When the time interval between two peripheral stimuli is long enough, reaction times (RTs) to targets presented at previously stimulated locations are longer than RTs to targets presented at new locations. This effect is widely known as "Inhibition of Return" (IOR). The effect is usually explained as an inhibitory bias against returning attention to previously attended locations. Thus, attentional disengagement is considered to be a necessary condition to observe IOR (Klein, 2000). We report data from three experiments with 2 different paradigms in which we show that IOR can be dissociated from the endogenous disengagement of spatial attention. Two main results are reported: 1) IOR is observed at an endogenously attended location in some situations, and 2) even after the endogenous disengagement of attention, facilitation instead of IOR is observed in other situations. We conclude that the endogenous disengagement of attention is neither sufficient nor necessary for IOR to be observed. However, by presenting an intervening event between the cue and the target, an IOR effect occurred in all conditions, indicating the importance of attentional segregation processes (exogenous disengagement) for generating IOR. These results are interpreted on the basis of cue-target event integration and segregation processes (Lupianez, 2010), which constitute our dynamical perceptual experience. IOR is explained as a cost in detecting the appearance of new attention-capturing information (i.e., the target) at locations where attention has been already captured by previous events (the cue).