Choice tests in animal welfare research

Implications of decision mechanisms for the design and interpretation of choice tests

PI: Dr Melissa Bateson
Student: Lucy Asher

Introduction. The results from behavioural choice tests have been widely used in animal welfare research as the basis for making recommendations regarding the husbandry of captive animals (Forbes et al 1997). For example, Mason et al (2001) showed that when mink were given a choice of seven different resources, they worked much harder for access to a swimming pool than they did for access to toys. On the basis of this research, Mason et al concluded that water is far more important to mink than toys. The applied outcome of this research is that some countries now require fur farmers to provide water baths for mink at considerable additional cost to the industry.

It is an implicit assumption underlying the majority of choice tests, including that of Mason et al, that the preferences obtained from the test are independent of the set of options available in the test. This follows from two largely untested assumptions about the mechanisms of choice: first, that animals use absolute evaluation mechanisms to assign value to options, and second, that the probability of choosing an option is proportional to the ratio between the value of that option and the sum of the values of the other options available. If either of these assumptions is incorrect then preferences can differ depending on the composition of the choice set (Bateson 2004). In the case of the mink for instance, if the animals use a comparative as opposed to an absolute mechanism to assign value to the various resources being compared, then the value of a resource is not an absolute property of that resource, but is determined by the other options in the choice set. For example, the mink assign very little value to toys when a pool is available, but in the absence of a pool the toys might be valued much more highly because they provide an alternative outlet for play behaviour. Given the economic implications of changing the requirements for the husbandry of captive animals it is clearly very important to establish to what extent the preferences emerging from choice tests are dependent on the precise context in which they were obtained.

So far there has been surprisingly little research into the effects of the size and composition of choice sets on animal choice. The only research of relevance comes from the foraging literature where recent work has demonstrated that foraging preferences recorded in a binary choice context can change when a third option is added to the choice set (Bateson 2002; Bateson et al 2002; Shafir et al 2002; Bateson et al 2003). Early indications suggest that foraging hummingbirds may use comparative evaluation mechanisms when faced with a choice of options that vary in more than one dimension of interest (in this case the volume and concentration of nectar in a flower) simultaneously (Bateson et al 2003). These results from foraging birds fit with data on human consumer behaviour that suggest that humans may use comparative evaluation mechanisms when faced with multidimensional choices (Huber et al 1982). It has been argued that although comparative evaluation mechanisms can sometimes result in irrational decisions, the algorithms involved are simpler and faster to implement. More research is necessary to understand first, the exact choice mechanisms used by animals, and second, the generality of these choice mechanisms.

Hypotheses In the course of this project we would like to test the following hypotheses:

1. That the changes in preference previously observed between binary and trinary choice contexts are the result of animals using comparative mechanisms to evaluate the options between which they are choosing.
2. That animals use comparative evaluation mechanisms when the options between which they are choosing differ in multiple dimensions.
3. That animals use comparative evaluation mechanisms either when they have to make a decision fast, or when the costs of making the wrong decision are not particularly great.