The other-race effect (ORE) refers to the impoverished recognition of other-race faces relative to own-race faces. The aim of this thesis was to investigate the ORE in two populations (single-race and multi-race) from infancy to adolescence/adulthood as a means of testing and clarifying both the perceptual expertise account and the categorisation-individuation model (CIM). Of six experiments using age-relevant measures, four investigated the ORE when all other factors (social/motivation) are equal, and two investigated when non-race based social categorisation is manipulated. Using a more refined set of stimuli (limited to external facial information), Experiments 1 to 4 supports the assertion that recognition is affected by racial experiences. Furthermore, the interpretations extended the perceptual expertise account by suggesting that such experiences can also differ across face gender and face race at different stages of development. Experiments 5 and 6 extended the categorisation-individuation model by suggesting that population differences and the subjective importance of motivation to individuate faces could modify the ORE when non-race based social categorisation (personality affiliation) was manipulated. These findings were interpreted using a principle of concept of broadly tuned versus narrowly tuned representation within Valentine's (1991) multidimensional face-space model. The findings from this thesis address the importance of considering a combination of face experiences (e.g., race, gender), the social importance of individuating faces at different stages of development (social relationships), and how broadly tuned face representation can have a different effect on ORE compared with a narrowly tuned face representation.