The Dog Project: Experiment 6B- Reward Preference
May 31, 2015
CPT was busy this Sunday conducting Experiment 6B- Reward Preference for The Dog Project. The experiment examines whether dogs prefer social reward (tactile contact and praise from the owner) or appetitive reward (treats). The experiment also examines whether the reward preference may vary depending upon deprivation, satiation, visual salience, or lateralization preferences, whether the optimal reward may vary randomly amongst dogs, or whether the ideal reward is correlated to the temperament profile of the dog.
The proper selection and presentation of rewards is essential for maximizing motivation and learning during training periods. There are training and behavioral pundits who insist on employing food training with all dogs. Diametrically, there are training and behavioral pundits who insist that food is not necessary and that all dogs can be trained successfully exclusively with praise.
At CPT we believe in an eclectic philosophy that matches the reward choice to the specific dog and that varies the reward with each dog dependent upon the temporal preferences and emotional state of the dog, the stage of training for the behavior, and the goals of the training session. Nevertheless, at CPT we prefer to use empirical data and valid science, rather than conjecture, when determining our training methodologies.
Therefore, we relish our participation in the The Dog Project, which is obtaining scientific information about canine cognition, emotions, sensory perception, receptive communication, and inhibitory control that will improve training methodologies for both working and pet dogs.
In Experiment 6B, the dogs are first conditioned to understand that their owner and the food may appear on either side, that the owner is sitting in a chair with his/her back turned, and that treats are placed within a yellow food bowl. Once we complete the conditioning stage, we remove the dog from the room. Then, we open the door to allow the dog to choose either its owner or the food. To determine the strength of lateralization preferences versus reward preferences, we switch the side (right-left) that we place the owner and food after each repetition.
As you can see from the videos, some dogs prefer the owner, others prefer food, and others are not constant in their choice. We expect that our upcoming paper that summarizes the data of our MRI and field Reward Preference experiments will receive excellent media attention. More importantly, the paper should provide value to the dog training community.
(Sandy Springs, GA)