Could Ultrasonic Vocalisations Provide The Elusive, Graded Measure Of Affective State Needed To Inform Refinements For The Laboratory Rat?

Animal research requires that studies consider the 3Rs (Reduce, Refine and Replace). Refinement involves using methods which cause the least suffering to the animal, and this can be more challenging to assess objectively. To achieve refinement requires the ability to quantify the animals affective or 'emotional' experience. Although overt changes in the animal's behaviour can provide some information, these do not offer sufficient sensitivity to understand the emotional impacts. Our research has pioneered the development of objective methods to measure animals' affective state. By translating behavioural methods from human psychology, we have developed and validated novel assays which accurately measure animals positive or negative affective experience of different interventions. We have used one of these assays, the affective bias test, to demonstrate that high frequency calls emitted by rats provide a graded measure of their individual emotional experience. This raises the exciting possibility that measuring these ultrasonic vocalisations may provide a simple and easy to use method to evaluate refinements. Our approach will not require costly equipment or specialist expertise meaning it could be readily set-up and utilised within any animal research facility.

Previous studies suggest that adult rats emit 50kHz or 22kHz calls when they experience putative positive or negative events, respectively. Although indirect evidence to support this idea has been achieved, we have not previously had a reliably objective method to accurately quantify the relationship between these calls and the animal's individual affective experience. Direct correlation between these ultrasonic vocalisations and affective state has so far only been tested in relation to human simulated play or 'tickling' in rats. The affective bias test is a simple, reward-based task which we have shown can provide a graded measure of the affective state of rats during acute psychosocial or pharmacological interventions. The task measures affective biases related to associative learning and memory. Affective bias describes the effect of emotional state on different cognitive processes and are known to modulate learning and memory and decision-making behaviour in human and non-human animals. Cognition can be biased in either a positive or negative direction influenced by both short-term changes in emotional state and arising from changes in core affect akin to mood. Using our affective bias test, we will test the hypothesis that ultrasonic vocalisations can provide an equally sensitive measure of animal's affective state but using a method which can be widely adopted across the scientific community. Our studies correlating calls with our measure of positive or negative affective state in the ABT will provide direct evidence of the validity of the approach. We will also test a cohort of animals in a putative depression-like state where we have found that negative affective biases are potentiated. Our aim is to establish whether vocalisations are similarly altered and hence provide a way of understanding animals core affective state. These validation studies will tell us whether recording animals calls can provide information about their emotional state both during acute experiences and in relation to core affect or 'mood'. In our final work package, we will translate the approach to testing in the home-cage and at an individual and populations level in response to manipulations commonly used the laboratory. We will also test two refinement methods, habituation to human handling and a novel caging system designed to increase space and social interactions whilst also being practical for handling. The outcomes of this project will be methods to enable many more researchers to study refinement and enable institutions to monitor the affective state of their colonies using a simple, non-invasive and non-specialist method.

The ability to quantify affective state accurately and objectively is critical for progress in the welfare of laboratory rodents. We recognise overt sign of distress, but we also need to understand the animal's affective state, something much harder to achieve through observations alone. Our research groups have pioneered the development of objective methods to quantify affective state in rodents. Although achieving excellent validity, the methods are specialist and not readily adopted by the wider community. We now propose a project where we use these methods to investigate if ultrasonic vocalisation, recorded from animals during specific affective manipulations, can provide a simple, graded measure of animal's affective experience. The affective bias test is sensitive to acute psychosocial and pharmacological manipulations exhibiting dose-dependent effects. We will use this method to establish if ultrasonic vocalisations (50kHz or 22 kHz), emitted by rats, can provide a simple, easily applied method to measure positive or negative affective state. We will quantify vocalisations and affective biases during acute state changes as well as in animals with chronic changes in core affect e.g. our depression models have found that those in a depressed state show exaggerated negative biases in the affective bias test which we predict will also be reflected in their USVs. In the final objective, this project will translate these laboratory focused studies into the 'real world' scenario testing vocalisations in the home-cage environment. We will test how vocalisation patterns relate to anticipation, or experience, of common procedures and investigate the positive welfare effects of two types of refinement, habituation to human handling and a novel, modular caging system. These findings will provide proof of concept and validation data as well as establishing a simple, home-cage method for quantifying vocalisations and hence affective state in rats.