFC. For reference, voxel-wise map of the correlation between [Neg>Neu] BOLD contrast and negative urgency is illustrated by inset (display threshold P = 0.01, k = 30, masked for [Neg>Neu] BOLD activation).

Figure 3.

Positive correlation between negative urgency and [Neg>Neu] BOLD activation in 8-mm-radius spherical region centered ([−20, −6, −18]) in the left amygdala. For reference, voxel-wise map of the correlation between [Neg>Neu] BOLD contrast and negative urgency is illustrated by inset (display threshold P = 0.01, k = 30, masked for [Neg>Neu] BOLD activation).

Figure 3.

Positive correlation between negative urgency and [Neg>Neu] BOLD activation in 8-mm-radius spherical region centered ([−20, −6, −18]) in the left amygdala. For reference, voxel-wise map of the correlation between [Neg>Neu] BOLD contrast and negative urgency is illustrated by inset (display threshold P = 0.01, k = 30, masked for [Neg>Neu] BOLD activation).

Relationships with General Risk-Taking

We next conducted a series of 3 separate mediation analyses using the INDIRECT SPSS macro provided by Preacher and Hayes (2008), which uses the product of coefficient approach and bootstrapping to examine the indirect effect. Confidence intervals (95%) that do not contain zero are considered significant indirect effects. Here, we tested negative urgency's mediation of: (1) mood valence ratings following the negative mood condition on risk-taking, and (2) brain activation on risk-taking. We also tested alternative reverse mediation models for each analysis (i.e., negative urgency as the independent variable and negative mood and brain activation as the mediator).

As hypothesized, negative urgency was a significant mediator of the relationship between mood valence ratings following the negative mood condition and general risk-taking (Fig. 4; regression weight, b = −2.48, SE = 2.02, 95% CI [−9.50, −0.12], controlling for sex and valence ratings following the neutral mood condition). The reverse mediation model (mood valence ratings following the negative mood condition as a mediator in the relationship between negative urgency and risk-taking) was not significant, nor were mediation analyses with arousal ratings following the negative mood condition (both as the independent variable and as a mediator). Conversely, positive urgency did not mediate the relationship between mood valence ratings following the positive mood condition and general risk-taking (controlling for sex and valence ratings following the neutral mood condition). The reverse mediation (with valence ratings following the positive mood condition as the mediator in the relationship between positive urgency and risk-taking) was also not significant.

Figure 4.

Mediation of negative mood valence and risk-taking through negative urgency. For clarity, only pathways with significant (*P < 0.05) regression weight (b) are depicted. Mediation was examined with the SPSS INDIRECT macro provided by Preacher and Hayes (2008) using the 95% confidence intervals. Risk-taking was assessed by total score on the Risky Behavior Scale.

Figure 4.

Mediation of negative mood valence and risk-taking through negative urgency. For clarity, only pathways with significant (*P < 0.05) regression weight (b) are depicted. Mediation was examined with the SPSS INDIRECT macro provided by Preacher and Hayes (2008) using the 95% confidence intervals. Risk-taking was assessed by total score on the Risky Behavior Scale.

Also as hypothesized, negative urgency mediated the relationship between left amygdala activation in the [Neg>Neu] contrast and risk-taking (b = 2.75, SE = 1.71, 95% CI [0.20, 7.40], controlling for sex), as well as the relationship between right lateral OFC activation in the [Neg>Neu] contrast and risk-taking (b = 3.42, SE = 2.58, 95% CI [0.02, 11.68], controlling for sex; see Fig. 5). The reverse mediation models (where BOLD activation in the left amygdala and the right lateral OFC mediated the relationship between negative urgency and general risk-taking) were not significant. There were no significant mediations using positive urgency and activation in response to [Pos>Neu] contrasts.

Figure 5.

The relationship of BOLD response to [Neg>Neu] images and risk-taking as mediated by negative urgency. For clarity, only pathways with significant (*P < 0.05) regression weights (b) are depicted. Two separate mediation models were examined with the SPSS INDIRECT macro provided by Preacher and Hayes (2008) using the 95% confidence intervals. OFC, orbitofrontal cortex. Risk-taking assessed as total score on the Risky Behavior Scale.

Figure 5.

The relationship of BOLD response to [Neg>Neu] images and risk-taking as mediated by negative urgency. For clarity, only pathways with significant (*P < 0.05) regression weights (b) are depicted. Two separate mediation models were examined with the SPSS INDIRECT macro provided by Preacher and Hayes (2008) using the 95% confidence intervals. OFC, orbitofrontal cortex. Risk-taking assessed as total score on the Risky Behavior Scale.

Discussion

The current study examined the relationship between urgency and BOLD responses to emotional stimuli, as well as how these variables relate to risk-taking behavior. As predicted from meta-analyses (Sabatinelli et al. 2011), emotional images (when compared with neutral images) provoked significant BOLD responses in the lateral OFC and amygdala. Moreover, right lateral OFC and left amygdala responses to negative images (when compared with neutral images) were related to negative urgency. Although positive images (when compared with neutral images) also activated both right and left lateral OFC, positive urgency was unrelated to this activation. Finally, negative urgency mediated the relationships between (1) negative image-induced mood ratings and (2) negative image-provoked activation in right lateral OFC and left amygdala and general risk-taking. Importantly, alternative mediation models were not significant. Collectively, the results suggest that the tendency to behave rashly in negative mood states is a key personality trait that mediates relationships between both (1) the report of negative emotions and (2) limbic brain activity during emotionally charged experiences and subsequent general risk-taking.

Negative urgency has previously been associated with increased ventromedial prefrontal cortex activation to alcoholic drink aromas (Cyders et al. 2013), reduced BOLD responses in the anterior and medial OFC and anterior cingulate to positively and negatively valenced high-arousal stimuli (Joseph et al. 2009; both of which were outside our regions of interest, and neither of which activated in our study), and increased amygdala activation during negative emotion maintenance and reappraisal (Albein-Urios et al. 2013), although this latter work used a clinical population with cocaine use and personality disorders. The current study sampled healthy, young adults and found similar patterns of activation to negative emotional stimuli that are related to both negative urgency and to risk-taking behavioral outcomes (except in the case of Joseph et al. 2009, which found reduced activation to a combination of both negatively and positively valenced high-arousal stimuli). This suggests that links between emotionally provoked brain activation and negative urgency are likely not limited to clinical contexts, but rather represent more of a general vulnerability to risk-taking and its subsequent problems.

The amygdala has long been acknowledged for its role in emotional processing. The amygdala responds to both negative and positive emotional stimuli, although findings are less robust with positive stimuli, related in part to the general lower arousal level of positive emotional stimuli (see Zald 2003). Additionally, negative stimuli tend to be lateralized to the left amygdala; there is less consistent lateralization of amygdala responding to positive stimuli (see Wager et al. 2003). The current study corroborates these patterns, showing responses in the left amygdala only to negative emotional stimuli, and no significant amygdala response to positive stimuli when compared with neutral stimuli. Responses in the amygdala were unrelated to self-reported valence and arousal, but this is not surprising since amygdala activation to emotional stimuli often occurs independently of conscious mood appraisal (see Zald 2003). Left amygdala responses to negative emotional stimuli were also related to negative urgency, further implicating the amygdala's function in personality trait dispositions of clinical relevance (Irwin et al. 1998; Canli et al. 2002; Schaefer et al. 2002).

The right lateral OFC, also related to negative urgency and risk-taking in the current study, receives direct inputs from the amygdala, which has reciprocal connections with the prefrontal cortex (see Ongur and Price 2000). In fact, Ongur and Price (2000) describe connections between area 12 (closest to our region of interest in the current study) and nuclei of the amygdala. Given previous work linking negative urgency with ventromedial prefrontal cortex activation to alcohol cues in these same subjects (Cyders et al. 2013), the current findings implicate a set of emotionally related brain regions that are responsive to negative emotional mood states and which, in turn, might influence negative mood-based rash action.

Previous work has yet to determine how negative urgency affects a wide range of risk-taking behaviors. This study suggests that BOLD activation to negative emotionally provocative content in the left amygdala and right lateral OFC is related to negative urgency, although not to self-reported mood changes. Additionally, negative urgency was itself largely unrelated to self-reported mood changes (as consistent with other work by Cyders et al. 2010). Such self-reported mood changes are often unrelated to amygdala function (Zald 2003) and are likely influenced by self-report and expectancy biases, making them only indirect proxies for neural responses to emotional stimuli. Therefore, other work that has reported no relationship between negative urgency and emotional responses may be limited by self-report methods in the assessment of mood responses (e.g., Cyders et al. 2010), with central nervous system physiological responses being a more direct and sensitive reflection of trait urgency's antecedent effects on behavior. As suggested by our models of mediation, negative urgency does not appear to impart risk for general risk-taking by increasing the limbic responses to emotional stimuli. Rather, increased left amygdala and lateral OFC activation from negative emotional stimuli would instead appear to engage negatively urgent impulses, which then lead to risky behavior.

Impulsivity is one of the most frequent criteria across disorders in the Diagnostic and Statistical Manual of Mental Disorders—5 (American Psychiatric Association 2013) and is a behavioral facet of borderline personality disorder, antisocial personality disorder, bulimia nervosa, attention deficit/hyperactivity disorder, mania, dementia, substance use disorders, paraphilias, and multiple other impulse-control disorders (Zapolski et al. 2010). Negative urgency may be the most clinically relevant impulsivity trait (see Smith et al. 2007), as it is related to many of these diagnoses (see Zapolski et al. 2010). Several lines of evidence have begun to indicate that negative urgency is related to multiple biological aspects, including genetic predispositions (Carver et al. 2011; Villafuerte et al. 2013) and brain responses (Joseph et al. 2009; Albein-Urios et al. 2013; Cyders et al. 2013). As our understanding of the cerebral foundations for this personality trait grows, we may then be able to determine some common mechanisms that underlie these disparate diagnostic entities. In addition, understanding the unique contribution of negative urgency to risk allows more targeted and effective prevention and treatment programs for vulnerable individuals. For example, Zapolski et al. 2010 propose that treatment approaches such as distress tolerance (such as found in Dialectical Behavior Therapy for Borderline Personality Disorder; Linehan 1987), experiencing emotions without externalizing action, and evaluating behavioral choices in terms of long-term goals might all be effective in reducing emotion-based risk-taking and the incidence of psychiatric illness.

There are limitations to the current study. First, the sample consists of social drinking, relatively young participants, which potentially limits generalizability of the findings to populations with more problematic behaviors. Second, the sample was modest, so null or trend-level results should be interpreted with caution, and the findings replicated in larger cohorts. In that vein, we could not compare responses across males and females, especially in regard to emotional reactions to positive emotional images, some of which were sexual in nature. The discrepant laterality of our effects (left amygdala, right OFC) is somewhat puzzling. This said, some research suggests a cerebral laterality in activation for externalizing behaviors (e.g., Berkman and Lieberman 2010; Speilberg et al. 2011), and this work further supports left lateralization of impulsive action at least in terms of the amygdala. The right lateral OFC activation was also related to negative urgency, further corroborating findings in the field relating activation in this region to emotional reactivity (Sabatinelli et al. 2011). Finally, the cross-sectional nature of the current study prevents a purely causal determination, and future research should seek to examine this model prospectively.

In summary, data in this study suggest that brain responses in regions known to respond to emotionally provocative stimuli are related to trait negative urgency and that negative urgency may mediate the relationship between these brain responses and the propensity to engage in risky behavior. This study provides further support for the role of negative urgency in multiple risk-taking behaviors and disorders, and it suggests a cerebral foundation for emotion-based rash action tendencies.

Supplementary Material

Supplementary Material can be found at http://www.cercor.oxfordjournals.org/online.

Funding

This work was supported by the NH (grant: K01AA020102 to MAC), The Indiana Alcohol Research Center (grant: P60 AA007611-26), and the Indiana Clinical Research Center (grant: M01 RR000750). DAK and MD were also supported in part by R01AA017661.

Notes

We also acknowledge the support of Michele Beal, Courtney Robbins and Rose Case. Conflict of Interest: None declared.

References

Albein-Urios
N
Verdejo-Roman
J
Sorlano-Mas
C
Asensio
S
Martinez-Gonzalez
JM
Verdego-Garcia
A
.
2013
.
Cocaine users with comorbid Cluster B personality disorders show dysfunctional brain activation and connectivity in the emotional regulation networks during negative emotion maintenance and reappraisal
.
Eur Neuropsychopharmacol
 .
23
:
1698
1707
.
American Psychiatric Association
.
2013
.
Diagnostic and Statistical Manual of Mental Disorders
 ,
5th ed
.
Arlington
(
VA
):
American Psychiatric Publishing
.
Anestis
MD
Smith
AR
Fink
EL
Joiner
TE
.
2009
.
Dysregulated eating and distress: examining the specific role of negative urgency in a clinical sample
.
Cognitive Ther Res
 .
33
:
390
397
.
Berkman
ET
Lieberman
MD
.
2010
.
Approaching the bad and avoiding the good: lateral prefrontal cortical asymmetry distinguishes between action and valence
.
J Cognitive Neurosci
 .
22
:
1970
1979
.
Brett
M
Johnsrude
IS
Owen
AM
.
2002
.
The problem of functional localization in the human brain
.
Nat Rev Neurosci
 .
3
:
243
249
.
Canli
T
Sivers
H
Whitfield
SL
Gotlib
IH
Gabrieli
JD
.
2002
.
Amygdala response to happy faces as a function of extraversion
.
Science
 .
296
:
2191
.
Carver
CS
Johnson
SL
Joorman
J
Kim
Y
Nam
JY
.
2011
.
Serotonin transporter polymorphism interacts with childhood adversity to predict aspects of impulsivity
.
Psychol Sci
 .
21
:
589
595
.
Coskunpinar
A
Dir
AL
Cyders
MA
.
2013
.
Multidimensionality in impulsivity and alcohol use: a meta-analysis using the UPPS model of impulsivity
.
Alcohol Clin Exp Res
 .
37
:
1441
1450
.
Costa
PT
McCrae
RR
.
1992
.
NEO-PI-R Professional Manual
 .
Odessa
(
FL
):
Psychological Assessment Resources
.
Cyders
MA
Coskunpinar
A
.
2010
.
Is urgency emotionality? Separating urgent behaviors from effects of emotional experiences
.
Pers Indiv Differ
 .
48
:
839
844
.
Cyders
MA
Dzemidzic
M
Eiler
WJ
Coskunpinar
A
Karyadi
KA
Kareken
DA
.
2013
.
Negative urgency and ventromedial prefrontal cortex responses to alcohol cues: fMRI evidence of emotion-based impulsivity
.
Alcohol Clin Exp Res
 .
38
:
409
417
.
Cyders
MA
Smith
GT
.
2008b
.
Clarifying the role of personality dispositions in risk for increased gambling behaviors
.
Pers Indiv Differ
 .
45
:
503
508
.
Cyders
MA
Smith
GT
.
2008a
.
Emotion-based dispositions to rash action: positive and negative urgency
.
Psychol Bull
 .
134
:
807
828
.
Cyders
MA
Smith
GT
.
2007
.
Mood-based rash action and its components: positive and negative urgency
.
Pers Indiv Differ
 .
43
:
839
850
.
Cyders
MA
Smith
GT
Spillane
NS
Fischer
S
Annus
AM
Peterson
C
.
2007
.
Integration of impulsivity and positive mood to predict risky behavior: development and validation of a measure of positive urgency
.
Psychol Assess
 .
19
:
107
118
.
Cyders
MA
Zapolski
TCB
Combs
JL
Settles
RF
Fillmore
MT
Smith
GT
.
2010
.
Experimental effect of positive urgency on negative outcomes from risk taking and on increased alcohol consumption
.
Psychol Addict Behav
 .
24
:
367
375
.
Dir
AL
Coskunpinar
A
Steiner
JL
Cyders
MA
.
2013
.
Understanding differences in sexting behaviors across gender, relationship status, and sexual identity, and the role of expectancies in sexting
.
Cyberpsychol Behav Soc Network
 .
16
:
568
574
.
Fischer
S
Smith
GT
.
2008
.
Binge eating, problem drinking, and pathological gambling: Linking behavior to shared traits and social learning
.
Pers Indiv Differ
 .
44
:
789
800
.
Fischer
S
Smith
GT
.
2004
.
Deliberation affects risk taking beyond sensation seeking
.
Pers Indiv Differ
 .
36
:
527
537
.
Fischer
S
Smith
GT
Annus
A
Hendricks
M
.
2007
.
The relationship of neuroticism and urgency to negative consequences of alcohol use in women with bulimic symptoms
.
Pers Indiv Differ
 .
43
:
1199
1209
.
Fischer
S
Smith
GT
Cyders
MA
.
2008
.
Another look at impulsivity: a meta-analytic review comparing specific dispositions to rash action in their relationship to bulimic symptoms
.
Clin Psychol Rev
 .
28
:
1413
1425
.
Irwin
W
Mock
BJ
Sutton
SK
Andermann
F
Kalin
NH
Sorenson
JA
Turski
PA
Davidson
RJ
.
1998
.
Ratings of affective stimulus characteristics and measures of affective reactivity predict MR signal change in the human amygdala
.
NeuroImage
 
7
:
S908
.
Joseph
JE
Liu
X
Jiang
Y
Lynam
D
Kelly
TH
.
2009
.
Neural correlates of emotional reactivity in sensation seeking
.
Psychol Sci
 .
20
:
215
223
.
Karyadi
KA
Coskunpinar
A
Cyders
MA
.
2012
.
Neurobiological underpinnings of dispositions to rash action
. In:
Cyders
MA
, editor.
Psychology of Impulsivity
 .
New York
:
Nova Science Publisher, Inc
. pp.
95
110
.
Kim
SH
Hamann
S
.
2007
.
Neural correlates of positive and negative emotion regulation
.
J Cognitive Neurosci
 .
19
:
776
798
.
Lane
RD
Reiman
EM
Bradley
MM
Lang
PJ
Ahern
GL
Davidson
RJ
Schwartz
GE
.
1997
.
Neuroanatomical correlates of pleasant and unpleasant emotion
.
Neuropsychologia
 .
35
:
1437
1444
.
Lang
PJ
Bradley
MM
Cuthbert
BN
.
1998
.
Emotion, motivation, and anxiety: Brain mechanisms and psychophysiology
.
Biol Psychiatry
 .
44
:
1248
1263
.
Lang
PJ
Bradley
MM
Cuthbert
BN
.
1999
.
International affective picture system (IAPS): instruction manual and affective ratings
 
(Tech. Rep. No. A-4)
.
Gainesville
(
FL
):
University of Florida, The Center for Research in Psychophysiology
.
Lang
PJ
Bradley
MM
Cuthbert
BN
.
1997
.
International Affective Picture System (IAPS): Technical Manual and Affective Ratings
 .
NIMH Center for the study of emotion and attention
.
Gainsville, FL: University of Florida
.
Levesque
J
Eugene
F
Joanette
Y
Paquette
V
Mensour
B
Beaudoin
G
Leroux
J-M
Bourgouin
P
Beauregard
M
.
2003
.
Neural circuitry underlying voluntary self- regulation of sadness
.
Biol Psychiatry
 .
53
:
502
510
.
Levesque
J
Joanette
Y
Mensour
B
Beaudoin
G
Leroux
JM
Bourgouin
P
Beauregard
M
.
2004
.
Neural basis of emotional self-regulation in childhood
.
Neuroscience
 .
129
:
361
369
.
Linehan
MM
.
1987
.
Dialectical behavior therapy for personality disorder: theory and method
.
Bull Menninger Clin
 .
51
:
261
276
.
Mak
AKY
Hu
Z
Zhang
JX
Xiao
Z
Lee
TMC
.
2009
.
Neural correlates of regulation of positive and negative emotions: an fMRI study
.
Neurosci Lett
 .
457
:
101
106
.
Michalczuk
R
Bowden-Jones
H
Verdejo-Garcia
A
Clark
L
.
2011
.
Impulsivity and cognitive distortions in pathological gamblers attending the UK national problem gambling clinic: a preliminary report
.
Psychol Med
 .
41
:
2625
2635
.
Ochsner
KN
Ray
RD
Cooper
J
Robertson
ER
Chopra
S
Gabrieli
JDE
Gross
JJ
.
2004
.
For better or for worse: neural systems supporting the cognitive down and up-regulation of negative emotion
.
Neuroimage
 .
23
:
483
499
.
Ongur
D
Price
JL
.
2000
.
The Organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans
.
Cereb Cortex
 .
10
:
206
219
.
Peters
JR
Upton
BT
Baer
RA
.
2013
.
Relationships between facets of impulsivity and borderline personality features
.
J Personal Disord
 .
27
:
547
552
.
Phan
KL
Wager
T
Taylor
SF
Liberzon
I
.
2002
.
Functional neuroanatomy of emotion: a meta- analysis of emotion activation studies in PET and fMRI
.
Neuroimage
 .
16
:
331
348
.
Preacher
KJ
Hayes
AF
.
2008
.
Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models
.
Behav Res Methods
 
40
:
879
891
.
Reiman
EM
Lane
RD
Ahern
GL
Schwartz
GE
Davidson
RJ
Friston
KJ
Yung
L
Chen
K
.
1997
.
Neuroanatomical correlates of externally and internally generated human emotion
.
Am J Psychiatry
 .
154
:
918
925
.
Russell
JA
Weiss
A
Mendelsohn
GA
.
1989
.
Affect grid: a single-item scale of pleasure and arousal
.
J Pers Soc Psychol
 .
57
:
493
502
.
Sabatinelli
D
Fortune
EE
Li
Q
Siddiqui
A
Krafft
C
Oliver
WT
Beck
S
Jeffries
J
.
2011
.
Emotional perception: meta-analyses of face and natural scene processing
.
NeuroImage
 .
54
:
2524
2533
.
Sabatinelli
D
Keil
A
Frank
DW
Lang
PJ
.
2013
.
Emotion perception: correspondence of early and late event-related potentials with cortical and subcortical function MRI
.
Biol Psychol
 .
92
:
513
519
.
Schaefer
SM
Jackson
DC
Davidson
RJ
Aguirre
GF
Kimberg
DY
Thompson-Schill
SL
.
2002
.
Modulation of amygdalar activity by the conscious regulation of negative emotion
.
J Cognitive Neurosci
 .
14
:
913
921
.
Smith
GT
Fischer
S
Cyders
MA
Annus
AM
Spillane
NS
McCarthy
DM
.
2007
.
On the validity and utility of discriminating among impulsivity-like traits
.
Assessment
 .
14
:
155
170
.
Speilberg
JM
Miller
GA
Engels
AS
Herrington
JD
Sutton
BP
Banich
MT
Heller
W
.
2011
.
Trait approach and avoidance motivation: lateralized neural activity associated with executive function
.
NeuroImage
 .
54
:
661
670
.
Thesen
S
Heid
O
Mueller
E
Schad
LR
.
2000
.
Prospective acquisition correction for head motion with image-based tracking for real-time fMRI
.
Magn Reson Med
 .
44
:
457
465
.
Verdejo-Garcia
A
Bechara
A
Recknor
EC
Perez-Garcia
M
.
2007
.
Negative emotion-driven impulsivity predicts substance dependence problems
.
Drug Alcohol Depend
 .
91
:
213
219
.
Villafuerte
S
Strumba
V
Stoltenberg
SF
Zucker
RA
Burmeister
M
.
2013
.
Impulsiveness mediates the association between GABRA2 SNPs and lifetime alcohol problems
.
Genes Brain Behav
 .
12
:
525
531
.
Wager
TD
Phan
KL
Liberzon
I
Taylor
SF
.
2003
.
Valence, gender, and lateralization of functional brain anatomy in emotion: a meta-analysis of findings from neuroimaging
.
NeuroImage
 .
19
:
513
531
.
Whiteside
SP
Lynam
DR
Miller
JD
Reynolds
SK
.
2005
.
Validation of the UPPS impulsive behavior scale: a four-factor model of impulsivity
.
Eur J Personality
 .
19
:
559
574
.
Zald
DH
.
2003
.
The Human amygdala and the emotional evaluation of sensory stimuli
.
Brain Res Rev
 .
41
:
88
123
.
Zapolski
TCB
Cyders
MA
Smith
GT
.
2009
.
Positive urgency predicts illegal drug use and risky sexual behavior
.
Psychol Addict Behav
 .
23
:
348
354
.
Zapolski
TCB
Settles
RF
Cyders
MA
Smith
GT
.
2010
.
Borderline personality disorder, bulimia nervosa, antisocial personality disorder, ADHD, substance use: common threads, common treatment needs, and the nature of impulsivity
.
Independent Practitioner
 .
30
:
20
23
.