Anhedonia, the loss of pleasure from previously rewarding activities, is implicated in several neuropsychiatric conditions, including major depressive disorder (MDD). In order to accelerate drug development for mood disorders, quantitative approaches are needed to objectively measure responsiveness to reward as a means to identify deficits. One such approach, the probabilistic reward task (PRT), uses visual discrimination methodology to quantify reward learning. In this computerized task, humans make visual discriminations, and probabilistic contingencies are arranged such that correct responses to one alternative are rewarded more often (rich) than correct responses to the other (lean). Healthy participants consistently develop a response bias in favor of the rich alternative. However, participants with MDD typically exhibit lower response biases, and this blunting correlates with current and future anhedonia. The present studies validated a touchscreen-based PRT in rodents with formal and functional similarity to the human task. First, rats were trained to discriminate between two lines that differed in length. Next, parametric manipulations of probabilistic contingencies, line-length stimuli, and drug treatment (amphetamine, 0.32–3.2 mg/kg; scopolamine, 0.1–1.0 mg/kg; oxycodone, 0.1–1.0 mg/kg) on response bias were evaluated. Results demonstrated orderly shifts in bias and discriminability that varied as a function of, respectively, the asymmetry of rich/lean probabilities and disparity in line lengths. Drugs that enhance reward responsiveness (amphetamine and scopolamine, but not oxycodone) increased bias, verifying pharmacological task sensitivity. Finally, performance outcomes under optimized conditions were replicated in female rats. Collectively, the touchscreen-based rodent PRT appears to have high preclinical value as a quantitative assay of reward learning.