There is now considerable evidence that infants are surprised when shown impossible events that violate their beliefs about objects, as indexed by reliably greater looking times at these events than at possible events that are consistent with their beliefs. Do infants also attempt, as older children and adults do, to generate explanations for these impossible events? Data from our laboratory provide intriguing hints that infants, like older children and adults, strive to reconcile what they observe with what they believe. This evidence comes from experiments in which infants were presented with an impossible and a possible event and were foundnotto show the typical preference for the impossible event. Five examples are presented, involving infants aged 3 to 10 months, and ranging over several facets of infants' physical reasoning. It is argued in each case that infants' failure to look preferentially at the impossible event stems from their having arrived at some explanation for the event. In some instances, infants' explanations appear to be entirely self‐produced, as when infants spontaneously posit hidden objects to make sense of otherwise impossible events. In other examples, infants' explanations appear to depend on additional clues provided, either deliberately or inadvertently, by the experimental situation. Despite these superficial differences, however, the examples are all related in suggesting that infants, like older children and adults, actively seek explanations for inconsistencies in their worl