Social Learning Theory and Animals: Observational/Imitation Learning

Social Learning Theory and Animals:  Observational/Imitation Learning

Cheryl Aguiar, Ph.D. on December 04, 2012

Social learning theory and animals: Does observational/imitation learning have a place in training or behavior in animals?

In human psychology, Albert Bandura, a self-described social-cognitive psychologist developed Social Learning Theory in the 1970’s (Bandura,1977). Bandura’s Social Learning Theory was rooted in Learning Theory but added the social element.  Social Learning Theory added that learning and new behaviors can occur through modeling, or, watching other people.  Can, or does, the same happen with animals? Do animals learn new behaviors by simply watching other animals? The research on this is complicated at this point and seems to have become muddled in the mechanisms involved and whether or not the behavior needs to be goal-directed. So, for now, we will need to “sort of” leave the research findings where they are at, at least to me: inconclusive.

Bandura’s famous experiment in 1961 was the “Bobo Doll” study (Bandura,1961).  Simply, in this study, he showed some children a film of a woman punching a Bobo Doll (these are those inflatable clowns with sand in the bottom) and yelling at it. He did not show this film to another group of children. He then had the children who saw the film play in a room with a Bobo Doll. Some of the children went up to the doll and started punching it and yelling at it. The group that did not see the film…none of the children did this. So the children imitated the aggression of the woman on the doll without any reinforcement.  So why did the children punch and yell at the doll? Social Learning Theory says they learned this behavior purely through observation and imitation.

Social learning theory in animals postulates that animals can learn by observation of, or interaction with, another animal (especially of the same species) or its actions (Box, 1984; Galef, 1988).

In learning theory, there are basically three types of experiences that can elicit novel behavior (Rescorla, 1988): (1) stimulus (verbal command, leash taut, moving away, etc.); (2) stimulus-stimulus (a click is paired with a treat, a verbal command is paired with a treat, lights on is paired with feeding, etc); and (3) response-reinforcement (the learner pushes on the dog door and gains entry into the house, the rat presses a lever and receives food, the learner puts up one paw and receives a treat, etc.).

From these three basic types, the last one, response-reinforcement is where social learning is most reasonably a subset of learning. However, unlike our examples above, it is the demonstrator, not the learner, who makes the response that is learned.  This effect of exposure to observing a response-reinforcement relationship then producing novel, somewhat matching behavior by the observer is social learning and I believe we should think about using it in our training of our own pets and of clients’ pets.

Personally, I can anecdotally cite many instances where it appears that an animal can and does learn by imitation.

• One dog is taught to use a pet door by giving encouragement and treats to go through the door. A new dog is introduced to the environment and learns to use the dog door after watching the experienced dog.
• A dog is afraid to swim. You bring a water-loving-retrieving-loving dog that bounds in and out of the water swimming and soon the other dog is swimming as well.
• You have a nervous barking dog. You know you did not elicit the behavior as you have had many dogs before and they have never been like this. You get a new puppy that comes from a good breeder. You meet all the dogs at the breeders…you may even KNOW these dogs and you have noted they are all confident, non-barkers. Your new puppy becomes a barker and nervous after just a few weeks.
• You tie two dogs together. One is a confident dog when going for walks. The other is fearful and unsure when going for walks. After a few days, the unsure dog is exhibiting confidence on the walks as well. After several times, you leave the original confident dog at home and just take the unsure dog and he is now confident as well.
• You have a new colt and he is about ready to go for a ride off the property. You ride an experienced and confident horse while you “pony” (just hold the lead rope and take the new colt with for the ride…riderless) the new colt several times.

At least at face value (face validity/anecdotally) social learning does occur with animals and, as a subset of learning theory, it seems that we can and should use it in developing new behaviors or changing unwanted behaviors.

Box, H. 0. (1984). Primate Behavior and Social Ecology. Chapman & Hall, London.
Bandura, A., Ross, D. & Ross, S.A. (1961). Transmission of aggression through imitation of aggressive models.  Journal of Abnormal and Social Psychology, 63, 575-82.
Bandura, A. (1977). Social Learning Theory. New York: General Learning Press. 
Galef, B .G. (1988). Imitation in animals: history, definition and interpretation of data from the psychological laboratory. In Social Learning: Psychological and Biological Perspectives (ed. T. R. Zentall and B. G. Galef), pp. 3-28. Erlbaum, Hillsdale, NJ.
Heyes, C. M. (1994) Social learning in animals: Categories and mechanisms. Biological Review, 69, 207-231.
Rescorla, A. (1988).Behavioral studies of Pavlovian conditioning. Annual Review of Neuroscience XI, 329-352.

---

Cheryl Aguiar, Ph.D. was trained in Social and Industrial/Organization Psychology. She received her Ph.D. at Colorado State University (CSU). Cheryl Directed an on-campus Institute at CSU for over a decade. While at CSU, she also was the successful recipient of over 3 million dollars in research and program grants, taught courses, served on student committees and published findings from her own research projects along with those of her students. She has been a dog trainer since 1974. Mainly she trains her own dogs in various performance or working roles such as Tracking, Obedience, Schutzhund and hunting trials. She also served in the 1980’s to 1990’s as Training Director for several years in a Schutzhund Club in the Twin Cities and more recently as Training Director in the local NAVHDA chapter. She is now the President of the Rocky Mountain Chapter of the North American Versatile Hunting Dog Association (NAVHDA). Cheryl has successfully titled dogs in AKC obedience, AKC tracking, AKC conformation, Schutzhund and NAVHDA. Since 2005, she pursued her true loves: education and dogs and founded E-Training for Dogs, the first online webinar entity for dog lovers.

IAABC Supporting Member
Cheryl Aguiar, Ph.D.
President, E-Training for Dogs, Inc
Online Dog Training Courses-We Bring the Seminar to You!
Office: (970) 231-9965
cheryl@e-trainingfordogs.com
www.e-trainingfordogs.com