Can Normal Pointing Hide Alcohol-Impaired Stopping? An Automatic-Correction Experiment

Quick answer: Alcohol can impair a person’s ability to stop an automatic hand movement even when some measures of movement speed and rapid online correction show no statistically detectable deterioration. In a 2013 controlled laboratory study, six men assigned to a countermanding task made more prohibited corrective movements after alcohol than after a no-alcohol drink: 38.0% versus 18.2% of target-displacement trials (p = .045). A different group of six men assigned to make the corrections showed no statistically detectable alcohol effect on their correction rate. These are separate task outcomes, not proof that alcohol universally spares automatic movement or that any particular driver was impaired.

The primary evidence is Johnston, Timney, and Goodale’s peer-reviewed original experiment, “Acute Alcohol Consumption Impairs Controlled but Not Automatic Processes in a Psychophysical Pointing Paradigm” (2013). It offers a useful forensic distinction: a person may still initiate and execute a familiar motor response while having greater difficulty suppressing that response when instructions or changing circumstances require restraint.

What exactly did the researchers test?

Twelve healthy right-handed male social drinkers, ages 19–31, participated. Six were randomly assigned to an automatic-correction group and six to a countermanding group. Every participant completed both an alcohol session and a no-alcohol session on separate days, with session order counterbalanced within each group. The two task groups were not the same six people: the researchers did not test every person on both kinds of instruction.

On a tabletop apparatus, a participant pointed rapidly with the right index finger toward a light. On 20% of trials, the illuminated target moved left or right just as the participant began reaching. The visual event and hand movement were similar across groups, but the instruction changed the required response. Each experimental session included 300 trials, of which 60 involved a target displacement.

Condition A: correct the reach

In the correction group, the participant was told to redirect the reach toward the newly illuminated location. This tests the ability to make a rapid online adjustment, often described as an automatic “autopilot” correction. Successful correction was classified using a participant-specific confidence interval based on stationary-target movements, rather than simply asking an observer whether the movement looked normal.

Condition B: stop the correction

In the countermanding group, the participant was instructed to terminate the movement when the target jumped. The investigators counted an “action slip” when the hand continued far enough to reach the displaced target despite that stop instruction. Thus the primary outcome was not whether the person could move a finger. It was whether the person could withhold a prepotent corrective action when the experimental rule required stopping.

How was the alcohol exposure established?

The alcohol drink was vodka mixed with fruit juice. Investigators calculated each dose to target a nominal blood alcohol concentration (BAC) of 0.08%, but that target was a dose calculation, not a measured forensic blood result. Participants drank over approximately 20 minutes; the task began when a Dräger Alcotest 7410 breath instrument yielded an estimated BAC of approximately 0.06% on the rising limb of the alcohol curve. The authors did not take breath readings during the approximately 50-minute performance task.

This matters for courtroom interpretation. A breath-device estimate, a planned dose-derived BAC, and a directly analyzed venous whole-blood ethanol concentration are different forms of evidence. The publication does not document contemporaneous whole-blood headspace gas chromatography during each pointing trial. Nor does it establish an exact ethanol concentration associated with every action slip.

The no-alcohol beverage was fruit juice served similarly, but the authors deliberately described it as “no-alcohol” rather than a validated placebo: they did not collect subjective-response measurements establishing successful masking. Expectations and session awareness therefore cannot be assumed fully controlled.

What changed under alcohol? The principal results

The principal countermanding contrast is substantial within this six-person group: alcohol was associated with an approximately 19.8-percentage-point increase in prohibited automatic corrections. The finding does not mean that 38% of alcohol-consuming motorists will fail to brake, that every drinker has the same deficit, or that the difference is a validated case-specific probability of impairment.

The automatic-correction result is a different kind of result: the study did not detect a difference in successful redirection in the six people instructed to correct. Absence of a statistically significant effect in a small sample does not demonstrate equivalence, prove immunity to alcohol, or show that all reflexes remain intact.

Did ordinary movement speed or accuracy deteriorate?

Several measurements of movement execution showed no statistically significant alcohol-versus-no-alcohol difference, even as the countermanding outcome changed. That contrast is the main scientific contribution of the study. It shows why global judgments based on whether someone can move normally may miss a narrower deficit in response selection or suppression.

Movement onset did not meet the study’s conventional .05 significance threshold; it should not be described as proven identical across sessions. Hand velocity was obtained from infrared markers tracked at 100 Hz, not from a roadside balance test, a neurological examination, or a clinical assessment of intoxication. The experiment also detected increased endpoint variability under alcohol despite no detected change in average radial displacement, a reminder that variability and average accuracy are not interchangeable.

Why can an automatic correction survive while an instructed stop fails?

The authors interpreted the two tasks as placing different demands on visuomotor control. Rapid redirection toward a suddenly moving visual target can rely heavily on automated online adjustments. Countermanding requires the person to override or suppress a correction they would otherwise execute. Alcohol’s selective effect in this paradigm is therefore compatible with greater vulnerability of controlled inhibition than of automatic redirection.

That interpretation is functional, not an independent measurement of a specific neural lesion or neurotransmitter. The study used hand tracking and behavioral classification; it did not directly assay prefrontal cortical ethanol concentrations, record neural inhibition, or demonstrate which particular brain circuit failed in each trial. The authors discussed possible prefrontal contributions as a mechanistic explanation, not as experimentally proven individualized localization.

A person can show a relatively ordinary movement trajectory while making an inappropriate choice to continue rather than stop. Conversely, a failure to stop does not by itself identify alcohol as the cause in an uncontrolled individual event. Instruction comprehension, attention, fatigue, familiarity with the task, neurological or physical conditions, and measurement methods can all matter outside this tightly controlled experiment.

What does this mean for DUI or collision reconstruction?

In litigation, it is tempting to equate visible coordination with preserved ability to make a safe decision. The results illustrate why that inference is not generally warranted. The test of initiating a reach, the test of adjusting a reach, and the test of suppressing a reach assessed distinct components of behavior. A witness who appeared able to make smooth movements could still have had difficulty with a specific inhibitory demand—although this study cannot establish that the witness did.

Driving involves brakes, steering, sustained attention, hazard anticipation, shifting task priorities, and multiple kinds of learned responses. No participant in this study drove a vehicle or simulator. Investigators did not measure lane keeping, emergency braking, collision avoidance, crash frequency, Standardized Field Sobriety Tests, or actual traffic outcomes. Extrapolating an observed finger-pointing “action slip” directly to a driver’s collision causation would therefore exceed the experiment.

The practical evidentiary question is narrower: Does apparently preserved simple movement rule out every alcohol-sensitive component of skilled behavior? The study supports answering no. It does not supply a universal impairment threshold, a personal correction factor for breath alcohol, or a validated causal calculation for a particular accident.

Why breath-based exposure, behavioral impairment, and causation must remain separate

Four different types of claims should be distinguished. First is exposure measurement: the breath analyzer provided a device-dependent ethanol estimate. Second is calculated exposure: the dose selection relied on a model target of 0.08%, not a forensic blood assay. Third is observed laboratory performance: infrared-tracked hand movements and trial classifications measured success or failure under instructions. Fourth is legal or clinical inference: whether a specific person’s real-world act was impaired by alcohol and whether that impairment caused an event. The first three did not directly establish the fourth.

For context on how breath and blood evidence can differ, see the site’s breath alcohol measurement overview. For the distinction between measured concentrations and pharmacokinetic inferences, see pharmacokinetics in forensic toxicology. The forensic toxicology evidence library provides additional methodological context. These links are background navigation, not extra scientific source studies for the numerical findings above.

What are the study’s main limitations?

Very small, male-only sample. Only twelve healthy right-handed male social drinkers participated, and each task group contained six people. The results cannot by themselves be generalized to women, older adults, injured persons, people with neurological disease, or every pattern of intoxication.

Different participants performed the two instructions. Although each person completed alcohol and no-alcohol sessions, the researchers assigned individuals to one task group. The study did not demonstrate that each individual retained automatic correction while personally losing countermanding ability in a paired comparison of both tasks.

No validated masking or task-time blood profile. The comparison drink was not confirmed as a successful placebo using subjective-response measures. Breath estimates were obtained before, but not continuously during, the behavioral testing. A planned 0.08% dose target does not establish a measured 0.08% blood concentration at the time of any particular movement.

Restricted ecological validity. These were seated finger-pointing movements with artificial, predictable trial structure and pretraining. A significant laboratory group effect is not a sensitivity or specificity estimate for roadside testing, and a nondetected effect in another endpoint is not evidence of clinical equivalence or universal absence of impairment.

Measurement and statistical uncertainty. Some outcomes were based on endpoint-classification thresholds constructed separately within participants and sessions. The source reports group-level means and significance tests; it does not furnish an individual forensic decision rule. Associations and differences must be interpreted within the reported design, not converted into a probability that a named person caused a collision.

Frequently asked questions

Can alcohol impair stopping even when a person can point accurately?

In this experiment, yes: the countermanding group had a higher rate of action slips under alcohol even though the publication found no significant alcohol effect on several ordinary movement kinematics. That result is task-specific, based on six men in that group, and not an individual diagnosis.

Did the study show that automatic movements are unaffected by alcohol?

No. It found no statistically detectable difference in the correction rate in a small group performing one online redirection task. That does not prove that all automatic actions are safe or resistant to alcohol.

Were blood alcohol concentrations measured directly?

No contemporaneous forensic whole-blood ethanol analysis is reported. Researchers used a Dräger breath device to obtain estimated BAC values and used a separate calculation to choose the alcohol dose. No breath readings were recorded during the pointing task itself.

Does 38% action slips mean a 38% crash risk?

No. The 38% figure is the mean fraction of relevant laboratory target-displacement trials classified as prohibited corrective movements in the six-person countermanding group under alcohol. It is not a road-crash incidence, a risk ratio, or a forensic causation probability.

Why distinguish variability from average pointing accuracy?

The experiment reported greater spatial variability of movement endpoints under alcohol while average radial displacement was not significantly different. A person can vary more from trial to trial even when the average distance from a target does not show a detected shift.

Can this experiment establish whether one particular driver was impaired?

No. It demonstrates a limited group-level laboratory effect. Individual forensic interpretation requires case-specific specimen identity, analytical quality, sampling time, clinical and behavioral context, alternative explanations, and the relevant legal standard.

What primary publication supports the numerical results?

Johnston K, Timney B, Goodale MA. Acute Alcohol Consumption Impairs Controlled but Not Automatic Processes in a Psychophysical Pointing Paradigm. PLOS ONE. 2013;8(7):e68682. doi:10.1371/journal.pone.0068682. The original open-access paper supplies the experiment, analyzed outcomes, and acknowledged interpretive context.

Bottom line for interpreting apparently normal movements

Johnston and colleagues showed that visible execution and inhibitory control can diverge: alcohol increased prohibited corrections in one small task group without producing a matching pattern of significant changes in several simple movement metrics. The evidence can inform a scientifically careful discussion of preserved coordination versus impaired response suppression. It cannot replace direct case evidence or establish an individual BAC, a real-world impairment conclusion, or an event’s cause.