EVIDENCE LIBRARY
Can a Harder Inhibition Task Show a Smaller Alcohol Effect?
A 20-man study found an alcohol-related rise in inhibition errors in easier task conditions but no significant increment in the hardest condition. Baseline difficulty and response preparation explain why this is not evidence that complex driving is safer.
Yes. In one experiment, alcohol increased inhibition errors in easier conditions while the hardest condition showed no significant added alcohol effect. That pattern does not mean difficult tasks became easy, that participants were unimpaired, or that adding workload protects driving. The sober baseline and the specific demands of the task are essential to interpretation.
What the study compared
Stock and colleagues analyzed behavioral and EEG data from 20 healthy men aged 19–32. Participants first completed a sober session and later an alcohol session. Most behavioral data had appeared in an earlier report; the 2022 paper added EEG analyses and included two additional participants previously excluded for lower response accuracy.
The alcohol session produced mean breath-derived BAC equivalents of approximately 1.01‰ before testing and 1.08‰ afterward. These were breath-instrument estimates, not laboratory blood measurements. Their increase also prevents assuming the entire session occurred on a declining concentration curve.
Participants saw rotated letters or numbers and either pressed a button or withheld a response. One block required a more complex distinction involving mirrored versus ordinary symbols; another used a simpler letter-versus-number rule. Rotation angle further changed the task demands.
The alcohol increment depended on the condition
In the simpler block, false alarms on no-go trials averaged 0.14% sober and 2.09% after alcohol. The alcohol effect was significant, p = 0.014, and supported by a nonparametric comparison.
In the more demanding block, alcohol increased errors for the smaller, 30-degree rotation: 8.89% sober versus 14.72% intoxicated. At the most demanding 150-degree rotation, the sober-versus-alcohol contrast was not significant, p = 0.413.
Thus, the null result applied to the hardest combination, not to every condition in the more demanding block. The paper also found a significant interaction among alcohol, block and rotation, supporting the need to specify the condition.
A smaller increment is not better absolute performance
Across sober and alcohol conditions, the more demanding block had substantially more errors than the simpler block. Therefore, absence of an additional detectable alcohol effect in the hardest condition did not establish superior performance or normal inhibitory ability.
The authors proposed that demanding response selection slowed or weakened automatic button-press tendencies. An action that is less strongly prepared may require less effort to stop, even while other processing demands are higher.
This is an explanation of the experimental balance between generating and withholding responses. It does not show that distracted driving is protective or that a busy road offsets alcohol effects.
The EEG findings concerned preparation as well as stopping
The clearest differences in the alcohol effect between easiest and hardest conditions occurred approximately 750–1,500 milliseconds after the stimulus. That timing was later than the usual early motor-inhibition window considered by the authors.
Theta- and alpha-band patterns were consistent with differences in working-memory engagement or preparation for upcoming stimuli. The study could not decisively separate these related explanations.
The EEG analysis used correctly performed no-go trials. Its brain-activity findings therefore were not direct recordings of the error mechanism on failed trials. Source locations were estimated from scalp EEG rather than directly measured with electrodes inside those brain regions.
Limits for a DUI opinion
The fixed sober-first session order and fixed block order leave practice, fatigue and sequence effects incompletely separated from alcohol. The all-male, small sample and single exposure range limit generalization. Reanalysis of overlapping participants is not an independent replication of the earlier behavioral result.
No driving task was performed. The study does not establish a personal impairment cutoff, predict a collision or show that passing one difficult test excludes deficits elsewhere. The site’s laboratory-to-driving calibration analysis addresses that additional inference. Its speech-listening study offers another example in which an alcohol effect depended on the endpoint tested.
Primary source
Stock A-K, Yu S, Ghin F, Beste C. How low working memory demands and reduced anticipatory attentional gating contribute to impaired inhibition during acute alcohol intoxication. Scientific Reports. 2022;12:2892. https://doi.org/10.1038/s41598-022-06517-9.