Reactive Risk-Taking: Anxiety Regulation Via Approach Motivation Increases Risk-Taking Behavior

Abstract

Experimental research and real-world events demonstrate a puzzling phenomenon—anxiety, which primarily inspires caution, sometimes precedes bouts of risk-taking. We conducted three studies to test whether this phenomenon is due to the regulation of anxiety via reactive approach motivation (RAM), which leaves people less sensitive to negative outcomes and thus more likely to take risks. In Study 1 (N = 231), an achievement anxiety threat caused increased risk-taking on the Behavioral Analogue Risk Task (BART) among trait approach-motivated participants. Using electroencephalogram in Study 2 (N = 97), an economic anxiety threat increased behavioral inhibition system-specific theta activity, a neural correlate of anxiety, which was associated with an increase in risk-taking on the BART among trait approach-motivated participants. In a preregistered Study 3 (N = 432), we replicated the findings of Study 1. These results offer preliminary support for the reactive risk-taking hypothesis.

Keywords

risk-taking anxiety approach motivation emotion regulation

American economist and Nobel Laureate (2013) Robert Shiller titled his 2000 book Irrational Exuberance to describe the then-booming dot-com market as an economic bubble bloated by high-risk investing. Five years later, Shiller (2005) declared that the U.S. housing market was also a bubble fueled by the same exuberant risk-taking. Although decidedly against popular opinion at the time, by recognizing the role of exuberant risk-taking, Shiller is now seen as an “oracle” for predicting the recessions that took place in the early and late 2000s. Exuberant risk-taking remains an important yet puzzling phenomenon. What leads people to exuberant risk-taking? For example, why do some people engage in dangerous sensation-seeking, risky driving, reckless gambling, and unsafe sexual practices?

Based on the Reactive Approach Motivation (RAM) model (McGregor et al., 2010a; see also Jonas et al., 2014), the present research examines a fundamental motivational process that may underlie certain forms of risky exuberance. While some forms of exuberant risk-taking may result from a dangerous spiral of positive feedback, experimental and qualitative research reveals that other instances can, paradoxically, follow anxious experiences. Although anxiety initially makes people more cautious (Gray & McNaughton, 2000), past research has demonstrated that anxious experiences cause some people to become approach motivated in an effort to alleviate the aversive state (McGregor et al., 2010a; Nash et al., 2011). Although approach motivation may regulate anxiety, it may also make people less sensitive to negative outcomes (Nash et al., 2012) and, counterintuitively, more open to taking risks than they would normally be (Hirsh et al., 2012).

Anxiety and RAM

Pharmacological, neuropsychological, and behavioral studies on humans and animals demonstrate a basic goal-regulation process initiated by experiences of anxiety (Gray & McNaughton, 2000). Anxiety is an uncomfortable affective state of heightened arousal and vigilance that arises from the uncertainty of goal conflict. Goal conflict can emerge when a focal goal is threatened, leading to a discrepancy between expectation and reality (e.g., when financial goals are impeded by looming economic uncertainty). The behavioral inhibition system (BIS) responds to this motivational uncertainty with anxiety by inhibiting the conflicted goal and redirecting behavior toward alternative, less conflicted goals (Gray & McNaughton, 2000; McGregor et al., 2010a). When an alternative goal is located, anxiety is muted by the pursuit of the new goal. This goal pursuit is governed by the behavioral approach/activation system (BAS; Gray & McNaughton, 2000). The BIS and the BAS tend to act in opposition (Corr, 2008). The BIS promotes increased sensitivity to threats and negative outcomes, whereas the BAS and approach motivation are thought to mute this sensitivity in the service of efficient and focused goal pursuit (Corr, 2002). Consistent with this, BAS-related neural activity predicts reduced sensitivity to negative stimuli (E. Harmon-Jones et al., 2009; Nash et al., 2012).

In sum, people may initiate approach-motivated states to resolve BIS-elicited anxiety (Jonas et al., 2014). Importantly, anxiety-inducing experiences cause RAM on self-report (e.g., personal goal striving and ideological conviction; McGregor et al., 2007, 2008; Nash et al., 2011), implicit (e.g., approach motivation implicit association test; McGregor et al., 2010a), behavioral (e.g., line bisection task; Nash et al., 2010), affective (e.g., emotional conviction; Nash et al., 2011), and electroencephalographic (EEG; e.g., heightened Relative Left Frontal EEG Activity; McGregor et al., 2009, 2010a) measures. Prior research reveals that individuals with approach-motivated personalities are particularly prone to RAM (McGregor et al., 2010a). For example, approach motivation-related dispositions like BAS sensitivity (Carver & White, 1994) and action control (Kuhl, 1994) predict RAM responses to anxious uncertainty (McGregor et al., 2010b).

Thus, although RAM can be an effective antidote to BIS-activation and anxiety, particularly among approach-motivated people, it may also leave people insensitive to negative outcomes and less restrained than they normally would be. Critically, approach-motivated eagerness may affect ensuing decisions and behaviors, particularly those related to risk. Yet, to our knowledge, no studies have examined RAM and risk-taking.

Risk-Taking Following Anxiety

The claim that anxiety can lead to risk-taking may appear inconsistent with past research. Trait and state anxiety have been associated with increased sensitivity to threatening cues and heightened activity in the amygdala—a brain structure important for threat detection and negative arousal (Bishop et al., 2004; Derryberry & Reed, 2002; Gray & McNaughton, 2000). Similarly, trait and state anxiety predict higher estimates of risk (Gasper & Clore, 1998; Lerner & Keltner, 2000). However, the notion that anxiety can sometimes lead to increased risk-taking does have empirical precedent. For example, professional stock traders who incur morning losses trade more aggressively in the afternoon (Garvey et al., 2007). Similarly, poker players become riskier after large and frustrating losses, despite knowing that riskier strategies generally lead to less success (Smith et al., 2009). This phenomenon also emerges in society at large. For example, during times of economic uncertainty, lottery ticket sales increase (Kumar, 2009). Furthermore, usage of slot machines in Christchurch bars increased by 19% following the devastating 2009 earthquake, although a sizable proportion of the slot machines were destroyed (Community and Public Health, 2012).

Anxiety-inspired risk-taking is not just bound to financial decisions. For example, under specific conditions, social anxiety predicts risk proneness in domains, such as drug use, heavy drinking, and unsafe sexual activity (Kashdan et al., 2006). High levels of anxiety also predict risky driving behavior, crashes, and driving under the influence (DUI) episodes (Dula et al., 2010). In sum, although anxious experiences initially inspire caution, they may also precede bouts of increased risk-taking for some people. We propose that these people may be regulating anxiety by becoming approach-motivated, leading to insensitivity to negative outcomes and increased proneness to risk.

RAM and Risk-Taking

Risk-taking is characterized by a reduced sensitivity to negative outcomes. Risk takers tend to undervalue negative outcomes, overvalue rewards, appraise risk as lower, and anticipate lower levels of anxiety during risk (Kahneman & Tversky, 1979; Zuckerman & Kuhlman, 2000). Risk-taking is also associated with reduced neural reactivity to negative outcomes (Santesso & Segalowitz, 2009). Disrupting function in the right prefrontal cortex, a brain area associated with the BIS and sensitivity to negative outcomes (Coan & Allen, 2004; Shackman et al., 2009), causes increased risk-taking (Knoch et al., 2006). Furthermore, there is compelling evidence that risk-taking is an approach-motivated phenomenon. Risk-taking and approach motivation are associated with the same dispositions, emotions, and neurobiological processes, such as sensation-seeking, impulsivity, dispositional BAS, positive affect, anger, testosterone, dopaminergic processes, and reward-related brain areas and approach-related patterns of resting brain activation (Corr, 2002; Gianotti et al., 2009; Kim & Lee, 2011; Lerner & Keltner, 2000; Loewenstein et al., 2001; Mohr et al., 2010). In sum, risk-taking involves approach motivation and reduced sensitivity to negative outcomes. As such, although anxiety is intuitively associated with caution, some people may become more approach-motivated to downregulate anxiety, leaving them prone to risky behavior.

Overview of Studies

We expected the distinct experience of anxiety to trigger approach motivation processes that would subsequently bolster risk-taking behavior among certain people. In contrast to past work that has focused on how risk-taking relates to trait anxiety, fear, or undifferentiated negative affect (Gasper & Clore, 1998; Lerner & Keltner, 2000; Maner & Gerend, 2007), we hypothesize that risk-taking can arise from reactions to state anxiety (i.e., anxious experiences) because this state elicits and is regulated by RAM.

Given that trait approach motivation is associated with RAM (Jonas et al., 2014; McGregor et al., 2010b) and risk-taking (Kim & Lee, 2011), we expected individuals prone to approach-motivated behavior to engage in more reactive risk-taking. Here, we measured trait approach motivation using the BAS scale of the widely used BISBAS scale (Carver & White, 1994). The BAS scale is characterized by increased sensitivity to rewards, heightened drive for achievement, and fun-seeking (Carver & White, 1994).

In Study 1, we test whether an anxiety manipulation causes increased risk-taking behavior, particularly among trait approach-motivated participants (reactive risk-taking effect). In Study 2, we use EEG to test whether an established neural correlate of anxiety—BIS-specific theta—mediates the relationship between anxiety manipulation and risk-taking behavior among trait approach-motivated participants. Finally, in a preregistered Study 3, we attempt to replicate the reactive risk-taking effect under conditions that would seemingly favor finding the opposite effect. All studies reported in this article received ethical approval (Protocols 00079790, 00087564, and 00084513) from the University of Alberta Research Ethics Board. All relevant data are available at https://osf.io/r3sv4/.

Study 1: Achievement Anxiety Causes Risk-Taking

In this study, we used an achievement anxiety manipulation that has reliably caused self-reported anxiety and RAM in past research (e.g., McGregor et al., 2008, 2009, 2010b; Nash et al., 2011). We preregistered the hypothesis that anxiety manipulation would cause an overall increase in risk-taking behavior. Subsequently, we report this analysis as well as a post hoc exploratory analysis relating to the moderating role of trait approach motivation, which informed the rationale for Studies 2 and 3.

Method

Open science

Our experimental design, a priori hypotheses, and confirmatory analysis plan were preregistered at the Open Science Framework, and the preregistration is available for download at: https://osf.io/9ezfp. All relevant materials are available for download at https://osf.io/msxuv/. As part of a separate line of research, we also preregistered two moderation effects concerning attachment and self-control that will be reported elsewhere.

Participants and design

Assuming a small to medium effect size based on reports from related studies (e.g., McGregor et al., 2010a; Schumann et al., 2014), we conducted an a priori power analysis using R (R Core Team, 2019) with p = .05, ΔR² = .035, and power = .80 (Aberson, 2019). This analysis produced a recommended target sample of 222 participants. A total of 253 undergraduate students from a Canadian University participated for class credit. Data from 235 participants (M_age = 19.15, SD_age = 2.12; 153 women, 82 men) were analyzed after exclusions (compliance: n = 14; study familiarity: n = 4). The study used a between-subjects design with random assignment into two experimental conditions (anxiety vs. control). Trait approach motivation was included as a moderator variable and the anxiety (vs. control) manipulation as a between-subjects factor (independent variable, coded 0 = control; 1 = anxiety). Risk-taking behavior served as the dependent variable.

Procedure

Participants were instructed to sit at an individual computer in a research lab. Responses were collected using Qualtrics and Inquisit software. After providing consent, participants completed a questionnaire that assessed demographic information and a trait approach motivation measure within a short battery of personality questionnaires (all data available upon request).¹ Next, participants were randomly assigned to an achievement anxiety condition or a control condition. Following the manipulation, all participants completed the BART before completing manipulation and compliance checks and a thorough debriefing.

Trait approach motivation

We measure trait approach motivation using the BAS scale of the widely used BISBAS measure (Carver & White, 1994). Participants rated the extent to which different statements generally apply to them (from Strongly Disagree = 1 to Strongly Agree = 5). Sample items include “If I see a chance to get something I want, I move on it right away” and “When I see an opportunity for something I like, I get excited right away.” The BAS scale exhibits strong psychometric reliability (Cronbach’s α = .76) and validity (Jorm et al., 1998).

Achievement anxiety (vs. control) manipulation

Participants were instructed to read a series of passages ostensibly from a popular psychology statistics textbook and told that we researchers were interested in how understandable the passages were. In the achievement anxiety condition, participants were presented with passages taken from an advanced graduate statistics textbook, dense with intimidating formulae and symbols. Key phrases and symbols were removed to make these passages even more unintelligible (Appendix A). In the control condition, participants were presented with more understandable passages about the benefits of statistics (Appendix B). Participants were then asked to report how well they understood each passage. This achievement anxiety manipulation (compared with the control) has produced self-reported anxiety and RAM in more than 10 separate published studies (e.g., McGregor et al., 2009; Nash et al., 2010).

Risk-taking dependent variable

Risk-taking behavior was measured using the Balloon Analogue Risk Task (BART; Lejuez et al., 2002). Participants were instructed to pump up virtual balloons on a computer screen by pressing the space bar. Each pump slightly inflates the balloon and earns the participant raffle entries toward a CAD$100 cash prize. However, each pump brings the balloon closer to a randomly determined threshold where the balloon will explode, and the participant will lose all the raffle entries earned on that trial. Importantly, participants can “collect” at any point prior to the balloon exploding, and the raffle entries earned from that trial will be added to their total (e.g., collect after 25 pumps, 25 entries are added to the total). Each participant completed 30 trials (i.e., 30 different balloons). In sum, each pump increases the potential risk and reward. Therefore, the average number of pumps is a measure of a participant’s risk-taking behavior. The BART shows strong reliability and construct and ecological validity as a measure of behavioral risk-taking in numerous psychological and behavioral economic studies (Lejuez et al., 2002). Furthermore, there are several conflicting studies examining anxiety and the BART (e.g., Hunt et al., 2005; Lejuez et al., 2002; Zhang & Gu, 2018), and we reasoned that our program of research, based on neurobiologically grounded conceptualizations and operationalizations of anxious states, could perhaps inform these conflicting findings.

Manipulation check

Participants were asked to retrospectively rate how the statistics comprehension task made them feel the following adjectives: good, happy, smart, successful, likable, meaningful, frustrated, confused, uncertain, empty, anxious, insecure, lonely, ashamed, and stupid (McGregor et al., 2010a). As a self-reported measure of anxiety, we created a Felt-anxiety composite (Cronbach’s α = .81) from all anxiety-related adjectives, including only anxious, confused, uncertain, frustrated, and insecure.

Results

Manipulation check

Felt-anxiety composite scores were calculated and standardized (z-score). A one-way analysis of variance (ANOVA) demonstrated that participants in the achievement anxiety condition reported higher Felt-anxiety-composite scores (M = 0.38, SD = 1.03) than participants in the control condition (M = −0.38, SD = 0.81), F(1, 224) = 38.35, p < .001, η²p = .15. Thus, the achievement anxiety task caused greater self-reported anxiety in our sample, compared with the control task.

Main effect of achievement anxiety on risk-taking

We first examined assumptions and checked for outliers. No outliers were identified. The dependent measure, adjusted mean pumps on the BART, was computed by taking the mean of all nonexploded balloons and then standardized (z-score). As in past research, we used the adjusted mean (i.e., the mean number of pumps on collected balloons only)² instead of an absolute mean because the number of pumps was necessarily constrained on exploded balloons, which limits the between-subject variability (see Lejuez et al., 2002). Thus, the standardized adjusted mean pumps variable (i.e., risk-taking variable) was entered into a one-way ANOVA to test whether the achievement anxiety manipulation caused increased risk-taking. As predicted, those in the achievement anxiety condition demonstrated increased risk-taking (M = 0.15, SD = 1.01) compared with those in the control condition (M = −0.15, SD = 0.97), F(1, 233) = 5.54, p = .019, η²p = .02.

Trait approach motivation moderation analysis

Participant’s trait approach motivation scores were calculated and then mean-centered (Aiken et al., 1991). We conducted a hierarchical linear regression analysis to determine whether trait approach motivation moderated the effect of achievement anxiety on risk-taking. The regression model was significant at Step 1, R² = .04, F(2, 232) = 4.77, p = .009. Achievement anxiety, B = 0.31, SE = .13, t(232) = 2.44, p = .016, 95% confidence interval (CI) = [0.06, 0.57], and trait approach motivation, B = 0.29, SE = .15, t(232) = 1.98, p = .049, 95% CI = [0.002, 0.58], were significant predictors of risk-taking. In support of our hypothesis, there was a significant trait approach motivation by achievement anxiety interaction at Step 2, R² = .06, ΔR² = .02, ΔF(1, 231) = 5.75, p = .017, B = 0.70, SE = .29, 95% CI = [0.12, 1.27].

A simple effects analysis demonstrated that among participants high in trait approach motivation (+1 SD), achievement anxiety increased risk-taking behavior compared with the control condition, B = 0.60, SE = .17, t(231) = 3.43, p < .001, 95% CI = [0.25, 0.94]. This finding remains significant after adjusting for multiple tests using the Bonferroni adjustment method (.05/4 = .0125). Among participants low in trait approach motivation (−1 SD), achievement anxiety had no effect on risk-taking behavior, B = 0.0076, SE = .18, t(231) = 0.04, p = .97, 95% CI = [−0.35, 0.36]. Note that these results remain significant after correcting for four exploratory analyses (Bonferroni Correction; .05/4 = .0125). These results support the hypothesis that trait approach-motivated participants would be the most prone to risk-taking behavior following an anxious experience (Figure 1).

Figure 1.

Study 1: Standardized mean adjusted pumps on the BART (i.e., risk-taking) as a function of trait approach motivation and condition (anxiety vs. control). Error bars represent standard errors of the mean.

Study 2: BIS-Specific Theta Mediates the Effect of Anxiety on Risk-TakingStudy 1 demonstrated that trait approach-motivated people respond to an achievement anxiety manipulation with increased risk-taking. Study 2 looked to extend this finding using a neurophysiological approach. Specifically, we used EEG to measure BIS-specific (i.e., medial right frontal) theta power (McNaughton et al., 2013) during an anxiety manipulation. Recall that BIS activation is integral to the expression of anxiety (Gray & McNaughton, 2000). A key property of neural activity in the BIS is low-frequency (“theta”) rhythmicity (McNaughton et al., 2007). Meta-analytical evidence suggests that this theta rhythmicity represents a neurophysiological mechanism of behavior adjustments (e.g., the BIS response) to anxiety and uncertainty (Cavanagh & Shackman, 2015). Indeed, BIS-specific theta has also been linked to trait anxiety (Neo & McNaughton, 2011), and anxiolytic drugs reduce this rhythmicity (McNaughton et al., 2013). By assessing the mediating role of BIS-specific theta, Study 2 aims to provide evidence that reactive risk-taking is a response to anxiety. Thus, in Study 2, we expected an anxious experience to increase BIS-specific theta, which would, in turn, lead to increased risk-taking behavior among trait approach-motivated participants.

Method

Open science

All relevant materials are available for download at https://osf.io/hjxe5/.

Participants and design

We conducted our power analysis based on pilot data, which allowed us to determine that the economic anxiety manipulation was indeed anxiety provoking in general, compared with an equivalent control, F(1, 57) = 66.449, p < .0001, η²p = .538 (large effect; Leota & Nash, 2021). Because the current study differed from the pilot, we used a much more conservative estimate of effect size using R (R Core Team, 2019) with p = .05, d = .58, and power = .80 (Aberson, 2019). This analysis produced a recommended target sample of 98 participants. Thus, 106 undergraduate students with corrected-to-normal vision were recruited from a first-year psychology class and were compensated with class credit. Data from 97 participants (M_age = 19.77, SD_age = 1.53, 56 women, 41 men, right-handed = 86, left-handed = 7, ambidextrous = 3, and 1 participant did not report) were analyzed after exclusions for poor EEG connection (n = 4), incomplete data (n = 2), and failing to reconsent (n = 3). The study used a between-subjects design with random assignment into two experimental conditions (economic anxiety vs. control). Trait approach motivation was included as a moderator variable and an economic anxiety (vs. control) manipulation as a between-subjects factor. BIS-specific theta power measured during the manipulation was included as a mediating variable, and risk-taking behavior served as the dependent variable.

Procedure

Participants first completed a written informed consent, then were fitted with a 64-channel EEG headset (Brain Products) and seated at a computer station in an electrically- and sound-shielded room. All materials were completed on a computer. Participants reported demographic information, including age, gender, ethnicity, completed the measure of trait approach motivation from Study 1 (BAS scale; Cronbach’s α = .71; Carver & White, 1994), before completing several other personality questionnaires (see https://osf.io/hjxe5/ for a list of all questionnaires) as part of a larger research project on individual differences in the neuroscience of self-regulation (all data available upon request). Participants were then randomly assigned to either the economic anxiety condition or the control condition. Participants then completed a passive auditory oddball task and a color-naming Stroop task before completing the BART (these measures of more basic cognitive processes are not considered here, manuscripts in prep). Then, participants answered manipulation and compliance checks and a thorough debriefing, had the headset removed and hair washed, and were thanked for their time.

Economic anxiety (vs. control) manipulation

Participants were tasked to generate a headline for an ostensibly real news article that had recently appeared on the Canadian news website, CBC.ca. In the economic anxiety condition, participants read an article that detailed a bleak and unsettling economic forecast for young Canadians (Appendix C). The forecast was said to be endorsed by leading Canadian researchers who concluded that a recession was imminent due to apparent economic red flags, such as wage stagnation, rising student debt, and declining long-term employment opportunities. Importantly, the article suggested that current Canadian students would be hit the hardest. As such, the article was tailored to elicit economic angst among our sample. Participants in the control condition were presented with a CBC.ca news article that detailed a more neutral economic picture for young Canadians (Appendix D). This forecast, also ostensibly endorsed by leading Canadian researchers, emphasized economic stability and adequate employment opportunities, particularly among current Canadian students. Importantly, both articles were derived from real economic forecasts recently published in the media. All participants were given 3 min to read the article and were later asked to submit their headline in a textbox before completing the BART.

EEG recording and preprocessing

Continuous EEG was recorded using the 64 Ag-AgCl channel ActiCHamp EEG system (Brain Products), positioned according to the 10/10 system and digitized at a sampling rate of 512 Hz (24-bit precision; bandwidth: 0.1–100 Hz). We used V16 SuperVisc High-Viscosity Gel for Active Electrodes and aimed to keep impedances <10 kOhms. During recording, signals were referenced online to the TP9 electrode positioned over the left mastoid. Off-line, EEG was re-referenced to the average mastoids (TP9-TP10), downsampled to 256 Hz, band-pass filtered between 0.1 and 30 Hz, and notch filtered at 60 Hz. Recorded EEG was segmented into the duration of reading for each participant of the economic anxiety manipulation. Blinks were statistically removed using the automatic ocular correction developed by Gratton and colleagues (1983, HEOG reference = FP2; VEOG reference electrode = FP1). Artifacts were then automatically detected using the following parameters: −100 to +100 μV min/max threshold, 50 μV maximum voltage step, 0.5 μV lowest allowed voltage (maximum–minimum) in 100-ms intervals.

BIS-specific theta

Contiguous artifact-free epochs of 2 s were extracted through a hamming window and overlapped by 75% to avoid data loss. Power spectra were calculated via fast Fourier transform, and power values (in μV²) were averaged over all artifact-free and blink-free epochs. Absolute power was averaged for the theta frequency band (3.5–7.5 Hz) at all nodes. The medial right frontal cortex is implicated in behavioral inhibition caused by goal conflict (Neo et al., 2011; Shackman et al., 2009). BIS-specific theta was thus calculated as theta power at the medial right frontal site, node F4 (Neo et al., 2011; see also McNaughton et al., 2013).

Risk-taking dependent variable

Due to time restrictions, risk-taking behavior was measured using a brief version of the BART with a lower maximum pumps per balloon threshold (30) and 20 balloon trials instead of 30. Much like the standard BART, the brief BART exhibited high reliability (Cronbach’s α = .89). The incentive structure was identical to Study 1, with pumps earning raffle entries toward a CAD$100 cash prize.

Manipulation check

The same Felt-anxiety composite (Cronbach’s α = .88) as Study 1 was created as a self-reported measure of anxiety from all anxiety-related adjectives, including only anxious, confused, uncertain, frustrated, and insecure.

Results

Manipulation check

Felt-anxiety composite scores were calculated and standardized (z-score). A one-way ANOVA demonstrated that participants in the economic anxiety condition reported higher Felt-anxiety composite scores (M = 0.49, SD = 0.77) than participants in the control condition (M = −0.72, SD = 0.86), F(1, 95) = 52.22, p < .001, η²p = .36. Thus, the economic anxiety task caused greater self-reported anxiety in our sample, compared with the control task.

Moderated mediation analyses

Participant’s trait approach motivation scores were calculated and then mean-centered (Aiken & West, 1991). The dependent measure, adjusted mean pumps on the BART, was computed by taking the mean of all nonexploded balloons. These scores were then standardized (z-score). We first examined assumptions and checked for outliers. No outliers were identified. We tested whether BIS-specific theta measured during the economic anxiety (vs. control) manipulation mediated the effect of economic anxiety on risk-taking behavior for trait approach-motivated participants. There was a strong relationship between gender (male = 1; female = 2) and BIS-specific theta, r = .26, p = .012. Thus, gender was included as a covariate in the model.³ The below analyses without the gender covariate are reported in Supplemental Material 1. We conducted mediational analyses using the PROCESS macro (Model 7; see A. F. Hayes, 2017; note results from Model 8 are reported in Supplemental Material 1) with 5,000 bootstrapped resamples. First, there was a marginal interaction effect between trait approach motivation and economic anxiety on BIS-specific theta, B = 0.40, SE = .23, t(93) = 1.77, p = .08, 95% CI = [−0.05, 0.85]. Importantly, a simple effects analysis revealed that among participants high in trait approach motivation (+1SD), economic anxiety significantly increased BIS-specific theta compared with the control condition, B = 0.29, SE = .13, t(91) = 2.28, p = .025, 95% CI = [0.04, 0.55]. Among participants low in trait approach motivation (−1SD), economic anxiety had no significant effect on BIS-specific theta, B = −0.03, SE = .13, t(91) = −0.25, p = .80, 95% CI = [−0.30, 0.23].

Second, these analyses revealed a significant moderated mediation effect on risk-taking behavior, Index = 0.18, bootstrapped SE = 0.12, and bootstrapped 95%CI = [0.001, 0.47] (see Figure 2). Importantly, the indirect effect of economic anxiety on risk-taking via BIS-specific theta was significant among those high in trait approach motivation (+1SD), B = .13, bootstrapped SE = .08, bootstrapped 95% CI [0.001, 0.33]. There was no significant indirect effect for those low in trait approach motivation (−1SD), B = −0.02, bootstrapped SE = .05, bootstrapped 95% CI [−0.12, 0.07]. These effects suggest that, among trait approach-motivated participants, the specific increase in risk-taking caused by the economic anxiety was mediated by the increase in BIS-specific theta.

Figure 2.

Study 2: Moderated mediation model (Model 7, 5,000 bootstrapped resamples; A. F. Hayes, 2017) and regression coefficients for the relationship between economic anxiety and risk-taking as mediated by BIS-specific theta (F4) and moderated by trait approach motivation. The regression coefficient for the effect of economic anxiety on BIS-specific theta at high trait approach motivation (+1SD) is in brackets. The regression coefficient for the direct effect of economic anxiety on risk-taking, controlling for BIS-specific theta, is in parentheses.

Study 3: Preregistered Replication Study

Study 1 demonstrates that an anxious experience leads to increased risk-taking behavior among trait approach-motivated participants. Study 2 further reinforced this phenomenon by demonstrating that BIS-specific theta mediated the effect of an anxious experience on risk-taking. As these effects were not preregistered, we follow open science recommendations and attempt to replicate the reactive risk-taking effect in a preregistered, high-powered Study 3. We also attempt to reconcile the reactive risk-taking hypothesis with past research on anxiety and risk aversion. Intuition suggests that anxiety, a state of heightened vigilance, should promote risk aversion. Past research has focused on the relationship between trait anxiety and risk aversion (e.g., Gambetti & Giusberti, 2012; Maner et al., 2007); however, empirical evidence is less clear than intuition may suggest. For example, past studies have failed to find a consistent relationship between trait anxiety and risk aversion (e.g., Gu et al., 2010; Hockey et al., 2000; Mitte, 2007), and some studies have even found a negative relationship (e.g., Lauriola et al., 2005; Nicholson et al., 2005). And, as we highlight in the Introduction, empirical evidence relating to the relationship between state anxiety and risk aversion is similarly equivocal. These contradictory findings may be explained by a person × situation interaction such that trait anxious individuals become risk-averse only after they experience state anxiety (Endler & Kocovski, 2001). Among trait anxious individuals, risk-averse decision-making processes may be shaped in part by anxiety experienced in the moment (Loewenstein et al., 2001; Maner et al., 2007). From a motivational perspective, cautious restraint (i.e., the avoidance of risk) may offer respite from anxious experiences for some (e.g., J. Hayes et al., 2016), much like eager risk-taking does for others.

To test the intuitive effect that trait anxiety, state anxiety, or their interaction is associated with risk aversion, we preregistered a conceptual replication of Studies 1 and 2, with two key changes. First, we measured trait anxiety prior to the experimental manipulation. Second, we altered the framing of the BART to promote more risk-taking behavior generally. People generally behave cautiously on the BART (i.e., most people fall below the optimal [and mid-point] number of pumps; Lejuez et al., 2002). This tendency toward risk-aversion may create a floor effect that would make it difficult to uncover significant deviations at the lower end of the continuum. Accordingly, we changed the framing of the BART in Study 3 from gains to losses. People tend to seek risks when a decision is presented in a negative frame (Tversky & Kahneman, 1981). Therefore, instead of pumping up a balloon to gain raffle tickets (Studies 1 and 2), we constructed the BART so that participants pumped up balloons to avoid losing raffle tickets (see Methods subsequently). We speculated that this would promote risk-taking more generally, allowing for easier detection of deviations toward risk-aversion. In doing so, we also create a robust test of the reactive risk-taking hypothesis given that changing the frame of the BART may in turn create a ceiling effect, making it more difficult to detect deviations toward greater risk-taking behavior.

We preregistered two hypotheses. First, we hypothesized that participants high in trait neuroticism would respond to an anxious experience with increased risk-aversion on the BART (the intuitive hypothesis). Second, we hypothesized that participants high in trait approach motivation would respond to an anxious experience with increased risk-taking on the BART (the reactive risk-taking hypothesis). We also preregistered two exploratory hypotheses relating to our conceptualization of trait anxiety: trait BIS and trait uncertainty aversion. In other words, we also analyzed these two anxiety-related traits as moderators in the intuitive hypothesis.

Method

Open science

Participants and design

Based on findings from Study 1, we conducted an a priori power analysis using R (R Core Team, 2019) with p = .05, ΔR² = .02, and power = .80 (Aberson, 2019). This analysis produced a recommended target sample of 380 participants. We recruited 499 people from Amazon’s Mechanical Turk (using the CloudResearch MTurk Toolkit; Litman et al., 2017) to participate in a “Psychology survey” for US$1.50. Because this study was conducted online, we screened participants to ensure high-quality data prior to any analyses. Participants were excluded if they failed a compliance check (n = 36; e.g., failing to “Click on the ‘Strongly Agree’ option”), self-reported poor-quality data (n = 14), failed to reconsent (n = 2), or reported that they knew the article was fabricated (n = 15; precise data-screening criteria are reported in Supplemental Material 3). Thus, data from 432 participants (M_age = 39.55, SD_age = 12.36, 207 women, 219 men, 2 other, 4 NA) were analyzed. The study used a between-subjects design with random assignment into two experimental conditions (anxiety vs. control). Trait neuroticism and trait approach motivation were included as moderator variables in separate analyses. Risk-taking served as the dependent variable.

Procedure

After providing consent, participants completed a questionnaire that assessed demographic information, including age, gender, and ethnicity, the measure of trait approach motivation from Studies 1 and 2 (BAS scale; Cronbach’s α = .77; Carver & White, 1994), a trait neuroticism measure (a primary measure of trait anxiety) within a short battery of personality questionnaires (including trait uncertainty aversion and trait BIS sensitivity). Next, participants were randomly assigned to an economic anxiety condition or a control condition. Following the manipulation, all participants completed the BART before completing manipulation and compliance checks and a thorough debriefing.

Trait neuroticism

We measured trait neuroticism using the Emotional Stability scale (reverse-coded) from the 10-item Big-Five personality inventory (Gosling et al., 2003). Participants rated the extent to which two statements generally applied to them (from Strongly Disagree [1] to Strongly Agree [5]). The two statements were “I see myself as: Anxious, easily upset.” and “I see myself as: Calm, emotionally stable” (reverse-coded; Cronbach’s α = .83). The scale exhibits adequate psychometric reliability and validity (Gosling et al., 2003).

Economic anxiety (vs. control) manipulation

The economic anxiety task from Study 2 was altered so that it was relevant to the American sample. In brief, participants were tasked to generate a headline for an ostensibly real news article that had recently appeared on the Wall Street Journal website. In the economic anxiety condition, participants read an article that detailed a bleak and unsettling economic forecast for Americans (Appendix E). In the control condition, participants read an article that detailed a more neutral economic picture for Americans (Appendix F).

Risk-taking dependent variable

Risk-taking behavior was measured using an adapted version of the BART (Lejuez et al., 2002). At the beginning of the BART, each participant was given 60 tickets and told that any remaining tickets could be entered into a draw to win a US$100 bonus. Before each round began, 3 tickets are temporarily subtracted from the participant’s total. However, the bigger the participant blew up the balloon, the fewer tickets they would potentially lose (each pump saves 0.1 tickets). At any time during each round, the participant could choose to click the button “accept current losses.” Doing so stopped the round and subtracted the current number of tickets from their total. For example, if a participant clicked “accept current losses” after 12 pumps, 1.8 tickets would be subtracted from their total (i.e., 3 tickets − [12 pumps × 0.1 tickets]). However, if the balloon exploded, the participant would lose all 3 tickets for that round. Like in Study 2, the balloon was set to pop after a random number of pumps between 1 and 30; thus, the explosion point randomly varied around 15 pumps. In essence, this version of the BART mirrors the version used in Study 2; however, it is played within the domain of losses and not the domain of gains.

Manipulation check

We administered the Discrete Emotions Questionnaire (DEQ; C. Harmon-Jones et al., 2016) to test whether the economic anxiety task directly raised anxiety. Participants rated the extent to which they experienced 32 different emotions while reading the Wall Street Journal article (from Not at all = 1 to An extreme amount = 7). The DEQ is sensitive to eight distinct state emotions: anxiety, anger, fear, disgust, sadness, happiness, relaxation, and desire. The anxiety scale includes items dread, anxiety, nervous, and worry (Cronbach’s α = .93). The DEQ exhibits strong psychometric reliability and validity (C. Harmon-Jones et al., 2016).

Results

BART in the domain of losses. We examined the raw grand mean of the adjusted mean pumps on the BART and compared with Study 2 to test whether changing the frame to the domain of losses increased risk-taking behavior. Overall, participants averaged 11.64 (SD = 4.42) adjusted mean pumps on the BART in Study 3, a significant increase from Study 2 (M = 9.22, SD = 3.63), F(1, 531) = 26.11, p < .001, η²p = .05. Thus, framing trials within the BART as a loss appears to have caused a general risk-taking increase.

Manipulation check

Discrete emotion scores from the DEQ were calculated and standardized (z-score). A one-way ANOVA demonstrated that participants in the economic anxiety condition reported higher discrete anxiety scores (M = 0.53, SD = 0.61) than participants in the control condition (M = −0.52, SD = 1.03), F(1, 430) = 168.21, p < .001, η²p = .28. Importantly, discrete anxiety accounted for the largest mean difference (MD = 1.06, SE = .08) and largest effect size out of all negative emotions (anxiety, anger, fear, disgust, sadness). The effect of economic anxiety on self-reported discrete anxiety remains significant after controlling for self-reported discrete anger, fear, disgust, and sadness, F(1, 426) = 27.37, p < .001, η²p = .06. Furthermore, the effects of economic anxiety on self-reported discrete anger, fear, disgust, and sadness all become nonsignificant after controlling for self-reported discrete anxiety, all ps > .06. Thus, the economic anxiety task caused greater self-reported discrete anxiety specifically in our sample, compared with the control task.

Trait neuroticism moderation analysis

Participant’s trait neuroticism scores were calculated and then mean-centered (Aiken & West, 1991). We conducted a hierarchical linear regression analysis to determine whether trait neuroticism moderated the effect of economic anxiety on risk-taking. The regression model was nonsignificant at Step 1, R² = .0001, F(2, 425) = 0.02, p = .98. Neither economic (state) anxiety, B = 0.04, SE = .069, t(425) = 0.062, p = .951, 95% CI = [−0.13, 0.14], nor trait neuroticism, B = −0.07, SE = .04, t(425) = −0.13, p = .59, 95% CI = [−0.09, 0.08], were significant predictors of risk-taking. The trait neuroticism by economic anxiety interaction at Step 2 was also nonsignificant, R² = .002, ΔR² = .002, ΔF(3, 424) = .292, p = .81, B = −0.08, SE = .08, 95% CI = [−0.24, 0.09]. These results fail to support the hypothesis that trait neuroticism individuals would be the most prone to risk-averse behavior following an anxious experience. Trait uncertainty aversion, trait BIS sensitivity, and their interaction terms with economic anxiety were also all unrelated to risk-averse behavior (all ps > .39, full analyses reported in Supplemental Material 4). Therefore, the nonsignificant effects cannot be attributed to the choice of the neuroticism scale.

Trait approach motivation moderation analysis

Participant’s trait approach motivation scores were calculated and then mean-centered (Aiken & West, 1991). As in Study 1, we conducted a hierarchical linear regression analysis to determine whether trait approach motivation moderated the effect of economic anxiety on risk-taking. The regression model was nonsignificant at Step 1, R² = .001, F(2, 426) = 0.28, p = .76. Neither economic anxiety, B = −0.01, SE = .10, t(426) = −0.12, p = .91, 95% CI = [−0.13, 0.14], nor trait approach motivation, B = −0.6, SE = .08, t(426) = −0.74, p = .46, 95% CI = [−0.21, 0.10], was significant predictors of risk-taking. In support of our hypothesis, there was a significant trait approach motivation by economic anxiety interaction at Step 2, R² = .02, ΔR² = .02, ΔF(1, 425) = 8.64, p = .003, B = 0.44, SE = .15, 95% CI = [0.15, 0.74].

A simple effects analysis demonstrated that among participants high in trait approach motivation (+1 SD), economic anxiety increased risk-taking behavior compared with the control condition, B = 0.27, SE = .14, t(425) = 1.99, p = .047, 95% CI = [0.004, 0.54]. Among participants low in trait approach motivation (−1 SD), economic anxiety decreased risk-taking behavior compared with the control condition, B = −0.29, SE = .14, t(425) = −2.17, p = .031, 95% CI = [−0.56, −0.03]. These results support the hypothesis that trait approach-motivated participants are prone to risk-taking behavior following an anxious experience (Figure 3).

Figure 3.

Study 3: Standardized mean adjusted pumps on the BART (i.e., risk-taking) as a function of trait approach motivation and condition (anxiety vs. control). Error bars represent standard errors of the mean.

Discussion

Past research has demonstrated that anxiety causes palliative RAM (McGregor et al., 2010a) and that approach motivation disrupts sensitivity to negative outcomes (Nash et al., 2012) and heightens sensitivity to rewards (Carver & White, 1994). The present studies demonstrate that anxiety can lead to risk-taking behavior. Furthermore, Study 2 shows that, among trait approach-motivated participants, as BIS-specific theta measured during an anxious experience increased, risk-taking increased. RAM responses to anxiety thus appear to leave approach-motivated people insensitive to negative outcomes and prone to risk-taking behavior. In sum, these results help illuminate the puzzling phenomenon of why, just after neurophysiological processes signal “Be careful!” some people sometimes strive forward more recklessly.

Emotion and Motivation in Risk-Taking

The current research adds to the evolving recognition of emotion’s role in risk-taking. For example, Loewenstein and colleagues (2001) reviewed findings that show the perception of risk and reward elicits anticipatory emotions that diverge from cognitive assessments and uniquely guide decisions. Several studies have also shown mood congruency effects—in general, negative affect increases risk-aversion whereas positive affect increases risk-taking (Bower, 1991; Johnson & Tversky, 1983). Lerner and Keltner (2001) have shown that fear increases risk-aversion whereas anger increases risk-taking. In this case, two negative emotions caused divergent decisions. These last results were taken to demonstrate that it is not the valence of the emotion that dictates the effect on risk-taking but the specific emotion. This does contradict a large body of research supporting a mood-congruency effect, however (e.g., Bower, 1991; Johnson & Tversky, 1983). Critically, the emotions fear and anger do differ on a separate, important dimension—motivational direction. Fear is an avoidance motivation whereas anger is an approach motivation (Carver & Harmon-Jones, 2009; Panksepp, 1998). This raises the possibility that past evidence does not reflect a mood congruency effect but rather an approach/avoidance congruency effect on risk-taking. There is a high degree of overlap between positive emotions and approach emotions, which could explain the evidence in favor of mood congruency (Carver & Harmon-Jones, 2009). Moreover, research has found that feelings of power, dominance, curiosity, and excitement—all approach-related states—predict risk-taking (Anderson & Galinsky, 2006; Demaree et al., 2009; Maner & Gerend, 2007). Future research should examine whether the degree to which emotions impact decision-making under risk and uncertainty are due to their implications for motivational direction rather than or in conjunction with emotional valence.

Implications for Financial Risk-taking

The present findings could offer new insights into instances of financial risk-taking. In the aftermath of the 2007–2008 subprime mortgage crisis, Zweig (2009) reported that some investors were behaving surprisingly reckless despite the declining market, transferring money from safer, balanced funds to far riskier ones. Zweig (2009) described this perplexing behavior as the financial equivalent of the “Hail Mary pass” in football—a last-ditch, low probability heave, typically made in desperation. Behavioral economics theories point to cold, cognitive explanations that are often devoid of incidental emotion (e.g., the break-even effect; Thaler & Johnson, 1990), despite the considerable angst and uncertainty present during market recessions. Instead, we offer a general, motivational explanation. We speculate that some individuals, specifically those high in trait approach motivation (who may indeed be over-represented among traders; Jana et al., 2021), may tend toward riskier financial options during anxious and uncertain times to alleviate the uncomfortable emotional state. Thus, we echo and extend the recommendations of Smith et al. (2009); investors may benefit from taking time off or being closely monitored after incurring anxiety-provoking losses. Our findings also show that financial reactive risk-taking can eventuate from anxious experiences unrelated to the financial domain (e.g., achievement anxiety in Study 1). It is possible that incidental angst and uncertainty caused by the COVID-19 pandemic contributed to instances of increased contrarian trading activity (responding to negative news about a stock by buying it) and the popularity of risky technology and gamble stocks on retail trading platforms like Robinhood (Pagano et al., 2021; Welch, 2021). Future research could examine this hypothesis.

Limitations and Future Directions

The current research examines the seemingly ironic phenomenon of anxiety-inspired risky behavior. In doing so, we provide novel experimental evidence that this phenomenon is a result of palliative RAM. However, certain questions remain. First, it is unclear from the current research precisely how RAM leaves approach-motivated people prone to reactive risk-taking. Decisions involving risk involve a choice between prospects (Kahneman & Tversky, 1979). That is, risk decisions require us to contrast the likelihoods of varying positive and negative outcomes. Thus, risk-taking can occur when positive outcomes are overvalued and/or negative outcomes are undervalued. There is substantial evidence that approach motivation heightens sensitivity to positive outcomes and motivationally salient stimuli. For example, traits related to approach motivation are associated with positive reactions to expected rewards (Carver & White, 1994), sensitivity to potential gains (Smillie & Jackson, 2005), and reward-maximizing strategies (Higgins, 1997). There is also evidence that approach motivation lessens sensitivity to negative outcomes. The joint subsystem hypothesis (Corr, 2002) proposes that to facilitate focused goal pursuit, the BAS inhibits BIS-mediated processes associated with sensitivity to negative outcomes and reactivity to aversive events. Consistent with this, approach motivation-related traits (Higgins, 1998) and patterns of brain activity (Nash et al., 2012) antagonize sensitivity to aversive stimuli. Although much research has focused on the interplay between approach and aversive motivation systems, future research could examine whether reactive risk-taking specifically is a consequence of inhibited BIS-related sensitivity to negative outcomes, enhanced BAS-related sensitivity to positive outcomes, or a combination of the two.

Finally, the current research focused on how trait approach-motivated individuals specifically respond to anxious experiences with respect to risk-taking behavior on the BART. Although the BART exhibits strong ecological validity, future studies could extend this research to other measures of risk-taking in different domains, for example, social or ethical domains. Relatedly, people display significant heterogeneity in how they respond to anxiety as well as in how they address risk. For example, some individuals respond to anxiety with withdrawal, reduced approach motivation, and increased avoidance motivation (J. Hayes et al., 2016; Park, 2010), responses that have been linked to risk aversion in past research (e.g., Friedman & Förster, 2002). For some, avoidance motivation may afford relief from anxious worry by providing security from uncertainty via the prevention of negative outcomes. This increased sensitivity to negative outcomes may in turn leave individuals less likely to engage in risk-taking behavior. Interestingly in Study 3, we did not find a reduced risk-taking response to anxiety among participants high in trait neuroticism, which is thought to reflect the sensitivity of the avoidance system (McNaughton et al., 2016). We did find partial evidence of a reduced risk-taking response to anxiety among individuals low in trait approach motivation in Study 3 (however no change in Study 1). Although substantial research has examined the link between avoidance motivation and risk-aversion, research investigating avoidance-motivated reactions to anxiety and their downstream consequences on risk-taking are limited. Future research could examine this possibly, particularly among less secure individuals (e.g., low self-esteem and insecurely attached), who may lack the self-confidence to respond to anxious experiences with unbridled approach motivation (Park, 2010). Similarly, future research could examine how different aspects of trait approach motivation (e.g., BAS-subscales; drive, fun-seeking, and reward-responsiveness) relate to reactive risk-taking.

Conclusion

This research shows that anxiety can lead to risk-taking behavior and provides further evidence that this effect is due to RAM processes. Our findings extend prior research on neural and psychological aspects of risk-taking and RAM. They also hint at how several types of problematic, risk-driven behavior may be curbed. If anxiety can lead to risk-taking then anxiolytic interventions could reduce risk-taking. Certain drugs are anxiolytic to the extent that they reduce anxiety symptoms and act directly on the BIS (McNaughton & Corr, 2004). Perhaps these anxiolytics could relieve the heightened tendency for risk-taking, like problem gambling, in the clinically anxious (e.g., Lloyd et al., 2010). For managing everyday anxieties in nonclinical individuals, several simple, social–psychological manipulations have a similar anxiolytic effect. For example, self-affirming personal values and strengths relieve anxiety (Creswell et al., 2005) as does attributing anxiety to external rather than internal sources (Nash et al., 2011). Perhaps these simple manipulations could also help reduce risky reactions to anxiety.

Supplemental Material

sj-docx-1-psp-10.1177_01461672211059689 – Supplemental material for Reactive Risk-Taking: Anxiety Regulation Via Approach Motivation Increases Risk-Taking Behavior

Supplemental material, sj-docx-1-psp-10.1177_01461672211059689 for Reactive Risk-Taking: Anxiety Regulation Via Approach Motivation Increases Risk-Taking Behavior by Josh Leota, Kyle Nash and Ian McGregor in Personality and Social Psychology Bulletin

Footnotes

Appendix A

Appendix B

Appendix C

Appendix D

Appendix E

Appendix F

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.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: J.L. was supported by the Alberta Gambling Research Institute (AGRI) and the Australian Government Research Training Program (RTP) scholarship.

ORCID iD

Josh Leota

Supplemental Material

Supplemental material is available online with this article.

Notes

References

Aberson

C. L.

(2019). Applied power analysis for the behavioral sciences. Taylor & Francis Group.

Aiken

L. S.

West

S. G.

Reno

R. R.

(1991). Multiple regression: Testing and interpreting interactions. SAGE.

Anderson

Galinsky

A. D.

(2006). Power, optimism, and risk taking. European Journal of Social Psychology, 36(4), 511–536. https://doi.org/10.1002/ejsp.324

Bishop

S. J.

Duncan

Lawrence

A. D.

(2004). State anxiety modulation of the amygdala response to unattended threat-related stimuli. The Journal of Neuroscience, 24, 10364–10368. https://doi.org/10.1523/JNEUROSCI.2550-04.2004

Bower

G. H.

(1991). Mood congruity of social judgment. In Forgas

(Ed.), Emotion and social judgments (pp. 31–54). Pergamon Press.

Carver

C. S.

Harmon-Jones

(2009). Anger is an approach-related affect: Evidence and implications. Psychological Bulletin, 135(2), 183–204. https://doi.org/10.1037/a0013965

Carver

C. S.

White

T. L.

(1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS scales. Journal of Personality and Social Psychology, 67(2), 319–333.

Cavanagh

J. F.

Shackman

A. J.

(2015). Frontal midline theta reflects anxiety and cognitive control: Meta-analytic evidence. Journal of Physiology-Paris, 109(1–3), 3–15. https://doi.org/10.1016/j.jphysparis.2014.04.003

Coan

J. A.

Allen

J. J.

(2004). Frontal EEG asymmetry as a moderator and mediator of emotion. Biological Psychology, 67(1–2), 7–50. https://doi.org/10.1016/j.biopsycho.2004.03.002

10.

Community and Public Health. (2012, December 1). Christchurch city health profile: Problem gambling [Press release]. https://www.healthychristchurch.org.nz/media/34858/problemgambling.pdf

11.

Corr

P. J.

(2002). J.A. Gray’s reinforcement sensitivity theory: Tests of the joint subsystems hypothesis of anxiety and impulsivity. Personality and Individual Differences, 33, 511–532. https://doi.org/10.1016/S0191-8869(01)00170-2

12.

Corr

P. J.

(2008). The reinforcement sensitivity theory of personality. Cambridge University Press.

13.

Creswell

J. D.

Welch

W. T.

Taylor

S. E.

Sherman

D. K.

Gruenewald

T. L.

Mann

(2005). Affirmation of personal values buffers neuroendocrine and psychological stress responses. Psychological Science, 16(11), 846–851. https://doi.org/10.1111/j.1467-9280.2005.01624.x

14.

Demaree

H. A.

DeDonno

M. A.

Burns

K. J.

Feldman

Everhart

D. E.

(2009). Trait dominance predicts risk-taking. Personality and Individual Differences, 47(5), 419–422. https://doi.org/10.1016/j.paid.2009.04.013

15.

Derryberry

Reed

M. A.

(2002). Anxiety-related attentional biases and their regulation by attentional control. Journal of Abnormal Psychology, 111, 225–236. https://doi.org/10.1037/0021-843X.111.2.225

16.

Dula

C. S.

Adams

C. L.

Miesner

M. T.

Leonard

R. L.

(2010). Examining relationships between anxiety and dangerous driving. Accident Analysis & Prevention, 42(6), 2050–2056. https://doi.org/10.1016/j.aap.2010.06.016

17.

Endler

N. S.

Kocovski

N. L.

(2001). State and trait anxiety revisited. Journal of Anxiety Disorders, 15(3), 231–245. https://doi.org/10.1016/S0887-6185(01)00060-3

18.

Friedman

R. S.

Förster

(2002). The influence of approach and avoidance motor actions on creative cognition. Journal of Experimental Social Psychology, 38(1), 41–55. https://doi.org/10.1006/jesp.2001.1488

19.

Gambetti

Giusberti

(2012). The effect of anger and anxiety traits on investment decisions. Journal of Economic Psychology, 33(6), 1059–1069. https://doi.org/10.1016/j.joep.2012.07.001

20.

Garvey

Murphy

(2007). Do losses linger? The Journal of Portfolio Management, 33, 75–83. https://doi.org/10.3905/jpm.2007.690608

21.

Gasper

Clore

G. L.

(1998). The persistent use of negative affect by anxious individuals to estimate risk. Journal of Personality and Social Psychology, 74(5), 1350–1363. https://doi.org/10.1037/0022-3514.74.5.1350

22.

Gianotti

L. R. R.

Knoch

Faber

P. L.

Lehmann

Pascual-Marqui

R. D.

Diezi

. . . Fehr

(2009). Tonic activity level in the right prefrontal cortex predicts individuals’ risk taking. Psychological Science, 20, 33–38. https://doi.org/10.1111/j.1467-9280.2008.02260.x

23.

Gosling

S. D.

Rentfrow

P. J.

Swann

W. B

Jr . (2003). A very brief measure of the Big-Five personality domains. Journal of Research in Personality, 37(6), 504–528. https://doi.org/10.1016/S0092-6566(03)00046-1

24.

Gratton

Coles

M. G.

Donchin

(1983). A new method for off-line removal of ocular artifact. Electroencephalography and Clinical Neurophysiology, 55(4), 468–484. https://doi.org/10.1016/0013-4694(83)90135-9

25.

Gray

J. A.

McNaughton

(2000). The neuropsychology of anxiety: An enquiry into the functions of the septo-hippocampal system. Oxford University Press.

26.

Jiang

Luo

Y. J.

(2010). Anxiety and outcome evaluation: The good, the bad and the ambiguous. Biological Psychology, 85(2), 200–206. https://doi.org/10.1016/j.biopsycho.2010.07.001

27.

Harmon-Jones

Bastian

Harmon-Jones

(2016). Detecting transient emotional responses with improved self-report measures and instructions. Emotion, 16(7), 1086–1096. https://doi.org/10.1037/emo0000216

28.

Harmon-Jones

Amodio

D. M.

Harmon-Jones

(2009). Action-based model of dissonance: A review, integration, and expansion of conceptions of cognitive conflict. Advances in Experimental Social Psychology, 41, 119–166. https://doi.org/10.1016/S0065-2601(08)00403-6

29.

Hayes

A. F.

(2017). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press.

30.

Hayes

Ward

C. L.

McGregor

(2016). Why bother? Death, failure, and fatalistic withdrawal from life. Journal of Personality and Social Psychology, 110(1), 96–115. https://doi.org/10.1037/pspp0000039

31.

Higgins

E. T.

(1997). Beyond pleasure and pain. American Psychologist, 52, 1280–1300. https://doi.org/10.1037/0003-066X.52.12.1280

32.

Higgins

E. T.

(1998). Promotion and prevention: Regulatory focus as a motivational principle. Advances in Experimental Social Psychology, 30, 1–46. https://doi.org/10.1016/S0065-2601(08)60381-0

33.

Hirsh

J. B.

Mar

R. A.

Peterson

J. B.

(2012). Psychological entropy: A framework for understanding uncertainty-related anxiety. Psychological Review, 119, 304–320. https://doi.org/10.1037/a0026767

34.

Hockey

G. R. J.

John Maule

Clough

P. J.

Bdzola

(2000). Effects of negative mood states on risk in everyday decision making. Cognition & Emotion, 14(6), 823–855. https://doi.org/10.1080/02699930050156654

35.

Hunt

M. K.

Hopko

D. R.

Bare

Lejuez

C. W.

Robinson

E. V.

(2005). Construct validity of the balloon analog risk task (BART) associations with psychopathy and impulsivity. Assessment, 12(4), 416–428. https://doi.org/10.1177/1073191105278740

36.

Jana

Kummer

Schmidt

(2021, June 25). The effects of approach/avoidance motivation and gain/loss-framing on the processing of information cues by non-professional investors. SSRN. https://doi.org/10.2139/ssrn.3664856

37.

Johnson

E. J.

Tversky

(1983). Affect, generalization, and the perception of risk. Journal of Personality and Social Psychology, 45, 20–31. https://doi.org/10.1037/0022-3514.45.1.20

38.

Jonas

McGregor

Klackl

Agroskin

Fritsche

Holbrook

. . . Quirin

(2014). Threat and defense: From anxiety to approach. In Olson

J. M.

Zanna

M. P.

(Eds.), Advances in experimental social psychology (Vol. 49, pp. 219–286). Academic Press. https://doi.org/10.1016/B978-0-12-800052-6.00004-4

39.

Jorm

A. F.

Christensen

Henderson

A. S.

Jacomb

P. A.

Korten

A. E.

Rodgers

(1998). Using the BIS/BAS scales to measure behavioural inhibition and behavioural activation: Factor structure, validity and norms in a large community sample. Personality and Individual Differences, 26(1), 49–58. https://doi.org/10.1016/S0191-8869(98)00143-3

40.

Kahneman

Tversky

(1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291.

41.

Kashdan

T. B.

Collins

R. L.

Elhai

J. D.

(2006). Social anxiety and positive outcome expectancies on risk-taking behaviors. Cognitive Therapy and Research, 30(6), 749–761. https://doi.org/10.1007/s10608-006-9017-x

42.

Kim

D. Y.

Lee

J. H.

(2011). Effects of the BAS and BIS on decision-making in a gambling task. Personality and Individual Differences, 50(7), 1131–1135. https://doi.org/10.1016/j.paid.2011.01.041

43.

Knoch

Gianotti

L. R. R.

Pascual-Leone

Treyer

Regard

Hohmann

Brugger

(2006). Disruption of right prefrontal cortex by low-frequency repetitive transcranial magnetic stimulation induces risk-taking behavior. Journal of Neuroscience, 26, 6469–6472. https://doi.org/10.1523/JNEUROSCI.0804-06.2006

44.

Kuhl

(1994). A theory of action and state orientations. Volition and Personality: Action versus State Orientation, 9–46.

45.

Kumar

(2009). Who gambles in the stock market? The Journal of Finance, 64(4), 1889–1933. https://doi.org/10.1111/j.1540-6261.2009.01483.x

46.

Lauriola

Russo

P. M.

Lucidi

Violani

Levin

I. P.

(2005). The role of personality in positively and negatively framed risky health decisions. Personality and Individual Differences, 38(1), 45–59. https://doi.org/10.1016/j.paid.2004.03.020

47.

Lejuez

C. W.

Read

J. P.

Kahler

C. W.

Richards

J. B.

Ramsey

S. E.

Stuart

G. L.

. . . Brown

R. A.

(2002). Evaluation of a behavioral measure of risk taking: The Balloon Analogue Risk Task (BART). Journal of Experimental Psychology, 8(2), 75–84. https://doi.org/10.1037/1076-898X.8.2.75

48.

Leota

Nash

(2021). Incidental anxiety caused by economic threat heightens white noise P3 amplitude [Manuscript in preparation]. Department of Psychology, University of Alberta.

49.

Lerner

J. S.

Keltner

(2000). Beyond valence: Toward a model of emotion-specific influences on judgement and choice. Cognition and Emotion, 14, 473–493.

50.

Lerner

J. S.

Keltner

(2001). Fear, anger, and risk. Journal of Personality and Social Psychology, 81(1), 146–159. https://doi.org/10.1037/0022-3514.81.1.146

51.

Litman

Robinson

Abberbock

(2017). TurkPrime. com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behavior Research Methods, 49(2), 433–442. https://doi.org/10.3758/s13428-016-0727-z

52.

Lloyd

Doll

Hawton

Dutton

W. H.

Geddes

J. R.

Goodwin

G. M.

Rogers

R. D.

(2010). How psychological symptoms relate to different motivations for gambling: An online study of internet gamblers. Biological Psychiatry, 68, 733–740. https://doi.org/10.1016/j.biopsych.2010.03.038

53.

Loewenstein

Weber

E. U.

Hsee

C. K.

Welch

(2001). Risk as feelings. Psychological Bulletin, 127, 267–286. https://doi.org/10.1037/0033-2909.127.2.267

54.

Maner

J. K.

Gerend

M. A.

(2007). Motivationally selective risk judgments: Do fear and curiosity boost the boons or the banes? Organizational Behavior and Human Decision Processes, 103(2), 256–267. https://doi.org/10.1016/j.obhdp.2006.08.002

55.

Maner

J. K.

Richey

J. A.

Cromer

Mallott

Lejuez

C. W.

Joiner

T. E.

Schmidt

N. B.

(2007). Dispositional anxiety and risk-avoidant decision-making. Personality and Individual Differences, 42(4), 665–675. https://doi.org/10.1016/j.paid.2006.08.016

56.

McGregor

Gailliot

M. T.

Vasquez

N. A.

Nash

K. A.

(2007). Ideological and personal zeal reactions to threat among people with high self-esteem: Motivated promotion focus. Personality and Social Psychology Bulletin, 33, 1587–1599. https://doi.org/10.1177/0146167207306280

57.

McGregor

Haji

Nash

K. A.

Teper

(2008). Religious zeal and the uncertain self. Basic and Applied Social Psychology, 30(2), 183–188. https://doi.org/10.1080/01973530802209251

58.

McGregor

Nash

K. A.

Inzlicht

(2009). Threat, high self-esteem, and reactive approach-motivation: Electroencephalographic evidence. Journal of Experimental Social Psychology, 45, 1003–1007. https://doi.org/10.1016/j.jesp.2009.04.011

59.

McGregor

Nash

K. A.

Mann

Phills

C. E.

(2010a). Anxious uncertainty and reactive approach motivation (RAM). Journal of Personality and Social Psychology, 99, 133–147. https://doi.org/10.1037/a0019701

60.

McGregor

Nash

K. A.

Prentice

(2010b). Reactive approach motivation (RAM) for religion. Journal of Personality and Social Psychology, 99, 148–161. https://doi.org/10.1037/a0019702

61.

McNaughton

Corr

P. J

. (2004). A two-dimensional neuropsychology of defense: Fear/anxiety and defensive distance. Neuroscience & Biobehavioral Reviews, 28(3), 285–305. https://doi.org/10.1016/j.neubiorev.2004.03.005

62.

McNaughton

DeYoung

C. G.

Corr

P. J.

(2016). Approach/avoidance. In Absher

J. R.

Cloutier

(Eds.), Neuroimaging personality, social cognition, and character (pp. 25–49). Academic Press.

63.

McNaughton

Kocsis

Hajos

(2007). Elicited hippocampal theta rhythm: A screen for anxiolytic and procognitive drugs through changes in hippocampal function? Behavioural Pharmacology, 18(5–6), 329–346. https://doi.org/10.1097/FBP.0b013e3282ee82e3

64.

McNaughton

Swart

Neo

Bates

Glue

(2013). Anti-anxiety drugs reduce conflict-specific “theta”—A possible human anxiety-specific biomarker. Journal of Affective Disorders, 148(1), 104–111. https://doi.org/10.1016/j.jad.2012.11.057

65.

Mitte

(2007). Anxiety and risky decision-making: The role of subjective probability and subjective costs of negative events. Personality and Individual Differences, 43(2), 243–253. https://doi.org/10.1016/j.paid.2006.11.028

66.

Mohr

Biele

Heekeren

(2010). Neural processing of risk. Journal of Neuroscience, 30, 6613–6619. https://doi.org/10.1523/JNEUROSCI.0003-10.2010

67.

Nash

Inzlicht

McGregor

(2012). Prefrontal EEG asymmetry characteristic of approach motivation predicts muted error-related negativity. Biological Psychology, 91, 96–102. https://doi.org/10.1016/j.biopsycho.2012.05.005

68.

Nash

McGregor

Inzlicht

(2010). Line bisection as a neural marker of approach motivation. Psychophysiology, 47, 979–983. https://doi.org/10.1111/j.1469-8986.2010.00999.x

69.

Nash

McGregor

Prentice

(2011). Threat and defense as goal regulation: From implicit goal conflict to anxious uncertainty, reactive approach motivation, and ideological extremism. Journal of Personality and Social Psychology, 101, 1291–1301. https://doi.org/10.1037/a0025944

70.

Neo

P. S. H.

McNaughton

(2011). Frontal theta power linked to neuroticism and avoidance. Cognitive, Affective, & Behavioral Neuroscience, 11(3), 396–403. https://doi.org/10.3758/s13415-011-0038-x

71.

Neo

P. S. H.

Thurlow

J. K.

McNaughton

(2011). Stopping, goal-conflict, trait anxiety and frontal rhythmic power in the stop-signal task. Cognitive, Affective, & Behavioral Neuroscience, 11(4), 485–493. https://doi.org/10.3758/s13415-011-0046-x

72.

Nicholson

Soane

Fenton-O’Creevy

Willman

(2005). Personality and domain-specific risk taking. Journal of Risk Research, 8(2), 157–176. https://doi.org/10.1080/1366987032000123856

73.

Pagano

M. S.

Sedunov

Velthuis

(2021). How did retail investors respond to the COVID-19 pandemic? The effect of Robinhood brokerage customers on market quality. Finance Research Letters, 43, Article 101946. https://doi.org/10.1016/j.frl.2021.101946

74.

Panksepp

(1998). Affective neuroscience. The foundations of human and animal emotions. Oxford University Press.

75.

Park

L. E.

(2010). Responses to self-threat: Linking self and relational constructs with approach and avoidance motivation. Social and Personality Psychology Compass, 4(3), 201–221. https://doi.org/10.1111/j.1751-9004.2009.00247.x

76.

Santesso

D. L.

Segalowitz

S. J.

(2009). The error-related negativity is related to risk taking and empathy in young men. Psychophysiology, 46, 143–152. https://doi.org/10.1111/j.1469-8986.2008.00714.x

77.

Schumann

McGregor

Nash

K. A.

Ross

(2014). Religious magnanimity: Reminding people of their religious belief system reduces hostility after threat. Journal of Personality and Social Psychology, 107(3), 432–453. https://doi.org/10.1037/a0036739

78.

Shackman

A. J.

McMenamin

B. W.

Maxwell

J. S.

Greischar

L. L.

Davidson

R. J.

(2009). Right dorsolateral prefrontal cortical activity and behavioral inhibition. Psychological Science, 20(12), 1500–1506. https://doi.org/10.1111/j.1467-9280.2009.02476.x

79.

Shiller

(2005). Irrational exuberance (2nd ed.). Princeton University Press.

80.

Smillie

L. D.

Jackson

C. J.

(2005). The appetitive motivation scale and other BAS measures in the prediction of approach and active avoidance. Personality and Individual Differences, 38(4), 981–994. https://doi.org/10.1016/j.paid.2004.09.013

81.

Smith

Levere

Kurtzman

(2009). Poker player behavior after big wins and big losses. Management Science, 55, 1547–1555. https://doi.org/10.1287/mnsc.1090.1044

82.

Thaler

R. H.

Johnson

E. J.

(1990). Gambling with the house money and trying to break even: The effects of prior outcomes on risky choice. Management Science, 36(6), 643–660. https://doi.org/10.1287/mnsc.36.6.643

83.

Tversky

Kahneman

(1981). The framing of decisions and the rationality of choice. Science, 211(4481), 453–458.

84.

Welch

(2021). The wisdom of the Robinhood crowd. Journal of Finance, Forthcoming. https://doi.org/10.2139/ssrn.3696066

85.

Zhang

(2018). Behavioral preference in sequential decision-making and its association with anxiety. Human Brain Mapping, 39(6), 2482–2499. https://doi.org/10.1002/hbm.24016

86.

Zuckerman

Kuhlman

D. M.

(2000). Personality and risk-taking: Common biosocial factors. Journal of Personality, 68, 999–1029. https://doi.org/10.1111/1467-6494.00124

87.

Zweig

(2009, February 21). As stock losses loom, don’t throw a “Hail Mary.” The Wall Street Journal. https://www.wsj.com/articles/SB123518055107638381

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB