The effects of caffeine-induced arousal on global versus local object focus were investigated in non-habitual consumers using a double-blind, within-subjects, repeated-measures design. Following an overnight fast, low caffeine consumers (N=36; M=42.5 mg/day caffeine) completed 5 counterbalanced test sessions (normal consumption, 0 mg, 100 mg, 200 mg, and 400 mg) separated by at least 3 days. During each session, volunteers either consumed their normal amount of caffeine or were administered 1 of 4 treatment pills. One hour later they completed two tasks assessing visual attention, in counterbalanced order. Measures of mood, salivary caffeine and cortisol were taken at multiple time points. Dose-dependent elevation of caffeine in the saliva demonstrated the experimental manipulation was effective. Furthermore, analyses of the mood and arousal measures detected consistent changes on arousal subscales and caffeine administration elevated saliva cortisol. Analyses of the visual attention tasks revealed that caffeine-induced physiological arousal produced global processing biases, after as little as 100 mg caffeine. These data suggest caffeine consumption may influence how individuals attend to and process information in their environment and could influence daily tasks such as face recognition, learning new environments and navigation, especially for those who normally consume little caffeine.