Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jul 7.
Published in final edited form as: Assessment. 2021 Jan 7;29(3):572–582. doi: 10.1177/1073191120983889

Cross-Study, Cross-Method Associations Between Negative Urgency and Internalizing Symptoms

Kevin M King 1, Max A Halvorson 1, Kevin S Kuehn 1, Madison C Feil 1, Liliana J Lengua 1
PMCID: PMC8260640  NIHMSID: NIHMS1677393  PMID: 33412920

Abstract

There is a small body of research that has connected individual differences in negative urgency, the tendency to report rash actions in response to negative emotions, with self-report depressive and anxiety symptoms. Despite the conceptual overlap of negative urgency with negative emotionality, the tendency to experience frequent and intense negative emotions, even fewer studies have examined whether the association of negative urgency with internalizing symptoms hold when controlling for negative emotionality. In the current study, we estimated the bivariate association between negative urgency and internalizing symptoms, tested whether they remained significant after partialling out negative emotionality, and tested whether these effects generalized to real-time experiences of negative emotions. We used data from five independent samples of high school and college students, assessed with global self-report (n = 1,297) and ecological momentary assessment (n = 195). Results indicated that in global self-report data, negative urgency was moderately and positively associated with depressive and anxiety symptoms, and the partial association with depressive symptoms (but not anxiety symptoms) controlling for negative emotionality remained significant and moderate in magnitude. This pattern was replicated in ecological momentary assessment data. Negative urgency may convey risk for depressive symptoms, independent of the effects of negative emotionality.

Keywords: self-regulation, negative urgency, internalizing symptoms, negative emotionality

Negative urgency, the tendency toward rash action in response to strong negative emotions (Cyders & Smith, 2008), is weakly to moderately associated with a wide range of psychopathology, and has been proposed as a potential transdiagnostic personality risk factor (Berg et al., 2015; Coskunpinar et al., 2013; Fischer et al., 2008; Smith et al., 2007). While early research focused on the association between urgency and externalizing disorders characterized by impulsive action, recent meta-analytic work suggests an even more robust association between negative urgency and internalizing disorders such as depression and anxiety (Berg et al., 2015). However, the evidence base for this is still relatively small, and relatively few studies considered how much of the association between negative urgency and internalizing symptoms was accounted for by generally higher levels of negative emotionality among those high in negative urgency. The current study aims to explore this question using a mix of global self-report and ecological momentary assessment (EMA) measures across multiple samples.

Negative Urgency and Internalizing Disorders

Theories of negative urgency hypothesize that this trait reflects a process where impulsive behaviors are negatively reinforced because they relieve the experience of distressing emotions (Cyders & Smith, 2008). Early research on urgency focused on externalizing behaviors that reflect dysregulated action that indirectly leads to problem outcomes such as binge eating, gambling, or substance use (Coskunpinar et al., 2013; Smith et al., 2007). More recent conceptualizations of negative urgency have suggested that it may also lead to internalizing symptoms because it reflects a more general tendency toward dysregulation in the face of negative affect, which may take the form of either dysregulated action or dysregulated inaction (such as avoidance or disengagement) aimed at relieving or avoiding distress. Indeed, we previously showed that negative urgency was strongly correlated with both reactive (such as catastrophizing and rumination) and disengagement (such as denial and avoidance) forms of emotion regulation using data from some of the same samples as the current study (King et al., 2018). Although these behaviors are likely aimed at relieving distress, they are also likely to have a backfire effect, where they inadvertently increase negative emotions in the long term because they fail to address the source of the negative emotion or stress. Moreover, as behaviors like withdrawal or avoidance are negatively reinforced, they may turn into symptoms of internalizing problems over time (Johnson et al., 2013; Smith & Cyders, 2016).

A recent meta-analysis of data from 15 studies with depressive symptoms (n = 5,162) and 14 studies with anxiety symptoms (n = 4,655) reported moderate correlations between negative urgency and both depressive (r = .45, 95% confidence interval [CI: 0.39, 0.41]) and anxiety (r = .40, 95% CI [0.32, 0.49]) symptoms (Berg et al., 2015). Although hundreds of prior studies had reported on the association between negative urgency and externalizing disorders, urgency’s association with internalizing symptoms were stronger than those reported in meta-analyses of the association between negative urgency and all other externalizing disorders, except borderline personality disorder (Berg et al., 2015). These findings suggest a robust, but relatively underexplored association between negative urgency and internalizing disorders, compared with externalizing disorders. Thus, the first aim of the current study was to contribute to the growing literature connecting negative urgency to internalizing symptoms. Using data from four samples (total n = 1,297), we examined the association of negative urgency with global self-reports of depressive and anxiety symptoms. To directly bolster the accumulated literature on the subject, we pooled our estimate with the meta-analytic estimate previously reported by Berg et al. (2015), increasing the sample size of the prior meta-analysis by 25% to 28%. The addition of our data provides further information about the stability of these associations across samples while also increasing the precision of the estimate.

Urgency and Momentary Negative Emotions

Very little research has examined the momentary emotional experiences associated with negative urgency. In the current study, we used EMA to measure the negative mood states that underlie internalizing symptoms in near real-time. Although negative mood states do not fully capture the real-time experience of anxiety or depression, persistent and strong negative moods have been consistently associated with internalizing disorders (Ebner-Priemer & Trull, 2009; Silk et al., 2011; Walz et al., 2014). Moreover, if it is true that negative urgency is characterized by impulsive and ineffective attempts at emotion regulation, we would expect to observe this backfire effect as an association between urgency and momentary negative emotions. One prior EMA study (n = 294) reported that negative urgency was correlated (r = .37) with dysphoria, a composite of two items indicating sadness or anxiety (Sperry et al., 2018). Thus, the final aim of the current study was to examine the association between negative urgency and ecologically valid measures of negative emotions core to internalizing disorders (unhappiness, anxiety, and generalized negative affect).

The Role of Emotionality

Although there is evidence of an association between negative urgency and internalizing symptoms, it remains unclear whether this association is better accounted for by shared variance between negative urgency and other traits that describe individual differences in people’s emotional experiences, such as negative emotionality and emotion reactivity. Negative emotionality is a trait that is described as a stable disposition to experience negative affect and distress (Tackett et al., 2013), while emotion reactivity is a tendency to experience intense and prolonged negative emotions (Nock et al., 2008). Negative urgency has been theorized to be associated with both negative emotionality (Cyders & Smith, 2008; Whiteside & Lynam, 2001) and emotional reactivity (Billieux et al., 2010; Lannoy et al., 2014). Both emotionality and reactivity have been associated with internalizing symptoms (Bylsma et al., 2008; Eisenberg et al., 2010; Klein et al., 2011; McLaughlin et al., 2010; Newman et al., 2013). It is possible that the relations between urgency and internalizing symptoms are attributed to a shared association with either negative emotionality or emotional reactivity. In other words, perhaps individuals high on negative urgency are prone to internalizing symptoms simply because they tend to experience high levels of negative emotions, or to experience strong emotional reactions, rather than their tendency to become behaviorally dysregulated in the face of such emotions. One study suggested the association of negative urgency with internalizing symptoms was independent of their shared covariation with negative emotionality (Smith et al., 2013), but we are aware of few if any studies that replicate this finding, or any studies that focused specifically on emotional reactivity. Thus, the third aim of the current study was to assess the degree to which the associations between urgency and internalizing symptoms were robust to controlling for measures of negative emotionality or emotional reactivity.

We tested the following hypotheses. In Samples 1 and 4 with cross-sectional data, (a) negative urgency would be moderately and positively correlated with depressive and anxiety symptoms (Aim 1). In Samples 4 and 5, with EMA data, we would observe (b) similar associations between negative urgency and average levels of negative affect, unhappiness, and anxiety. Finally, the correlations between urgency and (c) depressive and anxiety symptoms and (d) real-time negative emotions would be robust to the inclusion of negative emotionality or emotional reactivity (Aim 3).

Method

Participants

Table 1 provides demographic information on the samples used for the current study. All participants were independently sampled between 2008 and 2018. Procedures for all studies were approved by the institutional review board of the local university where data were collected. We have previously reported on analyses with other variables with Sample 1 (King et al., 2011), and with Samples 3, 4, and 5 (King et al., 2018).

Table 1.

Demographics and Descriptions of Samples.

Sample 1 Sample 2 Sample 3 Sample 4 Sample 5
Sample type College College Community College Community
Data collection method Cross-sectional Cross-sectional Cross-sectional Cross-sectional/EMA EMA
Internalizing measure ASR (G) ASR (G), BDI (G) CDI (G), MASC (G) PROMIS (G), PANAS (E) PANAS (E)
Negative emotionality measure N/A STAI, STAEI ERS ERS ERS
Hypotheses tested 1 1,2 1,2 1,2,3,4 3,4
Total N 457 413 283 144 51
% Female 57 52 55 53 66
Age, M (SD); range 19.26 (1.22); 18–24 18.46 (0.84); 18–23 16.52 (0.50); 16–17 19.45 (0.69); 18–22 16.62 (0.99); 15–19
% White 55 55 41 62 57
% Asian/Pacific Islander 33 28 16 37 14
% Other ethnicities 12 17 21 1 29
Open in a new tab

Note. ASR = Achenbach Adult Self Report; BDI = Beck Depression Inventory–II; CDI = Children’s Depression Inventory; MASC = Multidimensional Anxiety Schedule for Children; PROMIS = Patient Reported Outcomes Measurement Information System; PANAS = Positive and Negative Affect Schedule; (G) = Global self-report measure; (E) = Ecological momentary assessment measure; EMA = ecological momentary assessment measure; STAI = State-Trait Anxiety Inventory; STAEI = State-Trait Anger Expression Inventory; ERS = Emotional Reactivity Scale.

Samples 1 and 2.

Participants were undergraduates (n = 457 and n = 413, respectively) at a university in the Pacific Northwest who received course credit for survey participation. Content of the questionnaires varied, but in both cases, participants completed the study in a single lab-based session using a self-guided computerized interview.

Sample 3.

Participants were adolescents (n = 283) in a multisite community sample from three major cities in the United States (Boston, MA; Seattle, WA; and Pittsburgh, PA). Participants were recruited from the general community using advertisements in public transportation and other public places, Craigslist, and community centers. The current study used data from the baseline session of a multivisit data collection. Participants were paid $30 for their baseline interview.

Sample 4.

Participants (n = 144) were undergraduate students with an internet enabled smartphone at a university in the Pacific Northwest who received course credit for study participation. Following the completion of an in-lab questionnaire and a training session, participants completed 10 days of EMA. Each day, participants received three text messages per day (at random times during the morning, midday and evening) with a link to a brief web-based survey. Participants received surveys at least 2 hours apart and had two hours to complete each survey with one text message reminder for each survey. We obtained 3,707 total observations out of 4,410 possible, for a response rate of 84%. Only five participants (3.5%) completed fewer than 50% of EMAs; the response rate among those completing more than 50% of EMAs was 88.2%. Nine participants failed to complete more than 3 days of EMA and were dropped from analyses. There were no differences between included and excluded participants on the baseline data.

Sample 5.

Participants (n = 51) were high school students at a single high school in the Pacific Northwest who agreed to participate in an EMA study as part of a larger examination of a school-wide intervention. Following an online baseline assessment and telephone training session with a research assistant, participants completed the same EMA protocol described for Sample 4. Participants were paid $1 per EMA and $5 for the baseline survey. We obtained 1,406 total observations out of 1,830 possible, for a response rate of 77%. Only six (10%) participants completed fewer than 50% of EMAs; the response rate among those completing more than 50% of EMAs was 77.3%. We again excluded data from participants who completed less than 3 full days of data collection (i.e. ≥9 total EMAs) from the final analysis (n = 51). There were no differences between included and excluded participants on the baseline data.

Measures

Toward the aim of open and transparent science, we analyzed any data we had collected that included measures of negative urgency and internalizing symptoms, and any data that also had measures of negative emotionality. Across samples, we collected data with a variety of measures of internalizing symptoms and negative emotionality. This is both a potential strength and weakness. The degree to which we observed consistency in effects across studies with different measures improves the generalizability of the findings in the current analysis because it suggested that the associations we observed were not due to method effects. At the same time, it makes interpretation of inconsistent or null findings more difficult because we could not distinguish true null effects from method effects.

Negative urgency was measured in all samples using 12 items from the UPPS Impulsive Behavior scale (Whiteside & Lynam, 2001). Items included “when I am upset, I often act without thinking” and “It is hard for me to resist acting on my feelings.” Response options ranged from 1 (disagree strongly) to 4 (agree strongly) for all items. We computed a mean score for negative urgency. Cronbach’s alpha was high across studies (α >.85 for all studies).

Depressive symptoms.

Samples 1 and 2 used the Achenbach Adult Self-Report (Achenbach et al., 2001). Participants reported on their experiences in the past 6 months. Sample items included “I felt lonely,” and “I felt worthless or inferior.” Response options ranged from 1 (not at all like me) to 3 (very much like me). Depressive (15 items) symptom scores were obtained by using the Achenbach Diagnostic and Statistical Manual of Mental Disorders (DSM) oriented scales, which utilize clusters of items from the Achenbach Adult Self Report that were rated by experts to reflect DSM-based depressive symptoms (Achenbach et al., 2005, α = .79–.81). Sample 2 also used the Beck Depression Inventory–II (BDI-II; Beck et al., 1988), which is a symptom-oriented scale that is used to measure symptoms of depression. It consists of 21 items (we excluded an item on suicidality) that ask participants to choose from among three options to describe how they have felt in the past 2 weeks, α = .89.

Sample 3 used the Children’s Depression Inventory (CDI), a 26-item adaptation of the BDI suitable for children and adolescents (Saylor et al., 1984), α = .86. For Samples 1 to 3, we computed a mean score of depressive symptoms.

Samples 4 used 8 items from the short form of the Patient Reported Outcomes Measurement Information System (PROMIS) inventory depression scale (Reeve et al., 2007). The PROMIS has very strong psychometric properties, evidence of validity in both community and clinical samples (Schalet et al., 2016), and has nationally published norms. We computed a mean score for this scale and converted it to T-scores based on published norms (available at http://healthmeasures.net).

Sample 5 did not have a global self-report measure of recent internalizing symptoms.

Anxiety symptoms.

Samples 1 and 2 used the Achenbach Adult Self-Report, described above. We used the DSM Oriented Problems Anxiety Subscale (Achenbach et al., 2005), composed of seven items such as “I am too fearful or anxious” and “I worry a lot,” α = .75–.77.

Sample 3 used the Multidimensional Anxiety Scale for Children (MASC-C; March et al., 1997), a 39-item scale measuring a child’s most recent feelings of anxiety. The MASC captures four dimensions of anxiety: physical symptoms (tense/restless and somatic/autonomic), social anxiety (humiliation/rejection and public performance fears), harm avoidance (perfectionism and anxious coping) and separation anxiety. Participants respond on a 4-point Likert-type scale ranging from 1 (none of the time) to 4 (a lot of the time). We computed a mean of all items, α = .89.

Samples 4 used 8 items from the short form of the PROMIS inventory anxiety scale (Schalet et al., 2016). As with depressive symptoms, we computed a mean score for this scale and converted it to T-scores based on published norms.

Sample 5 did not have a global self-report measure of recent internalizing symptoms.

Ecological momentary assessment of negative emotions.

Samples 1 to 3 did not include an EMA measure of negative emotions.

In Samples 4 and 5, we measured the recent experience of negative emotions using items from the Positive and Negative Affect Schedule (Watson et al., 1988). Items included “irritable,” “unhappy,” “bored,” “anxious,” and “angry.” At each assessment, participants rated their experience of each emotion since the last assessment (or since they woke up) on a visual sliding scale with end anchors of “not at all” to “very much.” A score was generated from 0 to 100 based on the placement of the slider, though numbers were not visible to participants. We computed person-level negative affect as the mean of affect items across all measurement occasions. We also examined “unhappy” and “anxious” independently as the person-level mean of responses to those items. To estimate the reliability of the EMA measures when averaged across time, we estimated variance components as recommended by Shrout and Lane (2014) using the psych (Revelle, 2015) software package in R and the function multilevel.reliability. We estimated the coefficient RKR, representing the stability of rank-ordering of individuals in their levels of negative affect, which was high for negative affect, RKR = .95, in both samples. The reliability of the single item measures, estimated as intra-class correlations (ICC2) across 30 ratings (ICC2k) also using the psych package and the function ICC, was high in both samples, unhappiness ICC2k = .91–.91, and anxiety ICC2k = .94–.95.

Sample 4 had data on both depressive and anxiety symptoms at baseline, and reports of recent negative affect, unhappiness, and anxiety. Depressive symptoms were moderately and positively correlated with recent reports of negative affect (r = .46), unhappiness (r = .43), and anxiety (r = .39), all p < .001. Anxiety symptoms were also moderately and positively correlated with recent reports of negative affect (r = .48), unhappiness (r = .43), and anxiety (r = .47), all p < .001.

Negative emotionality.

Sample 1 did not have proxy variables for negative emotionality. In Sample 2, we measured trait anger and anxiety (Spielberger et al., 1999) using the State-Trait Anxiety Inventory and the State-Trait Anger Expression Inventory. Participants responded to 10 anger and 20 anxiety items asking how they generally feel using a 4-point Likert-type scale ranging from 1 (almost never) to 4 (almost always). Sample items included “I am quick tempered,” “I fly off the handle,” and “I feel nervous and restless” and “I am a steady person.” All items were coded such that 4 = more trait anger (α = .83) or anxiety (α = .92).

Emotion reactivity.

In Samples 3, 4 and 5, we used the Emotional Reactivity Scale (ERS; Nock et al., 2008). The ERS is a 21-item self-report measure designed to assess individuals’ experience of emotion reactivity. Prior research has shown the ERS to be moderated to strongly correlated with temperamental measures of negative emotionality (r = .30–.61; Nock et al., 2008). Though the ERS is a general measure of emotionality, its items either ask about negative emotions (“My feelings get hurt easily”) or about any emotions of either valence (“Even the littlest things make me emotional”). No items index positive emotions in particular. The ERS measures the extent to which an individual experiences emotions (a) in response to a wide array of stimuli (i.e., Emotion Sensitivity); (b) strongly or intensely (i.e., Emotion Intensity); and (c) for a prolonged period of time before returning to baseline level of arousal (i.e., Emotion Persistence). Participants were asked about how they experience emotions on a regular basis, responding on a 5-point Likert-type scale ranging from 1 (not at all like me) to 5 (very much like me). Items included “when I experience emotions, I feel them very strongly/intensely” and “even the littlest things make me emotional.” To index general negative emotionality, we used the full-scale ERS score (α > .89). Because we were working with existing data, the ERS was scored as a sum score in Sample 3, and as a mean in Samples 4 and 5. Because one is a linear transformation of the other, we did not change the ERS sum scores to means in Sample 3. Although missing data can bias summed scores more so than mean scores, there was no item level missingness in the ERS in Sample 3.

Analytic Approach

We used meta-analytic methods, which can be applied to small samples to integrate information across even a few studies (Goh et al., 2016). We conducted all analyses in R 3.4.1 (R Core Team, 2017) using the package meta (Schwarzer, 2007). We used random effects meta-analysis to account for cross-sample variation in the effect size due to method or demographic differences across studies (Borenstein et al., 2009), and used the restricted maximum likelihood estimator to estimate between study variances (Veroniki et al., 2016). This provided us with an estimate of the pooled correlation, which was computed as the weighted mean of all observed correlations (weighted by within and between study variance), and a CI of this estimate. Because Sample 2 had two measures of depression, we averaged their estimated correlations with negative urgency by first transforming the correlation to a Fisher’s Z score, averaging, and then back transforming the average to a correlation.

Results

Descriptive statistics for the samples are presented in Tables 1 and 2. Table 3 present results for global self-reports, while Table 4 presents results for reports of recent negative emotions.

Table 2.

Means and SD for Variables of Interest Across Samples.

Global self-reports Sample 1 Sample 2 Sample 3 Sample 4 Sample 5
M SD M SD M SD M SD M SD
Negative urgency 2.11 0.66 2.20 0.69 2.42 0.83 2.34 0.59 2.45 0.52
Depression symptoms
Achenbach Depressive Symptoms 1.31 0.28 1.41 1.00
Beck Depression Inventory 1.43 0.37
Child’s Depression Inventory 10.40 6.89
PROMIS Depressive Symptoms T-Score 49.58 10.63
Anxiety symptoms
Achenbach Anxiety Symptoms 1.64 0.43 1.60 0.43
Multidimensional Anxiety Scale for Children 1.99 0.40
PROMIS Anxiety Symptoms T-Score 54.74 8.98
Negative emotionality
Trait Anxiety Scale 2.05 0.51
Trait Anger Scale 1.79 0.47
Emotional Reactivity Scale 27.61 16.64 2.32 0.78 2.59 0.77
EMA reports
Unhappy 16.23 11.81 26.30 12.67
Anxious 25.56 16.2 38.84 18.34
Negative mood (average) 19.38 10.85 29.67 11.91
Open in a new tab

Note. EMA = ecological momentary assessment.

Table 3.

Hypotheses 1 and 3: Associations Between Negative Urgency and Global Self-Reports of Depressive and Anxiety Symptoms.

Hypothesis 1: Bivariate association Hypothesis 3: Partial correlation controlling for negative emotionality Hypothesis 3: Partial correlation controlling for emotional reactivity
r 95% CI r 95% CI r 95% CI
Depressive symptoms
Sample 1 (ASR) 0.39 [0.31, 0.46] — — — —
Sample 2 (ASR + BDI) 0.47 [0.39, 0.55] 0.18 [0.10, 0.32] — —
Sample 3 (CDI) 0.43 [0.34, 0.52] — — 0.16 [0.04, 0.27]
Sample 4 (PROMIS) 0.46 [0.33, 0.59] — — 0.19 [0.04, 0.35]
Pooled effect 0.47 [0.35, 0.48] — — 0.17 [0.08, 0.27)
Anxiety symptoms
Sample 1 (ASR) 0.31 [0.22, 0.39] — — — ——
Sample 2 (ASR) 0.39 [0.31, 0.47] 0.04 [−0.07, 0.13] — —
Sample 3 (MASC) 0.21 [0.10, 0.32] — — −0.09 [−0.20, 0.03]
Sample 4 (PROMIS) 0.53 [0.40, 0.63] — — 0.24 [0.08, 0.39]
Pooled effect 0.37 [0.23, 0.52] — — 0.08 [−0.24, 0.38]
Open in a new tab

Note. 95% Confidence interval (CI) for bolded coefficients do not contain zero. ASR = Achenbach Adult Self Report; BDI = Beck Depression Inventory–II; CDI = Children’s Depression Inventory; MASC = Multidimensional Anxiety Schedule for Children; PROMIS = Patient Reported Outcomes Measurement Information System.

Table 4.

Hypotheses 2 and 4: Association Between Negative Urgency and EMA Reports of Negative Affect, Unhappiness, and Anxiety.

Hypothesis 2: Bivariate Association Hypothesis 4: Partial correlation controlling for emotional reactivity
r 95% CI r 95% CI
Negative affect
Sample 4 0.45 [0.31, 0.57] 0.21 [0.05, 0.36]
Sample 5 0.32 [0.07, 0.53] 0.28 [0.02, 0.55]
Pooled effect 0.42 [0.29, 0.52] 0.27 [0.09, 0.47]
Unhappy
Sample 4 0.38 [0.23, 0.51] 0.21 [0.04, 0.36]
Sample 5 0.33 [0.05, 0.55] 0.28 [0.01, 0.52]
Pooled effect 0.37 [0.24, 0.48] 0.23 [0.09, 0.36]
Anxious
Sample 4 0.45 [0.31, 0.57] 0.23 [0.07, 0.38]
Sample 5 0.18 [−0.10, 0.44] 0.02 [−0.25, 0.30]
Pooled effect 0.34 [0.07, 0.57] 0.16 [−0.03, 0.34]
Open in a new tab

Note. 95% Confidence interval (CI) for bolded coefficients do not contain zero. EMA = ecological momentary assessment.

Hypothesis 1: Cross-Sectional Associations With Global Self-Reports (Samples 1 and 4)

Depressive Symptoms.

We first examined the correlation between negative urgency and depressive symptoms. For depression, all associations were positive and significant (r = .39–.47), and the pooled correlation was nearly identical (r = .47) to that reported in previous research (Berg et al., 2015, r = .45, 95% CI [0.39, 0.51]). Meta-analyzing our data with that previously reported estimate, the revised pooled correlation estimate was identical, r = .45, and more precise, 95% CI [0.43, 0.47].

Anxiety Symptoms.

For anxiety, all associations were also positive and significant (r = .21–.53), with a pooled effect of r = .37. When pooled with the estimate from Berg et al. (2015; r = .40, 95% CI [0.32, 0.48]), the revised pooled correlation estimate was again identical, r = .40, and more precise, 95% CI [0.37, 0.42].

Hypothesis 2: Associations of Urgency With Real-Time Emotions (Samples 4 and 5)

We examined the correlation of negative urgency with recent negative affect, unhappiness, and anxiety, as measured using EMA (see Table 4). Across both studies, negative urgency was moderately correlated with recent negative affect averaged across all EMAs (r = .42), and with recent unhappiness (r = .37) and recent anxiety (r = .34) specifically.

Hypothesis 3: Associations Controlling for Negative Emotionality or Emotional Reactivity

Depressive Symptoms.

The partial correlation between negative urgency and depression in our four samples was robust to covarying the effects of negative emotionality or emotional reactivity. For the sample controlling for emotionality, the estimate fell to .18, while for the two samples controlling for reactivity, the pooled estimate fell to r = .17, although with less precision in the effect estimate (95% CI = [.07, .27]). The attenuation of effects was consistent across samples, regardless of whether we controlled for negative emotionality (Sample 2) or reactivity (Samples 3 and 4).

Anxiety Symptoms.

The partial correlation between negative urgency and anxiety symptoms was not robust to covariation of negative emotionality. In Sample 2, which controlled for negative emotionality, the association fell to nonsignificant and small in magnitude (r = .04), as did the pooled association controlling for emotional reactivity (r = .08). Again, this attenuation was consistent regardless of the control variable.

Hypothesis 4: Real-Time Associations Controlling for Emotional Reactivity (Samples 4 and 5)

When partialling the effects of emotional reactivity, these associations weakened, but remained significant for average negative emotions (r = .27) and unhappiness (r = .23), but not anxiety (r = .15).

Discussion

Negative urgency is thought to produce risk for both externalizing and internalizing psychopathology because it reflects both automatic action and inaction in the face of distress (Johnson et al., 2013; Smith & Cyders, 2016). Urgency has been linked to the experience of depressive and anxiety symptoms in both global self-report and EMA data (Berg et al., 2015; Sperry et al., 2018). The current study provides further evidence that urgency is associated with the experience of internalizing symptoms as well as negative emotions central to internalizing disorders, and that this bivariate association is consistent across global self-report of symptoms and in-the-moment report of emotions. Our findings were highly consistent with prior work, and the inclusion of 1,297 additional participants left the estimate of the association between negative urgency and depressive symptoms unchanged while increasing the precision of the estimate (as measured by the CI) by 66%. We observed similar findings for anxiety symptoms: the bivariate association was unchanged, and the CI shrank by 32%.

Importantly, the associations of negative urgency with depressive symptoms and momentary unhappiness were robust, although reduced in magnitude, to the inclusion of measures of negative emotionality and emotional reactivity, which specifically indexed the frequency, intensity, and persistence of emotional experiences. On the other hand, its partial association with anxious symptoms controlling for negative emotionality or emotional reactivity was not significant in either global self-report or EMA data, regardless of which emotional trait was covaried. Relatively few studies had previously tested the robustness of the urgency-internalizing association to the inclusion of emotionality or reactivity, although both constructs measure aspects of emotional dysregulation. These findings suggest that the relation between anxiety and negative urgency may be driven by common variance in urgency and emotionality or reactivity. This could be driven either a propensity to experience negative emotions or strong emotional reactivity, either of which increases the chances of acting impulsively. It is possible that urgency or impulsivity is associated with avoidance or lack of activation that is not driven by strong emotions and that contribute to depressive symptoms independently (Smith & Cyders, 2016). It is also curious to note that only the high school samples (Samples 3 and 5) demonstrated weaker associations of urgency with anxiety symptoms or feelings of anxiety that were reduced to nonsignificant when controlling for negative emotionality. Although this is a very tentative hypothesis, it is possible that a unique association between urgency and anxiety only emerges in young adulthood.

Since its formulation by Whiteside and Lynam (2001), urgency has claimed a central place in research on impulsive traits and rash behavior/psychopathology. Urgency is most strongly correlated with outcomes across both the internalizing and externalizing spectra, and is represented well in most measures of impulsive traits (Berg et al., 2015; Sharma et al., 2014). However, the process by which the personality trait of urgency (i.e., the tendency to report impulsive acts in response to strong emotions) precipitates impulsive behaviors in the moment remains unarticulated. Carver et al. (2008) and Smith and Cyders (2016) have proposed that urgency reflects a reflexive responsivity to emotion. In the face of either strong positive or strong negative affect, some individuals tend to react automatically and without deliberation, which manifests as reflexive action or inaction. For example, we previously reported that negative urgency was robustly associated with emotion regulation styles characterized by disengagement or reactivity (King et al., 2018). As depression is often characterized as a disorder of inaction (anhedonia, low behavioral activation, hypersomnia, and fatigue), the unique association between negative urgency and depression that remains after controlling for negative emotionality could reflect this reactive inaction. Future research should aim to understand how internalizing symptoms emerge across development among those high on negative urgency, with particular attention to how the common and unique components of depressive and anxiety symptoms (such as inaction and emotional reactivity) may lead to differentiation over time.

One of the primary contributions of our study was to replicate and update current understanding of links between urgency and psychopathology, and in doing so, to contribute to cumulative psychological science. Calls for replication, confirmation, and extension of past empirical findings have been at the forefront of psychological science in recent years (Aarts et al., 2015; Ledgerwood, 2014; Mischel, 2006; Simonsohn et al., 2014). Statistical comparisons with past literature, including previous meta-analyses, can help theory and empirical summary to converge on stable consensus (Schmidt, 1996; Tsuji et al., 2014). By pooling our estimates with a prior meta-analysis, we hope that our work represents the type of summary necessary to build reliable knowledge across researchers, samples, and time.

The present study has some limitations. Except for Sample 3, our samples reflect convenience samples, and our findings may not generalize to more representative, diverse, or clinical samples. Sample 5 is especially small and was the largest outlier in terms of effect sizes and CIs, suggesting its information is less reliable than that from other samples. That said, we thought it was important to include data from Sample 5 both because it was a unique sample (EMA from high school students) and because the goal of this study was to report all data we had available. Moreover, we did not use clinical cutoffs, or EMA assessment of depressive or anxiety symptoms other than negative emotions. Our measurement of negative emotionality in Sample 2 substantially overlapped with anxiety symptoms, which likely reduced the shared variance between urgency and anxiety symptoms. Our measurements varied somewhat across studies, which may have increased variability in the observed correlations. However, it is important to note that negative emotionality is composed of tendencies toward depression, anxiety, and emotional reactivity in common models of personality (Soto & John, 2017), making our use of the trait anxiety and anger scales defensible in the current study. Our measure of negative emotionality was identical across all samples except Sample 2 and excluding the findings from Sample 2 did not change our inferences substantially. Finally, there are important concerns about inferences of partial correlations drawn from measures with low reliability (Westfall & Yarkoni, 2016). Future research should use structural equation modeling to test for shared variance between negative urgency and internalizing symptoms, after controlling for negative emotionality and accounting for measurement error.

Combined with the weight of the evidence for its role in externalizing symptoms (Foulds et al., 2015; Smith et al., 2007), this study provides further evidence that negative urgency may be a transdiagnostic risk factor for multiple forms of psychopathology. Future research should examine potential mechanisms (such as emotion regulation and appraisal systems; King et al., 2018) and explore how the association of negative urgency and internalizing symptoms emerges across development.

Acknowledgments

We thank Kate McLaughlin and Kathryn Monahan for their collaborative data collection efforts for Sample 3, and the many graduate and undergraduate students who assisted in collecting the data for all samples. Sample 2 data collection was supported by a grant from the Alcohol Beverage Manufacturer’s Association to Dr. King. Sample 3 data collection was supported by a Young Scholars Award to Dr. King, shared with Dr. McLaughlin and Dr. Monahan, and Sample 5 data collection was supported with an award from the Mindful Living Initiative at the Center for Child and Family Well-Being to Dr. King.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Preparation of this article was supported by a grant from the National Institute on Drug Abuse (DA047247), the National Institute on Alcohol Abuse and Alcoholism (AA027118), and the National Institute on Mental Health (MH117827). Sample 2 data collection was supported by a grant from the Alcohol Beverage Manufacturer’s Association to Dr. King. Sample 3 data collection was supported by a Young Scholars Award to Dr. King, shared with Dr. McLaughlin and Dr. Monahan, and Sample 5 data collection was supported with an award from the Mindful Living Initiative at the Center for Child and Family Well-Being to Dr. King.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

1.

The MASC has 4 subscales: harm avoidance, physical symptoms, social anxiety, and separation anxiety. Negative urgency was negatively correlated (r = −.17) with harm avoidance, and positively correlated (r = .18–.33) with the remaining subscales, although all subscales were positively correlated with one another (r = .24–.81). Although it is possible that this negative correlation attenuated the meta-analytic estimate, analyzing only the three positively correlated subscales did not substantially change the conclusions of the meta-analysis.

References

  1. Aarts AA, Anderson JE, Anderson CJ, Attridge PR, Attwood A, Axt J, Babel M, Bahník S, Baranski E, Barnett-Cowan M, Bartmess E, Beer JS, Bell R, Bentley H, Beyan L, Binion G, Borsboom D, Bosch A, Bosco FA, … Zuni K (2015). Estimating the reproducibility of psychological science. Science, 349(6251), Article 4716. 10.1126/science.aac4716 [DOI] [PubMed] [Google Scholar]
  2. Achenbach TM, Bernstein A, & Dumenci L (2005). DSM-oriented scales and statistically based syndromes for ages 18 to 59: Linking taxonomic paradigms to facilitate multitaxonomic approaches. Journal of Personality Assessment, 84(1), 49–63. 10.1207/s15327752jpa8401_10 [DOI] [PubMed] [Google Scholar]
  3. Achenbach TM, Dumenci L, & Rescorla LA (2001). Ratings of relations between DSM-IV diagnostic categories and items of the CBCL/6–18, TRF, and YSR. https://aseba.org/wp-content/uploads/2019/02/DSM6-18ratings.pdf
  4. Beck AT, Steer RA, & Carbin MG (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review, 8(1), 77–100. 10.1016/0272-7358(88)90050-5 [DOI] [Google Scholar]
  5. Berg JM, Latzman RD, Bliwise NG, & Lilienfeld SO (2015). Parsing the heterogeneity of impulsivity: A meta-analytic review of the behavioral implications of the UPPS for psychopathology. Psychological Assessment, 27(4), 1129–1146. 10.1037/pas0000111 [DOI] [PubMed] [Google Scholar]
  6. Billieux J, Gay P, Rochat L, & Van der Linden M (2010). The role of urgency and its underlying psychological mechanisms in problematic behaviours. Behaviour Research and Therapy, 48(11), 1085–1096. 10.1016/j.brat.2010.07.008 [DOI] [PubMed] [Google Scholar]
  7. Borenstein M, Hedges LV, Higgins JPT, & Rothstein HR (2009). Introduction to meta-analysis. John Wiley. 10.1002/9780470743386 [DOI] [Google Scholar]
  8. Bylsma LM, Morris BH, & Rottenberg J (2008). A meta-analysis of emotional reactivity in major depressive disorder. Clinical Psychology Review, 28(4), 676–691. 10.1016/j.cpr.2007.10.001 [DOI] [PubMed] [Google Scholar]
  9. Carver CS, Johnson SL, & Joormann J (2008). Serotonergic function, two-mode models of self-regulation, and vulnerability to depression: What depression has in common with impulsive aggression. Psychological Bulletin, 134(6), 912. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Coskunpinar A, Dir AL, & Cyders MA (2013). Multidimensionality in impulsivity and alcohol use: A meta-analysis using the UPPS model of impulsivity. Alcoholism: Clinical & Experimental Research, 37(9), 1441–1450. 10.1111/acer.12131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Cyders MA, & Smith GT (2008). Emotion-based dispositions to rash action: positive and negative urgency. Psychological Bulletin, 134(6), 807–828. 10.1037/a0013341 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ebner-Priemer UW, & Trull TJ (2009). Ecological momentary assessment of mood disorders and mood dysregulation. Psychological Assessment, 21(4), 463–475. 10.1037/a0017075 [DOI] [PubMed] [Google Scholar]
  13. Eisenberg N, Spinrad TL, & Eggum ND (2010). Emotion-related self-regulation and its relation to children’s maladjustment. Annual Review of Clinical Psychology, 6(1), 495–525. 10.1146/annurev.clinpsy.121208.131208 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. 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. Clinical Psychology Review, 28(8), 1413–1425. 10.1016/j.cpr.2008.09.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Foulds J, Veldheer S, Hrabovsky S, Yingst J, Sciamanna C, Chen G, Maccani JZJ, & Berg A (2015). The effect of motivational lung age feedback on short-term quit rates in smokers seeking intensive group treatment: A randomized controlled pilot study. Drug and Alcohol Dependence, 153(1), 271–277. 10.1016/j.drugalcdep.2015.05.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Goh JX, Hall JA, & Rosenthal R (2016). Mini meta-analysis of your own studies: Some arguments on why and a primer on how. Social and Personality Psychology Compass, 10(10), 535–549. 10.1111/spc3.12267 [DOI] [Google Scholar]
  17. Johnson SL, Carver CS, & Joormann J (2013). Impulsive responses to emotion as a transdiagnostic vulnerability to internalizing and externalizing symptoms. Journal of Affective Disorders, 150(3), 872–878. 10.1016/j.jad.2013.05.004 [DOI] [PubMed] [Google Scholar]
  18. King KM, Feil MC, & Halvorson MA (2018). Negative urgency is correlated with the use of reflexive and disengagement emotion regulation strategies. Clinical Psychological Science, 6(6), 822–834. 10.1177/2167702618785619 [DOI] [Google Scholar]
  19. King KM, Karyadi KA, Luk JW, & Patock-Peckham JA (2011). Dispositions to rash action moderate the associations between concurrent drinking, depressive symptoms, and alcohol problems during emerging adulthood. Psychology of Addictive Behaviors, 25(3), 446–454. 10.1037/a0023777 [DOI] [PubMed] [Google Scholar]
  20. Klein DN, Kotov R, & Bufferd SJ (2011). Personality and depression: Explanatory models and review of the evidence. Annual Review of Clinical Psychology, 7(1), 269–295. 10.1146/annurev-clinpsy-032210-104540 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lannoy S, Heeren A, Rochat L, Rossignol M, Van Der Linden M, & Billieux J (2014). Is there an all-embracing construct of emotion reactivity? Adaptation and validation of the emotion reactivity scale among a French-speaking community sample. Comprehensive Psychiatry, 55(8), 1960–1967. 10.1016/j.comppsych.2014.07.023 [DOI] [PubMed] [Google Scholar]
  22. Ledgerwood A (2014). Introduction to the special section on moving toward a cumulative science: Maximizing what our research can tell us. Perspectives on Psychological Science, 9(6), 610–611. 10.1177/1745691614553989 [DOI] [PubMed] [Google Scholar]
  23. March JS, Parker JDA, Sullivan K, Stallings P, & Conners CK (1997). The Multidimensional Anxiety Scale for Children (MASC): Factor structure, reliability, and validity. Journal of the American Academy of Child & Adolescent Psychiatry, 36(4), 554–565. 10.1097/00004583-199704000-00019 [DOI] [PubMed] [Google Scholar]
  24. McLaughlin KA, Kubzansky LD, Dunn EC, Waldinger R, Vaillant G, & Koenen KC (2010). Childhood social environment, emotional reactivity to stress, and mood and anxiety disorders across the life course. Depression & Anxiety, 27(12), 1087–1094. 10.1002/da.20762 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Mischel W (2006). Bridges toward a cumulative psychological science. In Bridging social psychology: Benefits of transdisciplinary approaches (pp. 435–446). Lawrence Erlbaum. [Google Scholar]
  26. Newman MG, Llera SJ, Erickson TM, Przeworski A, & Castonguay LG (2013). Worry and Generalized Anxiety Disorder: A review and theoretical synthesis of evidence on nature, etiology, mechanisms, and treatment. Annual Review of Clinical Psychology, 9(1), 275–297. 10.1146/annurev-clinpsy-050212-185544 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Nock MK, Wedig MM, Holmberg EB, & Hooley JM (2008). The Emotion Reactivity Scale: Development, evaluation, and relation to self-injurious thoughts and behaviors. Behavior Therapy, 39(2), 107–116. 10.1016/j.beth.2007.05.005 [DOI] [PubMed] [Google Scholar]
  28. R Core Team. (2017). R: A language and environment for statistical computing. R Foundation for Statistical Computing. [Google Scholar]
  29. Reeve BB, Hays RD, Bjorner JB, Cook KF, Crane PK, Teresi JA, Thissen D, Revicki DA, Weiss DJ, Hambleton RK, Liu H, Gershon R, Reise SP, Lai J-S, & Cella D (2007). Psychometric evaluation and calibration of health-related quality of life item banks: Plans for the Patient-Reported Outcomes Measurement Information System (PROMIS). Medical Care, 45(5 Suppl. 1), S22–S31. 10.1097/01.mlr.0000250483.85507.04 [DOI] [PubMed] [Google Scholar]
  30. Revelle W (2015). An overview of the psych package. Northwestern University [Google Scholar]
  31. Saylor CF, Finch AJ, Spirito A, & Bennett B (1984). The Children’s Depression Inventory: A systematic evaluation of psychometric properties. Journal of Consulting and Clinical Psychology, 52(6), 955–967. 10.1037/0022-006X.52.6.955 [DOI] [PubMed] [Google Scholar]
  32. Schalet BD, Pilkonis PA, Yu L, Dodds N, Johnston KL, Yount S, Riley W, & Cella D (2016). Clinical validity of PROMIS depression, anxiety, and anger across diverse clinical samples. Journal of Clinical Epidemiology, 73, 119–127. 10.1016/j.jclinepi.2015.08.036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Schmidt FL (1996). Statistical significance testing and cumulative knowledge in psychology: Implications for training of researchers. Psychological Methods, 1(2), 115–129. 10.1037/1082-989X.1.2.115 [DOI] [Google Scholar]
  34. Schwarzer G (2007). meta: An R package for meta-analysis. R News, 7(3), 40–45. [Google Scholar]
  35. Sharma L, Markon KE, & Clark LA (2014). Toward a theory of distinct types of “impulsive” behaviors: A meta-analysis of self-report and behavioral measures. Psychological Bulletin, 140(2), 374–408. 10.1037/a0034418 [DOI] [PubMed] [Google Scholar]
  36. Shrout PE, & Lane SP (2014). Psychometrics. In Mehl MR, & Conner TS(Eds.), Handbook of research methods for studying daily life (pp. 302–320). Guilford Press. [Google Scholar]
  37. Silk JS, Forbes E, & Whalen D (2011). Daily emotional dynamics in depressed youth: A cell-phone ecological momentary assessment study. Journal of Experimental Child Psychology, 110(2), 241–257. 10.1016/j.jecp.2010.10.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Simonsohn U, Nelson LD, & Simmons JP (2014). P-curve: A key to the file-drawer. Journal of Experimental Psychology: General, 143(2), 534–547. 10.1037/a0033242 [DOI] [PubMed] [Google Scholar]
  39. Smith GT, & Cyders MA (2016). Integrating affect and impulsivity: The role of positive and negative urgency in substance use risk. Drug and Alcohol Dependence, 163(Suppl. 1), S3–S12. 10.1016/j.drugalcdep.2015.08.038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. 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(2), 155–170. 10.1177/1073191106295527 [DOI] [PubMed] [Google Scholar]
  41. Smith GT, Guller L, & Zapolski TCB (2013). A comparison of two models of urgency: Urgency predicts both rash action and depression in youth. Clinical Psychological Science, 1(3), 266–275. 10.1177/2167702612470647 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Soto CJ, & John OP (2017). The next Big Five Inventory. Journal of Personality and Social Psychology, 113(1), 117–143. 10.1037/pspp0000096 [DOI] [PubMed] [Google Scholar]
  43. Sperry SH, Lynam DR, & Kwapil TR (2018). The convergence and divergence of impulsivity facets in daily life. Journal of Personality, 86(5), 841–852. 10.1111/jopy.12359 [DOI] [PubMed] [Google Scholar]
  44. Spielberger CD, Sydeman SJ, Owen AE, & Marsh BJ (1999). Measuring anxiety and anger with the State-Trait Anxiety Inventory (STAI) and the State-Trait Anger Expression Inventory (STAXI). In The use of psychological testing for treatment planning and outcomes assessment (2nd ed., pp. 993–1021). Lawrence Erlbaum. [Google Scholar]
  45. Tackett JL, Lahey BB, van Hulle C, Waldman I, Krueger RF, & Rathouz PJ (2013). Common genetic influences on negative emotionality and a general psychopathology factor in childhood and adolescence. Journal of Abnormal Psychology, 122(4), 1142–1153. 10.1037/a0034151 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tsuji S, Bergmann C, & Cristia A (2014). Community-augmented meta-analyses: Toward cumulative data assessment. Perspectives on Psychological Science, 9(6), 661–665. 10.1177/1745691614552498 [DOI] [PubMed] [Google Scholar]
  47. Veroniki AA, Jackson D, Viechtbauer W, Bender R, Bowden J, Knapp G, Kuss O, Higgins JPT, Langan D, & Salanti G (2016). Methods to estimate the between-study variance and its uncertainty in meta-analysis. Research Synthesis Methods, 7(1), 55–79. 10.1002/jrsm.1164 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Walz LC, Nauta MH, & aan het Rot M (2014). Experience sampling and ecological momentary assessment for studying the daily lives of patients with anxiety disorders: A systematic review. Journal of Anxiety Disorders, 28(8), 925–937. 10.1016/j.janxdis.2014.09.022 [DOI] [PubMed] [Google Scholar]
  49. Watson D, Clark LA, & Tellegen A (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54(6), 1063–1070. 10.1037/0022-3514.54.6.1063 [DOI] [PubMed] [Google Scholar]
  50. Westfall J, & Yarkoni T (2016). Statistically controlling for confounding constructs is harder than you think. PLOS ONE, 11(3), 1–22. 10.1371/journal.pone.0152719 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Whiteside SP, & Lynam DR (2001). The five factor model and impulsivity: Using a structural model of personality to understand impulsivity. Personality and Individual Differences, 30(4), 669–689. 10.1016/S0191-8869(00)00064-7 [DOI] [Google Scholar]

RESOURCES