In the study we considered the ability of the relative speed of processing-automaticity (RSOP-A) and contextual disintegration (CD) models of the Stroop interference effect to account for the age-related increase in Stroop interference typically observed in older adults. Findings from the first experiment were partially consistent with predictions of the RSOP-A model because response dominance was greater for older adults than for younger adults. However, the age-related increase in interference was independent of this increase in response dominance, suggesting that factors other than those postulated in the RSOP-A model contributed to the greater interference observed in older adults. Results of the second experiment were consistent with the CD model, which suggests that older adults had difficulty maintaining a color-naming strategy to guide task performance.