“External Influences on the Feeding of Carnivores § Demonstration of Dietary Aversion Learning”, Roger A. Mugford1977 ()⁠:

Most readers will be familiar with the phenomenon of “bait shyness” in rats and other vertebrates, induced as a result of temporal association between sickness and a characteristically flavored diet or drink (see reviews by Garcia et al 197450ya; Rozin & Kalat1971). The experiment to be described demonstrates this phenomenon in cats.

16 cats were assigned to two groups of 8 on the basis of matched sexes and body weight. On the first day of the experiment (Day 0, Figure 9) they were all individually fed 100g of an experimental meat-based canned diet (M, based mainly upon cow lungs). The cats’ normal maintenance diet was a complete dry cat food, so that M was quickly eaten because it was both palatable and novel. Immediately after they had finished eating, one group (aversion) was given lithium chloride (1% body weight of a 0.157N solution, as in Nachman & Ashe1973’s study with the rat). The other (control) group was treated with an equimolar solution of sodium chloride.

Figure 9: Mean intakes of canned meat by cats, demonstrating dietary aversion learning.

Analysis of variance (between groups, days 3–80) revealed a statistically-significant (p < 0.001) depression of intake by the aversion group of M, but not of their usual maintenance diet. Cats in the control group ate meals of M that were 3× larger than aversion cats’ meals of the same diet, whereas both groups continued to eat the maintenance dry cat food as before. Comparisons of intakes on each of days 3, 10, 20, and 40 confirmed that the dietary aversion was sustained for a period in excess of 1 month after only a single exposure to the U.C.S. (LiCl). By day 80, the two groups ate equally large meals of M.

However, this interval of 40–80 days probably does not encompass the time limit for retention of a dietary aversion, since the learned response would have suffered interference (or M-acquired properties of “learned safety”) in 5 meals when M was offered (see Rozin & Kalat1971, pg77). Nevertheless, the generality of the phenomenon of dietary aversion learning has been extended to include the cat on the basis of the results from this experiment.

One could speculate that this ability might have a parallel in nature, serving to protect wild felids from repeated ingestions of prey or organs of prey that might induce gastric distress. The widely reported avoidance of insectivorous rodents by predatory carnivores (eg. Macdonald1977; Pearson1966) might be an example of aversion learning occurring under natural conditions.