Alcohol, Task Difficulty, and Incentives in Drill Press Operation

Abstract

The effects of alcohol, task difficulty, and incentives on a drill press simulation were examined. On four nonconsecutive days the subject received four alcohol levels, one level per day. The levels were: 0.00%, 0.03%, 0.06%, and 0.09% blood alcohol concentration (BAC). The drill press simulation task consisted of placing a drill probe alternatively into two holes in a metal plate. For every level of alcohol, each subject received eight levels of task difficulty which were varied by changing the hole size and distance between holes. Task difficulty was in units of bits of information ranging from two to nine and was derived from a portion of Welford's (1968) version of Fitts' Law. For every level of task difficulty, each subject received two levels of incentive: productivity and accuracy which were enforced by monetary rewards. The subject's response was measured by hits and misses. A hit was defined as fully inserting the metal drill probe in either of two holes in the top metal plate of the workpiece without touching the plate. A miss was defined as a contact of the probe with the top plate. Alcohol had a significant effect on hits ( p = 0.0001) as evidenced by a 12% decrease in hits at the 0.06% BAC, and a 19% decrease at the 0.09% BAC. Task difficulty affected both hits and misses (p = 0.0001). The greatest decrease in hits between any two levels of task difficulty was an 89% decrease from the easiest task difficulty level to the most difficult level. The largest decrease between any two levels of difficulty in misses was 57% and occurred between the eight and five bit task difficutly level. Incentive had a significant effect on hits ( p = 0.0002) and misses ( p = 0.0001). Hits were 27.5% less in the productivity condition than in the accuracy condition. Misses were 74% less in the accuracy condition than in the productivity condition. Although the alcohol by task difficulty interaction was significant for hits ( p = 0.0001), and the incentive by task difficulty interactions were significant for both hits and misses ( p = 0.0001), plots of these effects were not subject to meaningful interpretation.