Errors in performance testing: a comparison of ethanol and temazepam

Both ethanol and benzodiazepines impair psychomotor function. Previous work has suggested that ethanol may have a greater effect on errors while benzodiazepines may cause greater slowing, but this has not been tested in a direct comparison. We assessed the effects of ethanol, at blood concentrations of approximately 80-100 mg/100 ml, compared to two doses of temazepam (20 mg and 30 mg) on psychomotor speed and accuracy and on long-term memory. Sixteen healthy volunteers (eight male, aged 20-25 years) took part in a four-period, placebo-controlled cross-over study. Performance was evaluated using analysis of covariance (critical significance level, p = 0.05) comparing the areas under the response-time curves. Performance on a psychomotor maze showed an almost complete dissociation, with ethanol leading to a substantial and significant increase in errors with little effect on speed, while temazepam slowed performance with no significant change in accuracy. Other tasks showed a similar pattern, but the dissociation was less complete. Handwriting size was substantially increased by ethanol, but not by temazepam. Information processing capacity and long-term memory formation were reduced by a similar amount both for ethanol and 30 mg temazepam. The faster, more error-prone, behaviour on ethanol than with a similarly impairing dose of temazepam has clear implications for the relative potential of the two drugs to contribute to accidents. The results are also important in understanding the differential effects of drugs with different mechanisms of action on human performance.