Alcohol and vigilance performance: a review.

In the literature on the effects of alcohol on driving-related skills, it is sometimes claimed that vigilance tasks are insensitive instruments whereas divided-attention tasks are extremely sensitive to the effects of alcohol. The results of the present review, based on the analysis of 38 comparisons of alcohol and placebo in vigilance tasks, require that these claims be restated. Both types of attentional task (concentrated and divided) are indispensable in test batteries, although not all types of vigilance and divided-attention task are equally sensitive, e.g. some types of vigilance task, using spatial stimuli, were sensitive to BAC levels of 0.03% whereas other types were insensitive to levels of 0.10%. In contrast, the usefulness of tasks of questionable validity and/or low sensitivity (such as the DSST, CFF, digit span, simple RT and choice RT) is questioned. Apart from issues of validity and sensitivity of tests, the ways in which alcohol may affect performance are also discussed. The main effect of moderate doses of alcohol is on attention and information processing. The capacity to divide and sustain attention is already impaired at BAC levels of 0.02-0.03%. Further, alcohol effects appear to some extent to be time-dependent, and are greatest during periods of sleepiness (the early afternoon and after midnight). Some current BAC levels concerning drinking and driving are far too generous. There is sufficient evidence from the literature on performance indicating that the BAC standard for driving should be lowered to 0.02% for driving after midnight and for special risk groups (young and less experienced drivers).