Previous research has suggested that infants are unable to make a corrective eye movement in response to a small base-out prism placed in front of one eye before 14-16 weeks . Three hypotheses have been proposed to explain this early inability, and each of these makes different predictions for the time of onset of a response to a larger prism. The first proposes that infants have a 'degraded sensory capacity' and so require a larger retinal disparity (difference in the position of the image on the retina of each eye) to stimulate disparity detectors . This predicts that infants might respond at an earlier age than previously reported  when tested using a larger prism. The second hypothesis proposes that infants learn to respond to larger retinal disparities through practice with small disparities . According to this theory, using a larger prism will not result in developmentally earlier responses, and may even delay the response. The third hypothesis proposes that the ability to respond to prismatic deviation depends on maturational factors indicated by the onset of stereopsis (the ability to detect depth in an image on the basis of retinal disparity cues only)  , predicting that the size of the prism is irrelevant. To differentiate between these hypotheses, we tested 192 infants ranging from 2 to 52 weeks of age using a larger prism. Results showed that 63% of infants of 5-8 weeks of age produced a corrective eye movement in response to placement of a prism in front of the eye when in the dark. Both the percentage of infants who produced a response, and the speed of the response, increased with age. These results suggest that infants can make corrective eye movements in response to large prismatic deviations before 14-16 weeks of age. This, in combination with other recent results , discounts previous hypotheses.