Exposure to novelty has been shown to induce anxiety responses in a variety of behavioural paradigms. The purpose of the present study was to investigate whether exposition of naïve rats to novelty would result in a comparable or a different pattern of responses in an open space versus enclosed space with or without the presence of an object in the centre of the field. Lewis and Wistar rats of both genders were used to illustrate and discuss the value and validity of these anxiety paradigms. We examined a wide range of measures, which cover several aspects of animals' responses. The results of this study revealed significant differences between the behaviour of animals in an open space and in the enclosed space. It also revealed significant differences in animal's responses to the presence and absence of an object in the open space and in the enclosed space. In the enclosed space, rats spent most of their time in the outer area with lower number of exits and avoided the object area except when there was an object, while in the open space rats displayed frequent short duration re-entries in the outer area and spent longer time in the object area in presence of an object. The time spent in the inner area (away from the outer area and the object area) was significantly longer and the number of faecal boli was significantly higher in the open space than in the enclosed space. In the present report, we will discuss the fundamental differences between enclosed space and open space models, and we will examine some methodological issues related to the current animal models of human behaviour in anxiety. In the enclosed space, animals can avoid the potential threat associated with the centre area of a box and chose the safety of walls and corners, whereas, in the open space animals have to avoid every parts of the field from which there was no safe escape. The response of animals to novelty in an open space model appears more relevant to anxiety than in an enclosed space. The present studies revealed no correlations between the measures of behaviour in enclosed space and the measures of behaviour in open space, which suggest that these two models do not involve the same construct. Our results suggest that the enclosed space model involves avoidance responses while the open space model involves anxiety responses. The open space model can be very useful in understanding the underlying neural mechanisms of anxiety responses, and in assessing the effects of potential anxiolytic drugs.