The central proposal of the current thesis is that stimuli cannot be evaluated in isolation but rather, can only be rendered meaningful on the basis of their inclusion in an appropriate and circumscribed set of related stimuli - that is, preliminary categorisation of stimuli is essential to their subsequent identification within the category. Thus the perceived properties of a stimulus are determined by the context in which it is presented, and this context may be explicitly defined by the conditions of presentation or may be inferred by the perceiver himself. Further, the level of analysis to which any stimulus will be subjected will depend on the task requirements of the situation and consequently on the judgemental criteria chosen by the subject to meet those requirements. It is also proposed that categorisation (or the establishment of a stimulus set) and the subsequent intracategory differentiations involve processes of constant comparing and contrasting - in other words, perceptual organisation of a stimulus is achieved by relational judgements. Concordant with the theoretical framework outlined above, two series of experiments (the first with rats, the second with children) were conducted with the aim of investigating further the operation of the stimulus set and the factors influencing its establishment and function. Taken altogether, then, the evidence from both series of experiments lends support to the current theoretical position - namely that the perceptual organisation (or structuring) of stimuli is contextually bound and is dependent on the prior categorisation of the stimulus into a connected ('contrastive') set within which its meaning can subsequently be specified. Further, evidence has been presented which attests to the important role of relational comparisons and judgements in perception. (Abstract shortened by ProQuest.).