Hide menu

Methods

The behavioral test and the experimental set-up

The study was conducted using three adult female Asian elephants housed at Kolmården Wildlife Park. The individuals were Saba (born 1968), Saonoi and Bua (born 1996 and 1997, respectively). The behavioral test was based on a food-rewarded two-alternative instrumental conditioning paradigm. The animals were taught to sample two odor ports and were food-rewarded when they performed an operant response (putting the trunk at a certain position of the experimental set-up) upon correctly identifying the rewarded odor. The training was carried out step-wise, where every new stage required a little bit more from the animals in order the gain access to the food-reward. The method was based on a voluntary participation of the animals and only positive reinforcement was used as a tool to shape the desired behavior.

In the method applied here, the animal had two options: (1) to correctly respond to the correct odor (correct decision), and (2) to falsely respond to the incorrect odor (incorrect decision). To measure performance the percentage of correct decisions per session was calculated for each individual and each session. In the initial learning tasks the criterion was set at 70 % hits, which corresponds to 21 correct out of 30 decisions in two consecutive sessions (corresponding to p < 0.01 two-tailed binomial test). In the discrimination tasks with structurally related odorants the criterion was set to 66.7 % correct in two consecutive sessions of 30 decisions each (corresponding to p < 0.05 two-tailed binomial test).

The odorants were presented to the animals in ventilated odor boxes and in pairs according to a previously determined order. A total of 10 experiments were conducted:

Experiments 1 and 2 were performed to demonstrate that the elephants could be trained to respond to a given odor and to discriminate between different odors. These experiments were also performed to assess the learning speed of the animals to master such tasks.
Experiments 3, 4, 5, 6 and 7 were performed to assess the elephant’s ability to perform intramodal stimulus transfer tasks and the learning speed of the animals in such tasks.
*During the negative transfer tasks, the rewarded odor (S+) was kept constant while a new S- was introduced as the unrewarded odor.
*During the positive transfer tasks, the unrewarded odor (S-) was kept constant while a new S+ was introduced as the rewarded odor.
*During the double-transfer task, both odors were replaced simultaneously by a new rewarded odor (S+) and a new unrewarded odor (S-).
Experiment 8 was performed to assess the discrimination ability of the animals with structurally related odorants. The odorants used in this experiment only differed from each other in carbon chain length.
Experiment 9 was performed to evaluate the long-term odor memory of the elephants. After two, four, eight and 16 weeks of recess in testing, previously learned odor combinations were presented to the animals.
Experiment 10 was a control experiment performed to affirm that the animals were actually responding to the odors and not to cues provided by the lids or ventilators of the odor boxes.
 

For the initial acquisition of the olfactory discrimination paradigm a set of eight odorants was used (amyl acetate, anethole, carvone, cineol, ethyl butyrate, limonene, pinene and 2-phenylethanol). For the assessment of odor discrimination capability with structurally related odorants, a set of four acetic esters was used (amyl acetate, butyl acetate, propyl acetate and ethyl acetate). The odorants were diluted at approximately equal subjective intensities, before presented to the animals.


Responsible for this page: Director of undergraduate studies Biology
Last updated: 05/20/11