Bilateral vestibular loss (BVL) causes a unique behavioural syndrome in rodents, with symptoms such as locomotor hyperactivity and changes in exploratory behaviour. Many of these symptoms appear to be indirect consequences of the loss of vestibular reflex function and are difficult to explain. Although such symptoms have been reported before, there have been few systematic studies of the effects of BVL using automated digital tracking systems in which many behavioural symptoms can be measured simultaneously with high precision. In this study, data were obtained from rats with BVL induced by intratympanic sodium arsanilate injections (n = 7) or sham injections (n = 8) and their behaviour in the open field was measured at 3 days and 23 days post-injection using Ethovision™ tracking software. BVL rats demonstrated reduced thigmotaxis, with more time spent in the central zones. Twenty-three days post-injection, BVL animals showed increased locomotor activity in the open field. The increase in activity was also reflected in the number of transitions between each zone of the field. In addition to increased activity, BVL animals showed increased whole body rotations following lesions. Using linear discriminant analysis (LDA) and random forest classification (RFC), we were able to show that the indirect behavioural effects of BVL, excluding direct measurement of vestibular reflex function, could correctly predict whether animals had received a BVL with a high degree of accuracy at both day 3 and day 23 post-BVL (83% and 100% for LDA, and 100% and 100% for RFC, respectively). RFC has been similarly successful in classifying other hyperactivity syndromes such as attention deficit hyperactivity disorder. These results suggest that BVL results in a unique behavioural signature that can identify vestibular loss in rats even without direct vestibular reflex measurements.
Read full abstract