The present report describes the emotional responses of different strains of mice to exposure to a novel open space model of anxiety using a 3D spatial navigation task. The 3D maze is modification of the radial maze with flexible arms that can be raised above or lowered below the horizontal level of a central platform. To access the arms animals need to cross a bridge linking the arms to the central platform. In this model, mice are exposed to novelty in an unfamiliar open space setting with no safe alternative. Fear from novelty is compounded with the need to explore. The drive to escape and the drive to approach are intermingled making this open space model radically different from the current models of anxiety which provide animals with the choice between safe and anxiogenic spaces. In a series of experiments, we examined the behaviour of different groups of mice from C57, C3H, CD1 and Balb/c strains. In the first experiment, different groups of C57 mice were tested in one of the three arms configurations. In the second experiment, C57 mice were compared to C3H mice. In the third experiment, C57 mice were compared to CD1 and Balb/c mice in the raised arm configuration over three successive sessions. In the fourth experiment, we examined the behaviour of C57 mice in the lowered arm configuration with an open and an enclosed central. In the final experiment, we examined the difference between C57 and C3H mice of both genders. Using several spatio-temporal parameters of the transition responses between central platform, bridges and arms, we have been able to show consistent results demonstrating significant differences between C57 and C3H mice, and between Balb/c and both C57 and CD1 mice. C3H appear more anxious than C57 mice, and Balb/c mice seem more anxious than C57 and CD1 mice. We also observed significant differences between sexes in C3H mice but not in C57 mice. C3H male mice appear more anxious than C3H female mice and than both C57 male and female mice. In the lowered arm configuration with an enclosed central platform, C57 mice took longer time to make a first entry to an arm, made more visits to bridges before first entry to an arm and required longer time between re-entries to arms, spent longer time on the central platform and shorter time on arms compared to mice in the other arm configurations. They also made frequent entries to the centre and bridges compared to mice in the lowered arm with an open central platform. These results demonstrate not only the sensitivity of the parameters of the test but also the consistencies and concordances of the results which make this 3D maze a valuable new tool in the study of the underlying neural mechanisms of anxiety responses in addition to learning and memory, and in assessing the effects of potential anxiolytic drugs. In this report we examine methodological issues related to the design of animal behavioural paradigms and question the value and the construct validity of the current models of human anxiety.