Experience with a stimulus changes a person forever... if not forever, at least for 17 years (Mitchell, 2006). Any process engaged by the brain leaves a trace behind that affects similar future processing. This form of plasticity aims to optimize the brain toward more efficient interaction with stimuli that have been encountered before, adapting the individual to the environment. Similar adaptation manifests itself as faster and more accurate performance to repeated stimuli. Repetition priming is the term used in the literature to describe this kind of facilitation. A crucial open question on repetition priming is whether experience makes stimulus processing more efficient or facilitates the selection of an appropriate response to a stimulus. In the literature these two alternatives are theorised respectively by facilitation in perceptual and conceptual networks and by rapid response learning (Chapter 1). In the work presented here, processing/response accounts are tested in person recognition. This thesis presents evidence that repetition priming in person recognition is mostly produced by facilitation in perceptual and conceptual networks, but, when stimuli are repeated three or more times, additional facilitation due to rapid response learning occurs as well (Chapters 2 - 3 - 4). Orthogonal tasks were used to distinguish the contribution of perceptual/conceptual facilitation and rapid response learning in behavioural facilitation (Chapter 5). Furthermore, LRPs/ERPs shed light on the neural processes that generate rapid response learning in person recognition: the response accessed from rapid response learning interacts with the response obtained from stimulus re-processing and determines benefits when congruent and costs when incongruent (Chapter 6). These results show that rapid response learning and facilitation in perceptual and conceptual networks refer to two complementary types of memory that can lead to repetition priming. This thesis proposes a framework in which facilitation in perceptual and conceptual networks is supplemented, under specific circumstances, by rapid response learning (Chapter 7).