Alcohol’s Effects During Uncertain and Uncontrollable Stressors in the Laboratory

Abstract

Alcohol’s effects on reactivity to stressors depend on the nature of the stressor and the reactivity being assessed. Research that identifies characteristics of stressors that modulate reactivity and clarifies the neurobehavioral, cognitive, and affective components of this reactivity may help prevent, reduce, or treat the negative impacts of acute and chronic alcohol use and have implications for other psychopathology involving maladaptive reactivity to stressors. We used a novel, multimeasure, cued, electric-shock-stressor paradigm in a greater-university community sample of adult recreational drinkers to test the effects of alcohol (n = 64), compared with no alcohol (n = 64), on reactivity to stressors that vary in both their perceived certainty and controllability. Preregistered analyses suggested alcohol significantly dampened subjective anxiety (self-report) and defensive reactivity (startle potentiation) more during uncertain stressors than during certain stressors regardless of controllability, which suggests that stressor uncertainty—but not uncontrollability—may be sufficient to enhance alcohol’s dampening of stress reactivity and thus negative reinforcement potential.

Keywords

addictive disorders psychopharmacology psychophysiology stress reactions threat open data open materials preregistered

For millennia, humans have used drugs such as alcohol to avoid or alleviate the detrimental effects of psychological stressors (Hanson, 2013; Sher, 1987). Frequent, excessive use of alcohol when “drinking to cope” with some stressors may eventually lead to exaggerated stressor reactivity, which may cause, maintain, or exacerbate alcohol use disorder and other psychopathology (Kaye et al., 2017; Koob & Volkow, 2010). Different types of stressors modulate reactivity in different ways, and alcohol may not diminish reactivity to all stressors equally (Sayette, 2017). Research that both identifies key characteristics of stressors that modulate reactivity and clarifies the neurobehavioral, cognitive, and affective components of this reactivity may help prevent, reduce, or treat the negative impacts of acute and chronic alcohol use behavior. This research may also guide similar efforts with other psychopathology involving maladaptive reactivity to stressors (e.g., pathological anxiety and depression; Grillon, 2008; Havranek et al., 2016; Insel et al., 2010; Maier, 2015).

Researchers have identified several distinct characteristics of stressors that modulate reactivity (Davis et al., 2010; Kaye et al., 2017; Maier, 2015; Zvolensky et al., 2000). In particular, parallel but separate lines of evidence indicate that stressors that are either uncertain or uncontrollable elicit greater reactivity and have more detrimental effects on physical health and cognitive performance than their certain and controllable counterparts (Henderson et al., 2012; Koolhaas et al., 2011; Mineka & Kihlstrom, 1978; Robinson et al., 2013). Characteristics of certainty and controllability have also emerged as important in the etiology and/or manifestations of some mood and anxiety disorders. Specifically, patients with either posttraumatic stress disorder or panic disorder display exaggerated reactivity selectively to uncertain stressors relative to certain stressors (Grillon et al., 2008; Grillon, Pine, et al., 2009). In parallel, exposure to uncontrollable stressors but not controllable stressors has figured prominently in etiologic models of depression such as “learned helplessness” in animals and “learned hopelessness” in humans (Abramson et al., 1978; Liu et al., 2015).

Recent research has also focused specifically on reactivity to uncertain stressors to clarify the negative reinforcing effects of anxiolytic medications and popular recreational yet addictive drugs such as alcohol (Davis et al., 2010; Kaye et al., 2017; Maier, 2015). Commonly prescribed anxiolytics (e.g., benzodiazepines, selective serotonin-reuptake inhibitors) appear to reduce reflexive defensive reactivity (i.e., startle potentiation) to a greater degree when experimentally manipulated stressors (e.g., cued threat of electric shock) are presented with uncertainty (e.g., unpredictable timing) rather than with certainty (e.g., predictable timing; Grillon, Chavis, et al., 2009; Grillon et al., 2006). Likewise, a programmatic series of experiments in our laboratory has demonstrated that single administrations of moderate to high doses of alcohol (e.g., a dose yielding blood alcohol concentrations [BACs] between 0.08% and 0.15%) temporarily reduce startle potentiation during uncertain stressors significantly more than certain stressors (Bradford et al., 2013, 2017; Hefner & Curtin, 2012; Hefner et al., 2013; Moberg & Curtin, 2009). Furthermore, Kaye et al. (2017) proposed that chronic use of alcohol causes etiologically relevant stress neuroadaptations (see Koob & Volkow, 2010) that manifest as persistent exaggerated reactivity to uncertain stressors in patients with alcohol use disorder. Moberg et al. (2017) provided preliminary support for this thesis (for comparable exaggerated reactivity to uncertain stressors in chronic marijuana users, see Hefner et al., 2018; for comparable exaggerated reactivity to uncertain stressors in chronic tobacco users, see Hogle et al., 2010).

It appears that stressors play an important role in the reinforcing effects of alcohol and other drugs and the etiology or manifestations of psychopathology involving addiction, anxiety, and depression. Furthermore, characteristics of these stressors that alter their certainty and controllability are crucial to consider as their impact is clarified. At least four important gaps remain in the understanding of these phenomena. First, research on stressor certainty and controllability typically focuses on only one of these characteristics, ignoring or even confounding the other (e.g., uncertain stressors are also uncontrollable in most experiments). This has prevented unambiguous identification of either the unique or interactive effects of each characteristic. Second, modern methods have largely focused on a single measure of a specific response domain (e.g., defensive reactivity). However, measures of defensive reactivity such as startle potentiation are affected by both affect and attention (Blumenthal, 2015). To assess clearly and comprehensively the multiple mechanisms affected by different stressor characteristics, a single study must use multiple measures.¹ Third, although Maier and others (e.g., Maier, 2015; Maier & Watkins, 2005) have conducted an elegant series of studies explicitly examining the effects of uncontrollable and controllable stressor exposure with animals, stressor controllability research using animal-to-human translational methods has only begun to be used in humans (Havranek et al., 2016; Löw et al., 2015; Wendt et al., 2017). Finally, although the effects of alcohol and other drugs on responses to uncertain stressors have received recent attention, no study to date has tested the effects of a pharmacological manipulation (e.g., alcohol) on responses to uncontrollable and controllable stressors.

These gaps leave three possibilities, each with unique implications about understanding of the effects of drugs on reactivity to stressors and that reactivity itself: (a) Alcohol’s effects on stressor reactivity are limited to stressors that are both uncertain and uncontrollable, which suggests a synergistic process across two stressor dimensions with implications for both the effects of alcohol and how stressor characteristics interact to affect reactivity; (b) alcohol reduces stressor reactivity more during uncertain stressors compared with certain stressors and also more during uncontrollable stressors compared with controllable stressors, which provides preliminary evidence for either distinct or shared drug-sensitive mechanisms for reactivity to both stressor types; and (c) alcohol selectively affects uncertain stressor reactivity without varying by stressor controllability, which provides an important boundary condition on the circumstances in which alcohol dampens stressor reactivity and additional evidence that human reactions to stressor certainty and controllability rely on distinct psychological and/or neurological mechanisms (Davis et al., 2010; Maier & Watkins, 2005; Wood et al., 2015). We present here the first attempt to discern between these possibilities by using a novel experimental stressor paradigm to test the effects of a moderate dose of alcohol (target BAC = 0.08%) on self-reported anxiety (subjective anxiety potentiation), defensive reactivity (startle potentiation), and emotionally motivated attention (suppression of the P3 event-related-potential [ERP] response to the startle probe, referred to hereafter as probe P3 ERP suppression; for a review of these measures, see Bradford et al., 2017) in a factorial 2 (uncertain, certain) × 2 (uncontrollable, controllable) fully crossed design.

Method

Open science and reproducibility

We preregistered our sample-size determination, data-processing parameters, and analysis plan at https://osf.io/q8d96/. Following recommendations about research transparency (Simmons et al., 2012), we have reported how we determined our sample size, all data exclusions, all manipulations, and all measures in the study. Following emerging open-science guidelines (Schönbrodt et al., 2015), we have made curated data and analysis scripts publicly available via OSF at https://osf.io/2zmxe/.

Participants

The study was approved by an institutional review board’s social and behavioral sciences human subjects committee, all procedures were carried out in accordance with the provisions of the Declaration of Helsinki, and informed written consent was obtained from all study participants. We recruited a final sample of 128 participants (64 female; for demographics, see Table 1) from the university community. Power analyses using the pwr (Version 1.3-0; Champely, 2020) package for the R software environment (Version 4.0.0; R Core Team, 2020) indicated that this sample size would provide at least 80% power to detect, at an α level of .05, the median effect size (partial η² = .07) observed across previous studies examining alcohol’s effects on startle potentiation during uncertain stressors compared with certain stressors (e.g., Bradford et al., 2013).

Table 1.

Demographics for Full Sample (N = 128)

Characteristic	Alcohol group	No-alcohol group
n	64	64
Female participants	50%	50%
Race/ethnicity
Non-White	22%	16%
Hispanic	5%	8%
Highest level of education
High school or equivalent degree	5%	5%
Some college or 2-year degree	30%	37%
College or advanced degree	65%	58%
Annual income
< $20,000/year	59%	63%
$20,000 to $50,000/year	29%	25%
$50,001 to ≥ $100,000/year	11%	11%
Did not respond	1%	1%
Age (years)
Mean	24.0	23.8
SD	3.2	2.8
Range	21–33	21–31
Current alcohol use
Frequency (occasions/week)
Mean	2.3	2.0
SD	1.7	1.4
Quantity (per occasion)
Mean	3.9	4.2
SD	2.2	2.4
Past-year alcohol problems
Mean	3.8	3.7
SD	2.1	2.1

Note: Self-reported alcohol problems were measured with the Young Adult Alcohol Problems Scale (Hurlbut & Sher, 1992).

Participants reported that they could read and write in English; were between 21 and 35 years old; had experience within the past year with drinking the quantities of alcohol needed to obtain the study dose of alcohol; reported no history of alcohol-related problems, no current psychiatric medication use, no uncorrected hearing or vision problems, and no alcohol-contraindicated medical condition; were not pregnant (verified by urine sample); and were sober on arrival (verified via breathalyzer; Alcosensor IV; Intoximeters Inc., St. Louis, MO). We paid participants $10/hr for their participation.

Beverage manipulation

Beverage manipulation procedures were consistent with our laboratory’s standards (e.g., Bradford et al., 2015). After a baseline assessment of general startle reactivity (see the Supplemental Material available online), we assigned participants at random to one of three beverage conditions: alcohol (n = 64), no alcohol/told alcohol (i.e., placebo; n = 32), and no alcohol/told no alcohol (i.e., true no alcohol; n = 32; for study flow diagram, see Fig. 1). Expectancies and other cognitions about alcohol and drinking may affect attentional measures (Testa et al., 2006), so we included separate placebo and true no-alcohol conditions to allow exploratory analysis of placebo effects. Past research suggests, however, that most of alcohol’s effects on affective responding to noxious stimuli, when present, appear to have a pharmacological basis (for a review, see Greeley & Oei, 1999). Furthermore, we have not found placebo effects in our past research on alcohol and threat uncertainty (Hefner et al., 2013), so our preregistered analyses focused on the contrast between the alcohol group and the combination of the two no-alcohol conditions (i.e., equal N for alcohol and no-alcohol groups). We provide more detail about the alcohol manipulation and results comparing placebo and true no-alcohol conditions for each dependent variable in the Supplemental Material.

Fig. 1.

Study flow diagram. Self-report individual measures were completed after posttask assessment for all participants. This figure is published under a Creative Commons CC-BY-4.0 license.

We divided beverages in all three conditions into two drinks, each consumed over 15 min, for a 30-min total drinking period. We began the experimental session approximately 15 min after the end of the drinking period. We measured BAC via breathalyzer (Alcosensor IV) immediately before and after completion of the main shock-threat task.

Shock workup procedures

After the beverage manipulation, we measured participants’ subjective tolerance to a series of 200-ms electric shocks of increasing intensity (7 mA maximum), delivered with a custom shock stimulator across the distal phalanges of the index and ring fingers of the nondominant hand, which follows standard procedures from our laboratory (Bradford et al., 2014). We set maximum shock intensity during the main task to each participant’s subjective maximum tolerance threshold to minimize individual differences in shock tolerance.

Task design

A PC-based MATLAB (The MathWorks, Natick, MA) script written with Psychophysics Toolbox (Version 3.0.13; Brainard, 1997; Pelli, 1997) presented visual and auditory stimuli. Participants viewed serially presented gray-scale square visual cues on a CRT monitor for 6 s each, separated by a variable intertrial interval of 9 to 15 s (M = 12 s). A number in black type in the center of each visual cue indicated impending shock intensity. Visual cues were presented in within-subjects blocks, and the same visual cue was repeated three times within each block. There were four types of shock block—certain and uncontrollable, uncertain and uncontrollable, certain but controllable, and uncertain but controllable—as well as one no-shock block type (see Fig. 2). Each type of shock block was presented twice (six cue presentations), and the no-shock block was presented three times (nine cue presentations), for a grand total of 11 blocks (33 cue presentations) presented in one of four counterbalanced orders.

Fig. 2.

Typical trial timeline. Each type of shock block was presented twice (three cues in each block). Three no-shock blocks (three cues in each block) with a black crosshair that turned gray when a joystick trigger was pulled were also presented (not shown). Intertrial intervals consisted of a black crosshair and no square (not shown). Electromyography startle and P3 ERP response to the startle probe were measured at 5.5 s into the cue presentation. Shocks occurred at 5.8 s and coterminated with cue offset at 6 s. Self-reported anxiety in response to each cue type was measured at a break in the middle and at the end of the task (not shown). This figure is published under a Creative Commons CC-BY-4.0 license.

For controllable blocks, we gave participants perceived control of a range of shock intensities, as opposed to, for example, control of shock occurrence or duration, so uncertain-shock trials could remain relatively uncertain even after being lowered (i.e., controlled). This design also allowed matching of number and average intensity of shocks received across block types to control for possible interactions of alcohol with intensity of the stress response (e.g., Moberg et al., 2011). Across all certain block cues, participants received shocks equal to 60% of their maximum tolerated shock intensity (12 shocks total). Across all uncertain-cue blocks, participants received a range of shocks: two shocks each at 20%, 40%, 80%, and 100% of their maximum tolerated shock intensity and four shocks at 60% intensity (mean intensity = 60%; 12 shocks total). This range was chosen, via pilot testing, to provide easily discernible intensities while ensuring that the lowest intensity was still aversive. Shock levels were pseudorandomized within each uncertain block. Each began with a mean-intensity (60%) shock to avoid interference with the control illusion that could have been caused by beginning with a higher intensity shock on controllable blocks.

We told participants they could receive 10 different shock intensities across blocks. In each of the two certain block types, the impending shock intensity was indicated by a specific number (e.g., 6) displayed in black type in the center of each cue. In each of the two uncertain block types, participants were told a range of possible shock values before block start (e.g., 4–8), but the exact impending shock intensity for each cue was unknown, and this was indicated by a black question mark in the center of the cue. We told participants they could lower the intensity of the shocks in certain blocks that were also deemed controllable (i.e., certain but controllable) by “two full levels” (e.g., 8 could be lowered to 6). Likewise, participants were told they could lower the range of possible shock intensities in uncertain blocks that were also deemed controllable (i.e., uncertain but controllable) by two full levels (e.g., 6–10 could be lowered to 4–8). We told participants they could not affect the shock intensity in either of the uncontrollable blocks (i.e., uncertain and uncontrollable; certain and uncontrollable). In all shock blocks, participants were aware that the 6-s cues for the shocks would coterminate with 200-ms shocks at the indicated intensity (i.e., shocks administered 5.8 s after cue onset).

Perceived control manipulation

We told participants they could control the shocks during controllable blocks using a custom made, fixed-position joystick with an on/off rocker switch and spring-loaded trigger. We instructed participants to use the rocker switch to turn on or off their ability to control the shocks before the start of controllable and uncontrollable shock blocks, respectively. Participants were then verbally asked, “Are you controlling the shocks now?” to which participants were directed to respond simply yes or no after they flipped the switch on or off. The switch controlled a custom light box illuminating the words “Shock Control ON” or “Shock Control OFF” positioned in front of the participant directly below the computer monitor. The rocker switch was also wired to an input channel on the data-collection computer in the experimenter control room to allow researchers to verify that the participant had properly used the switch. Regardless of the rocker switch position, the shock controller had no real effect on actual shock intensity because shock intensities were predetermined as described above.

To control for possible effects of motor movement or attentional effects on the startle or probe P3 responses, we instructed participants to quickly pull the trigger on the joystick after each cue was presented across all blocks regardless of shock controllability. We told participants to do this to show that they were paying attention during the task. Participants had the entire 6 s of the cue presentation to pull the trigger. The stimulus presentation computer recorded trigger pulls. On-screen instructions reminded participants to pull the trigger between every block. We discarded and replaced the data from three participants who failed to pull the trigger in time on a certain but controllable trial because this undermined the illusion of control. Data were not discarded if participants failed to pull the trigger on an uncertain but controllable trial (one participant) because the shock level remained uncertain on these trials. When participants pulled the trigger, the black text turned gray (on all trials) and shrunk in size (on controllable trials) to indicate that their trigger pull was recorded and that their control was effective. For a narrative summary of the presentation of each block type separately, see the Supplemental Material.

Midtask and posttask measures

During a short break at the midpoint and after the main task, participants rated how anxious/fearful they were on a 5-point rating scale (1 = not at all anxious/fearful, 5 = extremely anxious/fearful) when they saw each cue type. After the posttask anxiety ratings, participants completed a placebo manipulation check and a battery of self-report individual difference questionnaires (see the Supplemental Material) and were debriefed, compensated, and dismissed after reaching a BAC below 0.03%.

Startle-response measurement and processing

We measured the eye-blink startle response to acoustic startle probes (50 ms, 102 dB white noise with near instantaneous rise time). We presented six noise probes during a subset of the visual cues at 5.3 s after cue onset during the baseline procedure and 22 noise probes during a subset of the cues in the threat of shock blocks (16 probes; four for each cue type) and no-shock blocks (six probes) at 5.3 s after cue onset during the main task. We conducted offline data processing for all physiological signals using the PhysBox plugin (Version r267; Curtin, 2011) within the EEGLab toolbox (Version 2020.0; Delorme & Makeig, 2004) in MATLAB. We followed our preregistration and published guidelines (Blumenthal et al., 2005; Bradford et al., 2014) for startle-response reduction and processing (for more detail, see the Supplemental Material).

Probe P3 measurement and processing

We sampled the electroencephalogram (EEG) signal at 2500 Hz from three midline scalp sites (Fz, Cz, Pz,). We measured the P3 ERP response to the acoustic startle probes at electrode site Pz, consistent with our preregistration and recent published reports (Alius et al., 2015; Bradford et al., 2017; Nelson & Hajcak, 2017). Data from the two other electrode sites were collected for goals not relevant to the current study. EEG response reduction and processing followed published guidelines (Picton et al., 2000) and our preregistration (for more detail, see the Supplemental Material).

Analysis plan

Our analysis followed our preregistration plan and previous research (e.g., Bradford et al., 2013, 2017). We calculated general startle reactivity as mean startle response to startle probes during cues in the baseline procedure, startle potentiation as the mean increase in startle response to probes during cues in the shock blocks relative to cues in the no-shock blocks, subjective anxiety potentiation as the mean increase in anxiety in response to the shock cues relative to the no-shock cues, and probe P3 suppression as the mean decrease in the P3 ERP response to startle probes during cues in the shock blocks relative to cues in the no-shock blocks.

We analyzed each dependent variable in separate, general linear models (GLMs) using RStudio (Version 1.2; RStudio Team, 2016) for R with the lmSupport (Version 2.9.2; Curtin, 2015) package. All GLMs included between-subjects regressors for beverage group (alcohol vs. placebo vs. true no alcohol) and within-subjects factors for shock certainty (uncertain minus certain) and shock controllability (uncontrollable minus controllable). Beverage group was coded with two planned orthogonal contrasts to examine effects of alcohol (alcohol group vs. the no-alcohol group—a combination of the placebo and true no-alcohol groups—and the placebo manipulation (placebo group vs. true no-alcohol group). All a priori hypotheses concerning alcohol were interpreted using the alcohol contrast (no-alcohol group as reference), as preregistered. We tested the effects of controllability in the no-alcohol group as a manipulation check. Although not preregistered, we added additional tests of the effects of certainty (in the no-alcohol group) and the main effects of alcohol as additional manipulation checks for completeness and clarity. In the Supplemental Material, we report placebo effects (true no-alcohol as reference) and the preregistered description of covariate selection methods.

To address our substantive research questions, we tested for a three-way, Alcohol × Certainty × Controllability interaction for each dependent variable. Following our preregistration, in the presence of a significant three-way interaction, we tested three planned contrasts with comparing the certain and controllable reference condition with (a) startle potentiation in uncontrollable and uncertain, (b) startle potentiation in uncontrollable and certain, and (c) startle potentiation in controllable and uncertain. In absence of a significant three-way interaction, we tested the two, two-way Alcohol × Certainty and Alcohol × Controllability interactions as preregistered. For transparency, we report all predicted means and standard errors from GLMs for each individual threat-type variant by beverage group in the Supplemental Material.

We used our field’s standard p < .05 criteria for determining that results from tests were significantly different from expected if data were consistent with the null hypothesis. We report both η_p² and GLM coefficients (b) to describe effect sizes. We conducted outlier analysis (studentized residual with Bonferroni-corrected p < .05) in preliminary GLMs for each dependent variable. This resulted in the removal, without replacement, of two outliers for GLMs for subjective anxiety potentiation, eight outliers for GLMs for startle potentiation, and one probe P3 suppression outlier for GLMs for subjective anxiety potentiation.

Results

The mean BAC for participants administered alcohol was 0.07% (SD = 0.01%) immediately before the main task and 0.07% (SD = 0.01%) immediately after the task. Consistent with our previous work, the placebo manipulation was successful in establishing expectations of alcohol consumption and intoxication (for full analysis of the placebo manipulation and reliability of all measures, see the Supplemental Material).

In the no-alcohol group, subjective anxiety potentiation was significantly greater during uncertain threats than certain threats, η_p² = .42, b = 0.5, t(115) = 9.10, p < .001 (Fig. 3a). Startle potentiation was significantly greater during uncertain threats than certain threats, η_p² = .08, b = 4.0, t(114) = 3.10, p = .002 (Fig. 3b). Probe P3 suppression did not significantly differ during uncertain threats compared with certain threats, η_p² = .01, b = 0.8, t(121) = 1.25, p = .212 (Fig. 3c).

Fig. 3.

Effects of threat-certainty condition and alcohol group for each dependent variable. Error bars represent ±1 SE for point estimates of each dependent variable from the general linear model. Coefficients from the general linear model show the simple effect of alcohol for each stressor type and the Stressor Type × Alcohol interaction. Asterisks indicate significant differences between groups or conditions (*p < .05). This figure is published under a Creative Commons CC-BY-4.0 license.

In the no-alcohol group, subjective anxiety potentiation was significantly greater during uncontrollable threats than controllable threats, η_p² = .49, b = 0.7, t(115) = 10.31, p < .001 (Fig. 4a). Startle potentiation was significantly greater during controllable threats than uncontrollable threats, η_p² = .10, b = −4.3, t(107) = 3.44, p < .001 (Fig. 4b). Probe P3 suppression did not significantly differ during uncontrollable threats compared with controllable threats, η_p² = .00, b = 0.2, t(121) = 0.39, p = .700 (Fig. 4c).

Fig. 4.

Effects of threat-controllability condition and alcohol group for each dependent variable. Error bars represent ±1 SE for point estimates of each dependent variable from the general linear model. Coefficients from the general linear models how the simple effect of alcohol for each stressor type and the Stressor Type × Alcohol interaction. Asterisks indicate significant differences between groups or conditions (*p < .05). This figure is published under a Creative Commons CC-BY-4.0 license.

Across all stressors, alcohol significantly reduced subjective anxiety potentiation, η_p² = .07, b = −0.4, t(114) = 2.86, p = .005; startle potentiation, η_p² = .18, b = −13.0, t(107) = 4.89, p < .001; and probe P3 suppression, η_p² = .06, b = −3.2, t(121) = 2.78, p = .006. We did not observe a significant three-way Alcohol × Certainty × Controllability interaction for subjective anxiety potentiation, η_p² = .01, b = −0.1, t(115) = 1.11, p = .269. We observed a significant three-way Alcohol × Certainty × Controllability interaction for startle potentiation, η_p² = .08, b = −7.8, t(107) = 3.15, p = .002, but none of the follow-up contrasts were significant (ps > .192). We did not observe a significant three-way Alcohol × Certainty × Controllability interaction for probe P3 suppression, η_p² = .00, b = −0.7, t(121) = 0.57, p = .568.

The Alcohol × Certainty interaction was significant for subjective anxiety potentiation such that alcohol’s effect on subjective anxiety potentiation was significantly greater during uncertain threats, η_p² = .08, b = −0.5, t(115) = 3.25, p = .002, than certain threats, η_p² = .05, b = −0.3, t(115) = 2.26, p = .026; η_p² = .05, b = −0.2, t(115) = 2.48, p = .015 (Fig. 3a). The Alcohol × Certainty interaction was significant for startle potentiation such that alcohol’s effect on startle potentiation was significantly greater during uncertain threats, η_p² = .21, b = −15.3, t(107) = 5.30, p < .001, than certain threats, η_p² = .13, b = −10.7, t(107) = 3.92, p < .001; η_p² = .06, b = −4.6, t(107) = 2.55, p = .012 (Fig. 3b). The Alcohol × Certainty interaction was not significant for probe P3 suppression, η_p² = .03, b = −1.6, t(121) = 1.77, p = .080 (Fig. 3c).

The Alcohol × Controllability interaction was not significant for subjective anxiety potentiation, η_p² = .00, b = −0.1, t(115) = .63, p = .430 (Fig. 4a) startle potentiation, η_p² = .03, b = 3.2, t(107) = 1.85, p = .067 (Fig. 4b); or probe P3 suppression, η_p² = .00, b = −0.9, t(121) = 1.10, p = .275 (Fig. 4c).

Discussion

This preregistered laboratory study rests within a programmatic series of experiments showing evidence that alcohol dampens subjective anxiety and defensive reactivity more during uncertain stressors than certain stressors regardless of the type of uncertainty (e.g., uncertain probability, timing, location, or intensity; Bradford et al., 2017; Kaye et al., 2017). Like most modern stressor uncertainty research, however, the stressors in previous work were also uncontrollable, thus obscuring controllability’s influence on the effects of alcohol and uncertainty. Here, we used a novel design to simultaneously manipulate certainty and perceived controllability of stressor intensity. The presence of control significantly affected subjective anxiety and startle potentiation, but alcohol’s effects on these responses to uncontrollable stressors compared with controllable stressors did not differ. In fact, alcohol appeared to dampen subjective anxiety and startle potentiation more during uncertain stressors than during certain stressors regardless of controllability, which potentially establishes a boundary condition for the enhancement of alcohol’s effects on stressor reactivity. The present data suggest that stressor uncertainty—but not uncontrollability—is sufficient to enhance alcohol’s dampening of stress reactivity and thus negative reinforcement potential.

Although alcohol appears to consistently dampen reactivity to a greater degree during uncertain stressors than certain stressors (Kaye et al., 2017), in the present study, alcohol also dampened reactivity across all response domains, to some degree, regardless of stressor type. Thus, the current results may delimit the enhancement of alcohol’s effects on stressor reactivity but not the mere presence of such effects. In fact, alcohol may somewhat dampen reactivity to all types of stressors, thus reinforcing use and establishing addictive potential for many stressful situations. It remains unclear, however, how large of a dampening effect on stress reactivity would be sufficient to increase risk for addiction. Furthermore, three decades of so-called alcohol stress response dampening (SRD) laboratory research has made clear that alcohol’s effects on stress reactivity are often weakly observed, if observed at all (Sayette, 2017; Sher, 1987). The historical inconsistently of discernible alcohol SRD effects has reinforced the need to identify the stressor characteristics that most strongly and reliability predict it. The current results join other recent research (Bradford et al., 2017; Kaye et al., 2017) to suggest stressor certainty, perhaps more so than other stressor characteristics, is a particularly robust contributor to alcohol SRD.

Our findings may have implications for clarifying motivation for real-world drinking. The present data raise the interesting possibility that drinking to cope with stressors may be particularly reinforcing when those stressors are uncertain (e.g., you worry about the cost of an upcoming bill) regardless of whether it is uncontrollable (you think you cannot afford it) or controllable (you think you can afford it). Individuals with a large degree of uncertainty in their lives, such as people with a dearth of social support, separated families, or unstable housing, may be particularly sensitive to the negative reinforcing effects of alcohol even if they have resources, such as temporary housing or financial assistance, that would allow them to feel some control over those stressors.

Further clarifying the relative roles of certainty and controllability in alcohol’s acute and chronic effects on stressor reactivity could potentially inform the neurobiology of substance dependence etiology. Animal models of drug dependence postulate that repeated dampening of the neuropathways involved in response to uncertain stressors by alcohol or other drugs eventually leads to compensatory neuroadaptation that manifests as exaggerated negative affective responses to uncertain stressors that thus motivates further use to reduce negative affect (Kaye et al., 2017; Koob & Volkow, 2010). Complementary studies using stress paradigms with startle potentiation during uncertain stressors in recreational drinkers coupled with similarly conducted drug-deprivation experiments in individuals meeting criteria for drug dependence can provide cross-sectional evidence of the initial effects of drugs and the putative consequences of repeated use (for a review, see Kaye et al., 2017). Within this framework, the dampening of responses to uncertain stressors seen in the current study could be demonstration of the first stage of a proposed drug-induced neuroadaptation to uncertain stressors (Kaye et al., 2017). Preliminary cross-sectional research by our laboratory and others has demonstrated that individuals with alcohol use disorder show exaggerated startle potentiation to uncertain stressors, which potentially demonstrates compensatory changes in the consequent stage of these proposed neuroadaptations (Moberg et al., 2017; cf. Gorka et al., 2013; Gorka & Shankman, 2017). By ruling out the contribution of other stressor characteristics such as controllability, studies such as this one can implicate uncertain stressor reactivity and its associated neural substrates—rather than exaggerated reactivity broadly or other characteristics with distinct neural substrates—as key in at least the first stage of these neuroadaptations. Of course, more convincing evidence of these putative neuroadaptations requires a longitudinal approach, which could be accomplished in future research by assessing individuals before and after they have substantial exposure to alcohol and other drugs.

The current research joins a growing literature that uses modern experimental methods to assess responses to multiple stressor characteristics to inform understanding of stressor reactivity more broadly. Alcohol’s selective effects on uncertain stressor reactivity in humans are consistent with data from translational neuroscience research suggesting reactions to uncertain and certain stressors rely on distinct mechanisms (Davis et al., 2010). Likewise, the finding that alcohol’s effects vary for responses to uncertain stressors compared with certain stressors but not uncontrollable stressors compared with controllable stressors is consistent with research suggesting the mechanisms governing responses to stressor certainty are at least somewhat distinct from those governing responses to stressor controllability (Maier, 2015; Wood et al., 2015). The neurocircuits and neurotransmitter systems implicated in this research can be further clarified by more direct measurement and/or manipulation using neuroimaging techniques (e.g., functional MRI, magnetoencephalography) coupled with selective neurotransmitter agonists and antagonists and/or methods such as transcranial magnetic stimulation.

Whereas modern laboratory research on stressor reactivity tends to focus on one response domain, the current results demonstrate the importance of using a multimeasure approach to fully characterize stressor reactivity. In this study, subjective anxiety was greater during presentation of uncontrollable stressors compared with controllable stressors, as expected. We observed, unexpectedly, the opposite pattern for startle potentiation. The initial perceived risk of receiving higher intensity shocks (before control was enacted) may have activated neural structures responsible for increased defensive reactivity in the controllable condition at time of startle measurement but before any downstream effects of control. Thus, the use of multiple measures of different response domains raises interesting and testable questions about the timing of the neural and psychological mechanisms involved with the processing of stressor controllability. Furthermore, although startle potentiation has long been used to index increased defensive reactivity thought to reflect an increase in emotions such as anxiety, startle is also heavily influenced by attention (Blumenthal, 2015) because startle is lower when attention is focused away from the startle stimulus. Although this raises questions about the impact of attention, we could address this by including measurement of probe P3 suppression in the current study. Probe P3 suppression did not differ between controllable and uncontrollable stressors, which diminishes the likelihood that the results for startle potentiation were due to changes in attention.

Controllable and uncontrollable stressors appeared to demand comparable attentional resources given that each captured attention above the no-shock cues as measured by probe P3 suppression in the no-alcohol group. This is consistent with recent observations by our laboratory (Bradford et al., 2017) and others (Nelson et al., 2015) of comparable emotionally motivated attention to uncertain and certain stressors. Although some research has suggested that both uncertain and uncontrollable stressors have detrimental impacts on cognitive performance during complex cognitive tasks (Henderson et al., 2012; Robinson et al., 2013), our data suggest that basic, early attentional processes needed to appraise stressors initially in real time remain intact across multiple stressor characteristics including controllability, at least in sober individuals.

Despite differences in self-reported anxiety and defensive reactivity for uncontrollable stressors compared with controllable stressors in the no-alcohol group, alcohol did not appear to affect responses to these stressor types differently for any of the response domains studied. This may potentially diminish the apparent importance of stressor controllability relative to certainty, but future research should affirm this by clarifying remaining ambiguities and increasing the generalizability of the current results. For example, although the Alcohol × Certainty interaction was significant for self-reported anxiety and startle potentiation, the Alcohol × Certainty × Controllability interaction was statistically significant only for startle potentiation. Furthermore, although alcohol had a significantly bigger effect on probe P3 suppression during uncertain stressors compared with certain stressors in one past study (Bradford et al., 2017), this effect was only descriptively similar here. Thus, future research should continue to examine the robustness and replicability of alcohol’s effects on self-reported anxiety, startle potentiation, and emotionally motivated attention to various stressor characteristics with additional indices of emotion and attention (e.g., more granular self-report; eye-tracking methods), which aid in interpretation of the results.

Other important contributors to the relationship between drug use and response to stressors were beyond the scope of the current study. For example, theories such as the appraisal-disruption model (Sayette, 1993) recognize that drinking often occurs following a stressor rather than before it, as was the case in the current study. The appraisal-disruption model and other theories make specific predictions about the relative timing of stressor appraisal, exposure, and drinking. All of these components should be carefully manipulated and tested in future research using paradigms similar to here. Future research to clarify these and other ambiguities should also strive to include as representative of a sample as possible in terms of broad demographics and clinically relevant characteristics such as drinking history.

Additional research can also extend and bridge the current results to more ecologically valid and clinically relevant contexts. The use of electric-shock stressors in the current study is consistent with our program of research focused on careful translation of animal neuroscience that includes precise testing of stressor features to allow speculation about neural mechanism (Kaye et al., 2017). Future research using similar paradigms with different stressors, however, should further extend the current results to other translational paradigms from animal neuroscience (e.g., light/dark enhanced startle) and increase the generalizability of the results with stressors more consistent with common human experiences that include uncertainty (e.g., the Trier Social Stress Test). In fact, growing evidence suggests alcohol reduces negative affect more readily in social contexts when uncertain behavior is exhibited by others (e.g., “social-attribution approach”; Fairbairn & Sayette, 2014), and recent ecological momentary assessment research (EMA) supports the theory that these effects of alcohol may reinforce drinking in social contexts (e.g., Goodman et al., 2018). The current paradigm can be modified to include the presence of drinking partners or other social interactions to assess the contribution of social factors on alcohol stressor certainty effects, which was not addressed here. Furthermore, generalizability of the current results can be tested using EMA and mobile psychophysiology to examine responses to uncertain stressors and the effects of drug use in the real world.

More broadly, this study parses the mechanisms involved with two frequently confounded stressor dimensions assessed in various response domains using a novel, multiple-measurement paradigm and a powerful pharmacological manipulation. Similar approaches could be used for contemporary efforts to leverage psychophysiological markers toward clearer classification and/or diagnosis across psychopathology. Because of the perceived relevance of exaggerated response to uncertain stressors in anxiety disorders, the National Institute of Health’s Research Domain Criteria (RDoC) prominently features distinctions between certain and uncertain stressors, labeled respectively as “acute fear” and “potential threat (‘anxiety’)” in the Negative Valence Systems domain (Insel et al., 2010). Current RDoC criteria, however, do not currently speak to the controllability of those uncertain and certain stressors. Furthermore, RDoC does not yet include distinctions between uncontrollable and controllable stressors despite their potential relevance in the etiology of depression (Abramson et al., 1978; Maier & Watkins, 2005). Paradigms such as described here can be implemented to test for patient differences in reactions to stressor controllability and/or to rule out potential contributions of stressor controllability when differences in reactions to stressor certainty are of primary focus.

Supplemental Material

sj-pdf-1-cpx-10.1177_21677026211061355 – Supplemental material for Alcohol’s Effects During Uncertain and Uncontrollable Stressors in the Laboratory

Supplemental material, sj-pdf-1-cpx-10.1177_21677026211061355 for Alcohol’s Effects During Uncertain and Uncontrollable Stressors in the Laboratory by Daniel E. Bradford, Jack M. Shireman, Sarah J. Sant’Ana, Gaylen E. Fronk, Susan E. Schneck and John J. Curtin in Clinical Psychological Science

Footnotes

Acknowledgements

We thank Angelica DeFalco for drafting Figures 1 and 2 during the revision process.

Transparency

Action Editor: Tamika C. Zapolski

Editor: Jennifer L. Tackett

Author Contributions

D. E. Bradford, J. M. Sireman, and S. E. Schneck developed the study concept and design under the supervision of J. J. Curtin. Testing and data collection were performed by D. E. Bradford, J. M. Shireman and S. J. Sant’Ana. D. E. Bradford performed the data analysis and interpretation under the supervision of J. J. Curtin. D. E. Bradford drafted the manuscript, and D. E. Bradford, J. M. Shireman, S. E. Schneck, and G. E. Fronk provided critical revisions. All of the authors approved the final manuscript for submission.

ORCID iDs

Daniel E. Bradford

Susan E. Schneck

John J. Curtin

Supplemental Material

Additional supporting information can be found at

Notes

References

Abramson

L. Y.

Seligman

M. E.

Teasdale

J. D.

(1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87(1), 49–74. https://doi.org/10.1037/0021-843X.87.1.49

Alius

M. G.

Pané-Farré

C. A.

Löw

Hamm

A. O.

(2015). Modulation of the blink reflex and P3 component of the startle response during an interoceptive challenge. Psychophysiology, 52(1), 140–148. https://doi.org/10.1111/psyp.12295

Blanchard

D. C.

Griebel

Pobbe

Blanchard

R. J.

(2011). Risk assessment as an evolved threat detection and analysis process. Neuroscience & Biobehavioral Reviews, 35(4), 991–998. https://doi.org/10.1016/j.neubiorev.2010.10.016

Blumenthal

T. D.

(2015). Presidential address 2014: The more-or-less interrupting effects of the startle response. Psychophysiology, 52(11), 1417–1431. https://doi.org/10.1111/psyp.12506

Blumenthal

T. D.

Cuthbert

B. N.

Filion

D. L.

Hackley

Lipp

O. V.

van Boxtel

(2005). Committee report: Guidelines for human startle eyeblink electromyographic studies. Psychophysiology, 42(1), 1–15. https://doi.org/10.1111/j.1469-8986.2005.00271.x

Bradford

D. E.

Magruder

K. P.

Korhumel

R. A.

Curtin

J. J.

(2014). Using the threat probability task to assess anxiety and fear during uncertain and certain threat. JoVE Journal, 91, Article 51095. https://doi.org/10.3791/51905

Bradford

D. E.

Motschman

C. A.

Starr

M. J.

Curtin

J. J.

(2017). Alcohol’s effects on emotionally motivated attention, defensive reactivity, and subjective anxiety during uncertain threats. Social, Cognitive, and Affective Neuroscience, 12(11), 1823–1832. https://doi.org/10.1093/scan/nsx095

Bradford

D. E.

Shapiro

B. L.

Curtin

J. J.

(2013). How bad could it be? Alcohol dampens stress responses to threat of uncertain intensity. Psychological Science, 24(12), 2541–2549. https://doi.org/10.1177/0956797613499923

Bradford

D. E.

Starr

M. J.

Shackman

A. J.

Curtin

J. J.

(2015). Empirically based comparisons of the reliability and validity of common quantification approaches for eyeblink startle potentiation in humans. Psychophysiology, 52(12), 1669–1681. https://doi.org/10.1111/psyp.12545

10.

Bradley

M. M.

Codispoti

Lang

P. J.

(2006). A multi-process account of startle modulation during affective perception. Psychophysiology, 43(5), 486–497. https://doi.org/10.1111/j.1469-8986.2006.00412.x

11.

Brainard

D. H.

(1997). The psychophysics toolbox. Spatial Vision, 10(4), 433–436. https://doi.org/10.1163/156856897X00357

12.

Champely

(2020). Pwr: Basic functions for power analysis (Version 1.3-0) [Computer software]. Comprehensive R Archive Network. https://CRAN.R-project.org/package=pwr

13.

Curtin

J. J.

(2011). PhysBox: The psychophysiology toolbox. An open source toolbox for psychophysiological data reduction within EEGLab. http://dionysus.psych.wisc.edu/PhysBox.htm

14.

Curtin

J. J.

(2015). lmSupport: Support for linear models (Version 2.9.2) [Computer software]. Comprehensive R Archive Network. https://cran.r-project.org/web/packages/lmSupport/index.html

15.

Davis

Walker

D. L.

Miles

Grillon

(2010). Phasic vs sustained fear in rats and humans: Role of the extended amygdala in fear vs anxiety. Neuropsychopharmacology Reviews, 35(1), 105–135. https://doi.org/doi:10.1037/npp.2009

16.

Delorme

Makeig

(2004). EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134(1), 9–21. https://doi.org/10.1016/j.jneumeth.2003.10.009

17.

Fairbairn

C. E.

Sayette

M. A.

(2014). A social-attributional analysis of alcohol response. Psychological Bulletin, 140(5), 1361–1382. https://doi.org/10.1037/a0037563

18.

Goodman

F. R.

Stiksma

M. C.

Kashdan

T. B.

(2018). Social anxiety and the quality of everyday social interactions: The moderating influence of alcohol consumption. Behavior Therapy, 49(3), 373–387. https://doi.org/10/gdkhng

19.

Gorka

S. M.

Nelson

B. D.

Shankman

S. A.

(2013). Startle response to unpredictable threat in comorbid panic disorder and alcohol dependence. Drug and Alcohol Dependence, 132(1–2), 216–222. https://doi.org/10.1016/j.drugalcdep.2013.02.003

20.

Gorka

S. M.

Shankman

S. A.

(2017). Preliminary evidence that reactivity to uncertain threat is an endophenotype for alcohol use disorder. Drug & Alcohol Dependence, 180, 265–271. https://doi.org/10.1016/j.drugalcdep.2017.08.023

21.

Greeley

Oei

(1999). Alcohol and tension reduction. In Leonard

K. E.

Blane

H. T.

(Eds.), Psychological theories of drinking and alcoholism (2nd ed., pp. 14–53). Guilford Press.

22.

Grillon

(2008). Models and mechanisms of anxiety: Evidence from startle studies. Psychopharmacology, 199(3), 421–437. https://doi.org/10.1007/s00213-007-1019-1

23.

Grillon

Baas

J. M.

Pine

D. S.

Lissek

Lawley

Ellis

Levine

(2006). The benzodiazepine alprazolam dissociates contextual fear from cued fear in humans as assessed by fear-potentiated startle. Biological Psychiatry, 60(7), 760–766. https://doi.org/10/1016/j.biopsych.2005.11.027

24.

Grillon

Chavis

Covington

M. F.

Pine

D. S.

(2009). Two-week treatment with the selective serotonin reuptake inhibitor citalopram reduces contextual anxiety but not cued fear in healthy volunteers: A fear-potentiated startle study. Neuropsychopharmacology, 34(4), 964–971. https://doi.org/10.1038/npp.2008.141

25.

Grillon

Lissek

Rabin

McDowell

Dvir

Pine

D. S.

(2008). Increased anxiety during anticipation of unpredictable but not predictable aversive stimuli as a psychophysiologic marker of panic disorder. American Journal of Psychiatry, 165(7), 898–904. https://doi.org/10.1176/appi.ajp.2007.07101581

26.

Grillon

Pine

D. S.

Lissek

Rabin

Bonne

Vythilingam

(2009). Increased anxiety during anticipation of unpredictable aversive stimuli in posttraumatic stress disorder but not in generalized anxiety disorder. Biological Psychiatry, 66(1), 47–53. https://doi.org/10.1016/j.biopsych.2008.12.028

27.

Hanson

D. J.

(2013). Historical evolution of alcohol consumption in society. In Boyle

Boffetta

Lowenfels

A. B.

Burns

Brawley

Zatonski

Rehm

(Eds.), Alcohol: Science, policy and public health (pp. 3–12). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199655786.003.0001

28.

Havranek

M. M.

Bolliger

Roos

Pryce

C. R.

Quednow

B. B.

Seifritz

(2016). Uncontrollable and unpredictable stress interacts with subclinical depression and anxiety scores in determining anxiety response. Stress, 19(1), 53–62. https://doi.org/10.3109/10253890.2015.1117449

29.

Hefner

K. R.

Curtin

J. J.

(2012). Alcohol stress response dampening: Selective reduction of anxiety in the face of uncertain threat. Journal of Psychopharmacology, 26(2), 232–244. https://doi.org/10.1177/0269881111416691

30.

Hefner

K. R.

Moberg

C. A.

Hachiya

L. Y.

Curtin

J. J.

(2013). Alcohol stress response dampening during imminent versus distal, uncertain threat. Journal of Abnormal Psychology, 122(3), 756–769. https://doi.org/10.1037/a0033407

31.

Hefner

K. R.

Starr

M. J.

Curtin

J. J.

(2018). Heavy marijuana use but not deprivation is associated with increased stressor reactivity. Journal of Abnormal Psychology, 127(4), 348–358. https://doi.org/10.1037/abn0000344

32.

Henderson

R. K.

Snyder

H. R.

Gupta

Banich

M. T.

(2012). When does stress help or harm? The effects of stress controllability and subjective stress response on stroop performance. Frontiers in Psychology, 3, Article 179. https://doi.org/10.3389/fpsyg.2012.00179

33.

Hogle

J. M.

Kaye

J. T.

Curtin

J. J.

(2010). Nicotine withdrawal increases threat-induced anxiety but not fear: Neuroadaptation in human addiction. Biological Psychiatry, 68(8), 687–688. https://doi.org/10.1016/j.biopsych.2010.06.003

34.

Hurlbut

S. C.

Sher

K. J.

(1992). Assessing alcohol problems in college students. Journal of American College Health, 41(2), 49–58. https://doi.org/10.1080/07448481.1992.10392818

35.

Insel

Cuthbert

Garvey

Heinssen

Pine

Quinn

Sanislow

Wang

(2010). Research domain criteria (RDoC): Toward a new classification framework for research on mental disorders. The American Journal of Psychiatry, 167(7), 748–751. https://doi.org/10.1176/appi.ajp.2010.09091379

36.

Kaye

J. T.

Bradford

D. E.

Magruder

K. P.

Curtin

J. J.

(2017). Probing for neuroadaptations to unpredictable stressors in addiction: Translational methods and emerging evidence. Journal of Studies on Alcohol and Drugs, 78(3), 353–371. https://doi.org/10.15288/jsad.2017.78.353

37.

Koob

G. F.

Volkow

N. D.

(2010). Neurocircuitry of addiction. Neuropsychopharmacology Reviews, 35(1), 217–238. https://doi.org/10.1038/npp.2009.110

38.

Koolhaas

J. M.

Bartolomucci

Buwalda

de Boer

S. F.

Flügge

Korte

S. M.

Meerlo

Murison

Olivier

Palanza

Richter-Levin

Sgoifo

Steimer

Stiedl

van Dijk

Wöhr

Fuchs

(2011). Stress revisited: A critical evaluation of the stress concept. Neuroscience & Biobehavioral Reviews, 35(5), 1291–1301. https://doi.org/10.1016/j.neubiorev.2011.02.003

39.

Linden

D. E. J.

(2005). The p300: Where in the brain is it produced and what does it tell us? The Neuroscientist, 11(6), 563–576. https://doi.org/10.1177/1073858405280524

40.

Liu

R. T.

Kleiman

E. M.

Nestor

B. A.

Cheek

S. M.

(2015). The hopelessness theory of depression: A quarter century in review. Clinical Psychology, 22(4), 345–365. https://doi.org/10/gg887w

41.

Löw

Weymar

Hamm

A. O.

(2015). When threat is near, get out of here: Dynamics of defensive behavior during freezing and active avoidance. Psychological Science, 26(11), 1706–1716. https://doi.org/10.1177/0956797615597332

42.

Maier

S. F.

(2015). Behavioral control blunts reactions to contemporaneous and future adverse events: Medial prefrontal cortex plasticity and a corticostriatal network. Neurobiology of Stress, 1, 12–22. https://doi.org/10.1016/j.ynstr.2014.09.003

43.

Maier

S. F.

Watkins

L. R.

(2005). Stressor controllability and learned helplessness: The roles of the dorsal raphe nucleus, serotonin, and corticotropin-releasing factor. Neuroscience and Biobehavioral Reviews, 29(4–5), 829–841. https://doi.org/10.1016/j.neubiorev.2005.03.021

44.

Mineka

Kihlstrom

J. F.

(1978). Unpredictable and uncontrollable events: A new perspective on experimental neurosis. Journal of Abnormal Psychology, 87(2), 256–271. https://doi.org/10.1037/0021-843X.87.2.256

45.

Moberg

C. A.

Bradford

D. E.

Kaye

J. T.

Curtin

J. J.

(2017). Increased startle potentiation to unpredictable stressors in alcohol dependence: Possible stress neuroadaptation in humans. Journal of Abnormal Psychology, 126(4), 441–453. https://doi.org/10.1037/abn0000265

46.

Moberg

C. A.

Curtin

J. J.

(2009). Alcohol selectively reduces anxiety but not fear: Startle response during unpredictable vs. predictable threat. Journal of Abnormal Psychology, 118(2), 335–347. https://doi.org/10.1037/a0015636

47.

Moberg

C. A.

Weber

Curtin

J. J.

(2011). Alcohol dose effects on stress response to cued threat vary by threat intensity. Psychopharmacology, 218, 217–227. https://doi.org/10.1007/s00213-011-2304-6

48.

Nelson

B. D.

Hajcak

(2017). Defensive motivation and attention in anticipation of different types of predictable and unpredictable threat: A startle and event-related potential investigation. Psychophysiology, 54(8), 1180–1194. https://doi.org/10.1111/psyp.12869

49.

Nelson

B. D.

Hajcak

Shankman

S. A.

(2015). Event-related potentials to acoustic startle probes during the anticipation of predictable and unpredictable threat. Psychophysiology, 52(7), 887–894. https://doi.org/10.1111/psyp.12418

50.

Pelli

D. G.

(1997). The VideoToolbox software for visual psychophysics: Transforming numbers into movies. Spatial Vision, 10(4), 437–442. https://doi.org/10.1163/156856897X00366

51.

Picton

T. W.

Bentin

Berg

Donchin

Hillyard

S. A.

Johnson

Miller

G. A.

Ritter

Ruchkin

D. S.

Rugg

M. D.

Taylor

M. J.

(2000). Guidelines for using human event-related potentials to study cognition: Recording standards and publication criteria. Psychophysiology, 37(2), 127–152. https://doi.org/10.1111/1469-8986.3720127

52.

R Core Team. (2015). R: A language and environment for statistical computing (Version 4.0.0) [Computer software]. R Foundation for Statistical Computing. http://www.R-project.org

53.

Robinson

O. J.

Bond

R. L.

Roiser

J. P.

(2015). The impact of threat of shock on the framing effect and temporal discounting: Executive functions unperturbed by acute stress? Frontiers in Psychology, 6, Article 1315. https://doi.org/10.3389/fpsyg.2015.01315

54.

Robinson

O. J.

Vytal

Cornwell

B. R.

Grillon

(2013). The impact of anxiety upon cognition: Perspectives from human threat of shock studies. Frontiers in Human Neuroscience, 7, Article 203. https://doi.org/10.3389/fnhum.2013.0020

55.

RStudio Team. (2016). RStudio: Integrated development for R. (Version 1.2) [Computer software]. RStudio. http://www.rstudio.com/

56.

Sayette

M. A.

(1993). An appraisal-disruption model of alcohol’s effects on stress responses in social drinkers. Psychological Bulletin, 114(3), 459–476. https://doi.org/10.1037/0033-2909.114.3.459

57.

Sayette

M. A.

(2017). The effects of alcohol on emotion in social drinkers. Behaviour Research and Therapy, 88, 76–89. https://doi.org/10.1016/j.brat.2016.06.005

58.

Schönbrodt

F. D.

Maier

Heene

Zehetleitner

(2015). Voluntary commitment to research transparency. http://www.researchtransparency.org

59.

Sher

K. J.

(1987). Stress response dampening. In Blane

H. T.

Leonard

K. E.

(Eds.), Psychological theories of drinking and alcoholism (pp. 227–271). Guilford Press.

60.

Simmons

J. P.

Nelson

L. D.

Simonsohn

(2012). A 21 word solution. SSRN. https://doi.org/10.2139/ssrn.2160588

61.

Testa

Fillmore

M. T.

Norris

Abbey

Curtin

J. J.

Leonard

K. E.

Mariano

K. A.

Thomas

M. C.

Nomensen

K. J.

George

W. H.

VanZile-Tamsen

Livingston

J. A.

Saenz

Buck

P. O.

Zawacki

Parkhill

M. R.

Jacques

A. J.

Hayman

L. W.

Jr. (2006). Understanding alcohol expectancy effects: Revisiting the placebo condition. Alcoholism: Clinical and Experimental Research, 30(2), 339–348. https://doi.org/10.1111/j.1530-0277.2006.00039.x

62.

Wendt

Löw

Weymar

Lotze

Hamm

A. O.

(2017). Active avoidance and attentive freezing in the face of approaching threat. NeuroImage, 158, 196–204. https://doi.org/10.1016/j.neuroimage.2017.06.054

63.

Wood

K. H.

Wheelock

M. D.

Shumen

J. R.

Bowen

K. H.

Ver Hoef

L. W.

Knight

D. C.

(2015). Controllability modulates the neural response to predictable but not unpredictable threat in humans. NeuroImage, 119, 371–381. https://doi.org/10.1016/j.neuroimage.2015.06.086

64.

Zvolensky

M. J.

Lejuez

C. W.

Eifert

G. H.

(2000). Prediction and control: Operational definitions for the experimental analysis of anxiety. Behaviour Research and Therapy, 38(7), 653–663. https://doi.org/10.1016/S0005-7967(99)00090-X

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

0.16 MB