Development of a raceway method to assess aversion of domestic fowl to concurrent stressors

Abstract

The requirement for assessing the effects of stressor combinations in improving the welfare of animals has not been widely recognised. Knowledge of the effects of concurrent stressors is needed to improve environments such as transport, where animals are presented with many simultaneous challenges. However, no method for measuring the effects of different stressors with a common unit is currently available. A locomotor passive avoidance method was developed as a common currency measure of the aversion of domestic fowl to concurrent stressors, using vibrational and thermal stressors as an exemplar. Juvenile fowl, fasted overnight, were trained to run a raceway into a goal-box for small food rewards (FR1). When running consistently, the reinforcement schedule was superimposed with a FR5 treatment schedule (60 min confinement in the goal-box with either a control of no other stressors [N] or concurrent vibration and thermal stressors [VT]). Subsequent latency to return to the goal-box was recorded as a measure of aversion. The factors affecting bird response were addressed in a series of experiments to optimise the method and clarify interpretation of results. Pre-feeding (20% ration 2 h prior to testing) did not affect response, but increasing the number of treatment presentations facilitated learning and increased method sensitivity. Treatment responses were consistent across experiments; overall VT was avoided (P<0.001), but N was not. However, there was large individual variation in response to VT. A final experiment indicated that, given a visual discriminatory cue, birds were capable of learning the required association between entering the goal-box and receiving the treatment, suggesting that the delay responses were due to aversion rather than the immediate impact of treatment on ability to respond. Further work is required to test the singular stressors, but the method retains common currency potential for assessing aversion to multiple stressors.

Introduction

It is important to assess how domestic animals respond to management, transportation and handling procedures that are imposed as part of normal farming practice. Many of these procedures may be perceived as aversive, and methods for quantifying animal aversion have been developed and reviewed (Rutter, 1989, Rushen, 1996). However, environmental stressors are rarely experienced individually and to understand their true impact on animals, responses to various combinations of stressors require characterisation. To achieve this, new methods are needed to assess the effects of very different types of stressors using the same unit of currency. The combination of vibrational and thermal stressors was chosen as an exemplar to assess aversion to concurrent stressors, but is also relevant to the welfare of transported birds (Weeks and Nicol, 2000). Vibration is immediately perceived, novel and likely to be fear-eliciting, whereas there is a time-lag before a hot temperature induces heat stress, it is unlikely to induce fear and may have been experienced previously. Thus responses common to both stressors are few (Abeyesinghe et al., 2001).

A locomotor passive avoidance method was devised to circumvent these problems as a common currency of aversion. The premise of the raceway method is to titrate motivation to obtain food against motivation to avoid an aversive stimulus. The method is similar to both Miller's (1944) work on conflict behaviour, and Petherick et al.'s (1992) investigation of laying hens' motivation in response to food and a negative reinforcer. Miller trained rats to run an alleyway into a goal-box for food and, after subjecting them to a brief electric shock while eating, measured the distance from the goal-box at which they stopped on a subsequent run. This was assumed to represent the point of equal motivation to approach and avoid. However, there was no indication of the time period over which animal responses were measured or accounting of factors other than attraction and avoidance which might be influencing responses. Petherick et al. (1992) ran hens once per day and always rewarded them with food, even when they also received an aversive stimulus. It is likely that the single daily trials and the consistency of the reward interfered with the hens' ability to learn the association between the aversive stimulus and the goal-box. The risk of the aversive treatment may have been perceived as worthwhile to obtain the guaranteed food, reducing sensitivity of the measure.

In the current study, once birds were running a raceway consistently for food, a treatment schedule was imposed, which took precedence over the food reward. Subsequent latency to return to the goal-box was recorded as a measure of aversion. Birds were therefore required to associate a place (the goal-box) with both stimuli and respond to memory of the events occurring there. Assuming that both vibration and thermal stress were equally associated with the goal-box, this approach allowed equal comparison of the stressors on an equal basis and, with confinement of sufficient duration to allow the thermal stressor to take effect, over different time periods (Abeyesinghe et al., 2001). The major factors affecting bird responses in the raceway were investigated in a series of experiments. In experiment 1, the effect of hunger on technique sensitivity was examined. In experiment 2, the number of treatment exposures was increased to facilitate learning of the association between receipt of the treatment and the goal-box. In experiment 3, the ability of birds to learn the required association (goal-box: treatment) was investigated.