Environmental enrichment is a very important issue for zoos and is becoming more important for farm animals. Evaluation of the effect of enrichment in mostly done by behavioral measurement, but it is often difficult to apply statistics, due to low numbers. However, statistical techniques are available, for instance single case analysis and meta-analysis, that - alone or in combination - could replace analysis of variance and can be applied to a single animal. Characteristic of the mentioned techniques is a very careful preplanning of the experiment, i.e. the number and the order of treatments given to the subject(s). Our research in farm and zoo animals is concentrated on comparison of time budgets related to housing conditions. One of the tools to find individuals that show aberrant time budgets is correspondence analysis [2]. In addition to this work enrichment experiments are done to ameliorate the keeping conditions of animals. Single case analysis is used to evaluate enrichment statistically. Examples of the use of this technique are:

• single case analysis and meta-analysis of the effect of food deprivation on crowing in the rooster [3],
• single case analysis of environmental enrichment in an Amur tiger, in which the amount of stereotypies was drastically reduced (P<0.001), and
• single case analysis and meta-analysis of feeding enrichment in a group of three orangutans [4].

As an example, the research on orangutans will be explained here in some more detail. Time budgets of orangutans (Pongo pygmaeus pygmaeus) show large amounts of spare time. Feeding enrichment is one of the means to increase appetitive activity related to feeding. In a randomized design we presented three conditions available to the orangutans (A= feed dispenser; a log with holes filled with peanut butter, B= raisins hidden under the straw substrate, and C= control, the normal condition). During one month on 14 different days these conditions were randomly presented to the orangutans at the beginning of the day. Behavior was measured for 5 hours after releasing the animals in their enclosure. The time budgets of the animals were compared between the three conditions by Single Case Randomization Tests [1]. Single case analysis shows significant increases in activities on an individual level (P<0.05). In conclusion, the comparison of time budgets of animals between zoos with different keeping conditions gives indications of welfare of individual animals related to housing and keeping conditions. Carefully planned experiments, statistical tests at the level of the individual and combining results of these tests in meta-analysis provide sophisticated methods for evaluation of environmental enrichment in zoos with only one or few subjects.

References

1. Damme, G. van; Onghena, P. (1993). SCRT 1.1: Single Case Randomization Tests, User Manual. ProGamma, University of Groningen, The Netherlands.
2. Koene, P. (1995). The use of time budget studies in captive propagation and zoo biology. In U. Ganslosser, J.K. Hodges and W. Kaumans (eds.). Research and Captive Propagation, 272-284. Fuerth: Filander Verlag.
3. Koene, P. (1996). Temporal structure of red jungle fowl crow sequences: single-case analysis. Behavioural Processes, 38, 193-202.
4. Koene, P.; Sieler, C. (1996). Feeding enrichment in orangutans: statistical evaluation by single-case randomization tests. Ethologentreffen (Groningen, 1996), 18.