EVIDENCE LIBRARY
Can a Cognitive Score Make an Older Driver Equivalent to an Intoxicated Young Driver?
MedDrive responded to alcohol in young adults, but older adults could have lower scores while driving well. Equal cognitive scores did not establish equal driving impairment.
No. Similar cognitive-test scores do not establish equivalent driving impairment from aging and alcohol. The MedDrive development study found that alcohol reduced scores in young adults, yet many older adults who drove well had scores below those of the alcohol-exposed young adults. The context and timing of a change mattered.
For DUI interpretation, this cautions against translating an age-related score into an “equivalent BAC,” or treating an alcohol-sensitive test as a universal fitness-to-drive threshold.
The paper combined several different studies
Vaucher and colleagues developed MedDrive to assess visual processing and cognitive functions associated with aging. Its tasks examined visual recognition, attention shifting, movement detection with distractors and spatial memory.
The project included an initial development sample, a separate validation sample of 182 older drivers, a 17-person repeated-testing study, a 47-person psychophysical study and a 20-person alcohol experiment. These were different samples answering different questions. The paper did not administer alcohol to all 182 older drivers.
In the older-driver validation study, instructors assessed on-road performance without knowing the cognitive results. The alcohol experiment instead studied healthy adults aged 23–40 using a simulator.
Alcohol calibration showed responsiveness
The alcohol study was a randomized, double-blind, placebo-controlled crossover with nominal targets of 0, 0.5, 0.65 and 0.8 g/L. Participants had practiced MedDrive five times before experimental testing.
Alcohol was monitored using breath instruments and maintained with repeated drinks. The paper expresses exposure as BAC, but these were breath-derived values, not independently assayed blood concentrations. Mean achieved values were approximately 0.49, 0.64 and 0.78 g/L in the active conditions.
At the 0.5 g/L target, the overall score was approximately 5.3 points lower than placebo (p = 0.008). Fifteen of 20 participants showed a decrease. All task outcomes were significantly influenced at the highest target, but not every component showed a significant change at the lowest target.
The study also found increased simulator lane-position variability with alcohol. Changes in MedDrive and simulator performance were related in the authors’ models. That establishes a within-experiment relationship, not a validated BAC estimator or a rule for classifying every new driver.
The older-driver findings prevent a simple equivalence claim
In the separate older-driver sample, the composite score’s association with on-road performance was statistically significant but weak: R² was approximately 0.053. That describes roughly 5.3% of variation in the analyzed driving score, not “5.3% accuracy.”
The authors observed that more than half of the very good older drivers scored below any of the young participants in the highest alcohol condition. Thus, a lower cognitive score could not be assigned the same behavioral meaning across the two settings.
The authors proposed that gradual changes allow adaptation, whereas abrupt alcohol effects create a different situation. That mechanism is a plausible interpretation, not something the study directly established through longitudinal brain imaging or a lifetime experiment.
Why a threshold would mislead
An explored score threshold in the older-driver sample had limited screening performance. Its estimated association with poor driving was imprecise and not statistically significant. The investigators warned that using the score alone would produce many incorrect fitness judgments.
A composite score also depends on its component tasks, weighting and prior practice. In the reliability study, scores improved across repeated sessions. Reliability and a learning effect can coexist: a test may consistently distinguish participants while their performance changes with familiarity.
The site’s analysis of laboratory alcohol calibration explains why detecting an alcohol effect and predicting driving are separate validation questions.
Limits and case-specific use
The aging comparisons were cross-sectional, participants volunteered, and some older-driver measures required imputation. The alcohol sample was small and young. Blinding was incomplete, and alcohol increased simulator-sickness symptoms, although the investigators examined that potential influence.
The study does not establish that aging resembles intoxication, that an older person’s low score proves alcohol exposure, or that a particular score establishes safety. It supports evaluating test results in their population, task and temporal context.
Actual DUI evidence still requires the measured matrix or breath result, timing and observed driving. For another example of why age and task outcome need separate treatment, see the site’s older-versus-younger recognition-memory experiment.
Primary source
Vaucher P, Cardoso I, Veldstra JL, Herzig D, Herzog M, Mangin P, Favrat B. A neuropsychological instrument measuring age-related cerebral decline in older drivers: development, reliability, and validity of MedDrive. Frontiers in Human Neuroscience. 2014;8:772. https://doi.org/10.3389/fnhum.2014.00772.