Drunk driving is one of the main causes of car accidents. Often, drunk drivers do not realize that they are impaired, believing that they 'only had 2 drinks' or that they 'feel fine'. Five drivers were selected, and their driving reaction times in an obstacle course were measured after having consumed no alcoholic drinks and after consuming 2-3 alcoholic drinks. Using the observed sample data, which of the following statistical procedures would be most appropriate to test the claim that driving reaction times are affected by alcohol consumption? Assume that any necessary normality requirements hold.
a. Two-tailed (non-directional), two-sample paired/dependent t-test of means
b. Two-tailed (non-directional), two-sample independent t-test of means