Dairy cows' use of colour cues to discriminate between people

Abstract

The role of the colour of overalls as a cue used by cows to discriminate between people was studied in four experiments using a total of 66 Danish Friesian cows kept in tie-stalls. In all experiments, the average distance cows kept from the person was calculated from 12 scores during 1 min with the person standing in front of the cow. Experiment A. The distance cows (n=24) kept, from two unfamiliar people wearing red or yellow overalls, was tested. There was no difference in the distance the cows kept from the two people and the two colours (P>0.10). Experiment B. The distance cows (n=18) kept from an unfamiliar person wearing familiar coloured overalls (blue) was compared to when the same unfamiliar person was wearing red or yellow overalls. There was no effect due to the colour of the overalls (P>0.10). Experiment C. Cows (n=12) were handled repeatedly, by two handlers wearing overalls of the same colour. Each person handled six cows gently and six cows aversively. There was no significant difference in distance kept from the gentle and the aversive handler, after 11 treatments (P>0.10). Experiment D. Cows (n=12) were handled repeatedly by one person wearing either red or yellow overalls. The person wore one colour when handling the cows gently and the other colour when the cows were handled aversively. The cows kept a longer distance when the handler wore the colour worn during the aversive treatment than when the handler wore the colour worn during the gentle treatment (P<0.01). The cows also responded according to the colour of the clothes when worn by an unfamiliar person, although they kept a shorter distance from the unfamiliar person than to the handler.

Introduction

Aversive handling of farm animals, and the animals' resulting fear of people, can substantially reduce both the production and the welfare of the animals, and increase the risk of injury to the handlers (Hemsworth and Coleman, 1998). Aversive handling has been shown to reduce milk production in dairy cows (Seabrook, 1984; Breuer et al., 1997). Dairy cows from higher-producing herds allow people to approach them more closely (Seabrook, 1984; Hemsworth et al., 1995), while the presence of an aversive handler during milking can reduce milk yield by 10% (Rushen et al., 1997). Such fear of people as a result of aversive handling, presumably arises from the animal learning to associate aversive handling with the person performing the handling through classical conditioning (Hemsworth et al., 1993; De Passillé et al., 1996). Ideally, aversive handling of farm animals should be reduced to a minimum, but some aversive handling is usually unavoidable, e.g., during veterinary treatment. It is therefore important to understand the factors that might reduce the animals' tendency to become fearful of people, particularly of their normal caretaker, as a result of such treatment.

One important issue is the extent to which animals can discriminate between different persons as opposed to generalising their fear to all people. At a practical level, it would be advantageous if animals which had been handled aversively by one person would not generalise their fear to other people. On the other hand, attempts to tame animals by increased gentling (e.g., Boissy and Bouissou, 1988; Boivin et al., 1994) will be more successful if the animals do not strongly discriminate between different people but show a reduced fear to people in general (Hemsworth et al., 1993; Jones, 1994). Dairy calves (De Passillé et al., 1996) and dairy cows (Munksgaard et al., 1997) have been found to distinguish readily between people who handle them aversively and those who handle them gently. However, calves do not always discriminate between familiar and unfamiliar people (Arave et al., 1992; De Passillé et al., 1996).

The ability to discriminate between different people is likely to be enhanced if the animals have sufficient cues available (Hemsworth et al., 1993), and cattle and pigs have been shown to discriminate people by the colour of the clothes (De Passillé et al., 1996; Munksgaard et al., 1997; Tanida and Nagano, 1998).

The aim of this series of experiments was to examine in more detail the ability of dairy cows to discriminate between handlers based on the colour of the clothes worn. We examined whether: (1) cows respond differently to people according to the colour of the clothes worn; (2) cows react differently to people wearing unfamiliar colours rather than the colour of clothes normally worn by the stockpersons; (3) cows can learn to discriminate between two people wearing the same colour of clothes; (4) cows can learn to approach or avoid the same person based on the colour of clothes being worn.