THE EFFECTS OF SMOKING ON PERFORMANCE ON THE GARNER SPEEDED CLASSIFICATION TASK

The Stroop task has been used by several investigators to examine the effects of nicotine and smoking on human selective attention, but this research has produced inconclusive results. In this article a new tas k is described, the Garner speeded classification task, that can be us ed to explore the influences of nicotine on human selective attention in a more detailed fashion than has been reported previously. In a stu dy using this task reported here, 52 smokers performed the Garner task twice. Half the subjects smoked a cigarette between the first and sec ond completion of the task, and the remainder did not smoke. The main findings were that smoking reduced the size of Garner interference for both reaction time and error measures, and that smoking reduced the s ize of Stroop interference for the error measure but not the reaction time measure. The degree of nicotine deprivation of the subjects at te sting did not substantially affect this result. Moreover, there was a suggestion that the effect of smoking on Stroop interference was secon dary to the effect on Garner interference, indicating that smoking, an d thus presumably nicotine, principally attenuates the disruptive infl uence of task-irrelevant, but varying, dimensions in selection. (C) 19 98 John Wiley & Sons, Ltd.