In a linear regression of SS>GS against SSRT, we identified the correlates of response inhibition for drinkers (regression I). To examine the effects of duration of use and drinking problems on response inhibition, we regressed SS>GS against the AUDIT score and years of alcohol use in drinkers in a multiple regression (regression II). Thus, the latter model identified correlates of response inhibition each in association with drinking problems and duration of use and, by comparison with the linear regression of SS>GS against SSRT (regression I), allowed us to identify regional activities specific to response inhibition and modulated by the AUDIT score and years of alcohol use. Because the duration of drinking was significantly correlated with age (see Results), we regressed SS>GS against age to localize age-related changes in non-drinkers (regression III) and to specify regional activations to response inhibition as influenced by years of drinking (regression II) but not age.
