How do people know what categories objects belong to? Traditional accounts of categorisation typically assume that concepts comprise perceptual or functional features. By contrast, recent accounts of conceptual structure emphasise the dual role of sensorimotor (i.e., perception-action experience of the world) and linguistic distributional information (i.e., statistical distribution of words in language). This thesis contains a literature review and four empirical papers describing pre-registered experiments, which explore how the degree of sensorimotor-linguistic representational overlap between category- and member-concepts may drive object categorisation. Chapter 1 presents a review of the literature, covering relevant theories of categorisation, as well as sensorimotor-linguistic theories of conceptual processing. Chapter 2 presents a study which explored the role of sensorimotor and linguistic distributional information in processing advantages. This study found that overlap in sensorimotor and linguistic distributional representations between category (e.g., dog) and member (e.g., Labrador) concepts reliably predicted performance (accuracy, RT) in a speeded picture category verification task. Chapter 3 reports a study that contrasted the traditional prediction of a basic-level advantage with the sensorimotor-linguistic prediction that representational overlap, not taxonomic level, is more important to categorisation. In a forced-choice categorisation task using labels only, participants decided between a basic- (e.g., dog) and superordinate-level label (e.g., animal) for a target object label (e.g., Labrador). While basic-level labels were overall chosen faster and more frequently, an exploratory analysis suggested that basic-level categorisation was slowed down when sensorimotor-linguistic overlap was greater between the target object label and the superordinate label. Chapter 4 describes the collection of a normed set of 800 photographs of 200 natural objects and artefacts, and their most frequent names. An exploratory analysis of the object recognition latencies associated with each photograph found that word frequency and length averaged over all names given to an image predicted object recognition time better than the word frequency and length of the most frequent response. Chapter 5 reports a study that used the images and names collected in the study reported in Chapter 4, and examined the role of sensorimotor and linguistic distributional information in an ultra-rapid object categorisation paradigm with backwards masking. This study found evidence for the effect of linguistic distributional on sensorimotor information on categorisation accuracy, but not RT, nor was there a systematic relationship between perceptual processing time and sensorimotor-linguistic information. In summary, the findings presented in this thesis provide support for a novel account of object categorisation based on sensorimotor and linguistic distributional representational overlap between category and member concepts.