This post started out as one thing and transformed into another as I went along, as many of mine do. I have been familiar for a while with the term local enhancement for a type of social learning in dogs. I had some videos that I felt were good examples. But while researching this post and putting the clips together into a movie, I learned that the concepts and definitions were a lot less cut and dried than I thought.
This topic is up for lots of interpretation and discussion in the literature and I have found it to be underrepresented in discussions about dog behavior. I felt that at least an introduction to the subject would be helpful. I have gone with the most thorough, most recent, and most cited sources. I am open to additional information and hope for a good discussion.
Terms and Definitions
There are several different types of socially facilitated behaviors and social learning. These are two separate terms since behaviors can be socially facilitated without subsequent learning (Heyes, 1994, p. 214). Also the types of social facilitation overlap, and more than one can be going on at the same time. Among the types are behavioral contagion, local enhancement, stimulus enhancement, observational conditioning, copying, emulation, and imitation.
I got interested in local enhancement since I was pretty sure I saw it happening with my dogs. Like most of the other types, it involves animals performing similar behaviors as a result of observation or other perception of another animal. But it is not classified as imitation.
Here are a definition and an example of local enhancement from textbooks:
Local enhancement occurs when, after or during a demonstrator’s presence, or interaction with objects at a particular location, an observer is more likely to visit or interact with objects at that location (Hoppitt, 2013, p. 66).
…When local enhancement is in play, a model simply draws attention to some aspect of the environment by the action he undertakes there (for example, digging for worms). Once the observer is drawn to the area, he learns on his own (Dugatkin, 2004, p. 154-5).
Note that the observer animal doesn’t have to see the demonstrator animal. The observer can happen upon odors the demonstrator left or other signs of its actions in the area.
But if you have more than one dog, I bet you have seen local enhancement now and again.
Socially Facilitated Behaviors Without Learning
One thing that tripped me up is that it turns out local enhancement doesn’t have to involve learning (Thorpe, 1963, p. 154). Sometimes behavior is elicited socially but there is no behavior change in the future. The examples in my movie are probably of this type.
Some researchers say that local enhancement only takes place if the observer animal interacts at the location after the demonstrator has left (Heyes, 1994, p. 215). That is true in the first of my video examples but is not required by most definitions.
William Hoppitt (2013, p.66), whose definition I included first above, believes that the term local enhancement should be inclusive:
…We suggest that local enhancement be retained to refer to all such location effects, irrespective of whether they result in learning.
He also includes in his definition that the demonstrator animal may be either present or absent. Under that definition, both of the examples in my movie would qualify. When the demonstrator animal is still there, the classification of the observer’s behavior is more difficult. If the observer is interacting at the location at the same time as the demonstrator, we could be seeing general social facilitation. This is the tendency of animals to behave as others in their group are doing (Shettleworth, 2010, p. 467). Consider such contagious behaviors as yawning in humans and barking or fence running in dogs. In one of my examples in the movie, the dogs are attracted to a location but also running around excitedly in a group. Local enhancement and social facilitation are both probably involved.
Thus, local enhancement can end up with two animals doing the same thing at more or less the same place. But it is different from imitation or emulation. These are separate and precisely defined learning methods.
Not Imitation or Emulation
The term imitation has a specific meaning in learning theory.
Imitation: Performing the same action as a demonstrator by virtue of having seen the action performed. The action must be novel… (Shettleworth, 2010, p. 468)
Some definitions stipulate that the observing animal must use the same body parts to perform the behavior they observe. For example, in one study, marmosets watched a demonstrator open a canister. The marmosets that observed a demonstrator using its hands to remove the lids used only their hands. The marmosets that observed a demonstrator using its mouth also used their mouths to remove the lids (Voelkl, 2000). That difference marked their behavior as true imitation.
Emulation means that the observer copies only some of the elements of a complex action (Shettleworth, 2010, p. 468). The behavior by the observer may be different and may or may not achieve the same end as the demonstrator.
Local enhancement is a much looser concept than both of these. But the more I read about it, the more obvious it seems to me that since animals of the same species would respond similarly to the same stimuli in the same location, it would make sense for them to pay attention to what their conspecifics are doing and where. This could be advantageous and selected for.
When Do We See Local Enhancement?
Almost all studies of local enhancement in the natural environment involve foraging behavior. For instance, one animal will see that another has found a good source of food and will go to that area. Or an animal will happen on the scent of a conspecific and will learn to consume the food in that area or of that type.
Lab experiments follow this model as well. Rather than involving foraging, they generally involve a learned behavior that results in food.
Several domesticated species respond to humans in ways that involve local enhancement. One study shows local enhancement behaviors in horses as a response to the presence of a human near food (Krueger, 2011). There are several studies with dogs. Some of the human gestural and pointing studies with canids may involve local enhancement.
One of my examples shows two of my dogs investigating a spot in the grass after another dog had appeared to snap at and possibly eat an insect there. The two other dogs waited until the first dog left, then both went to the spot and sniffed for a while. Anthropomorphically speaking, here’s what I imagine going through their heads. “That was interesting. Is it something I need to know more about? Did she maybe leave a piece or is there another one of those? Do they live here?” In the second example, one dog discovers something alive and exciting under a step on my back porch. This is the one where you can see both local enhancement and socially facilitated behavior. After all the dogs arrived, they ran around excitedly and tried to get at the animal (which stayed safe).
Link to the video in case the above embed doesn’t work for you.
Social Learning Is…Learning
Some dog trainers treat social learning as exempt from learning theory. Nothing could be further from the truth. Depending on the type, social learning includes antecedents, behaviors, consequences, and/or classical associations. It’s just that some of the elements are a little different from what we are used to.
How about your dogs or other animals? Do you see local enhancement? How about between different species?
Dugatkin, L. A. (2004). Principles of animal behavior (No. Sirsi) i9780393976595). New York: WW Norton.
Heyes, C. M. (1994). Social learning in animals: categories and mechanisms. Biological Reviews, 69(2), 207-231.
Hoppitt, W., & Laland, K. N. (2013). Social learning: an introduction to mechanisms, methods, and models. Princeton University Press.
Krueger, K., Flauger, B., Farmer, K., & Maros, K. (2011). Horses (Equus caballus) use human local enhancement cues and adjust to human attention.Animal cognition, 14(2), 187-201.
Shettleworth, S. J. (2009). Cognition, evolution, and behavior. Oxford University Press.
Thorpe, W. H. (1956). Learning and instinct in animals.
Voelkl, B., & Huber, L. (2000). True imitation in marmosets. Animal Behaviour, 60(2), 195-202.
Thank you to Yvette Van Veen and Debbie Jacobs for leading me to some good resources on this topic. All conclusions are my own.
Copyright 2016 Eileen Anderson
8 thoughts on “Local Enhancement and Socially Facilitated Behaviors in Dogs”
Eileen, you impress me immensely! Please keep up the amazing work; I love your critical analysis.
Peggy, you are very kind! The hard part about this one was not to be too definite. There is so much up in the air about this subject!
“Depending on the type, social learning includes antecedents, behaviors, consequences, and/or classical associations.”
Is there any evidence you ran across that suggests that dogs can overcome significant fears through these processes? For example, would a dog who is terrified of the hose be able to overcome their fear solely through seeing other dogs interact with it?
I didn’t run across any evidence, sorry. But that’s not what I was looking for. Good question!
Eileen, I really enjoy reading your blog! There are so many useful articles and comments which help me improve my training. What made me make a comment in the first place, was the post “Lumping It: A Public Service Announcement” and how you trained Zani to nose target a piece of tape. I did the same, but used a Post-it sticky note instead. I did not experience the same lumping problem you did. I could move directly from increment #2 to #7 by pointing in the direction of the note with my finger. Often I find gestures and pointing to be helpful in my training. Now to why I am posting this here: For the past year or so I have explored social learning and imitation. Inspired by Claudia Fugazza’s Do As I Do training method, I can make my dog do new tasks by demonstrating what I want him to do. If I want him to do the Post-it trick, I first put my own nose to the note and then I tell him to do the same, i.e “Do It!”. Another example is to get him to put one of his toys into a shoe box. It was amazing how quickly he figured out what to do. A few repetitions was sufficient. The best part of it is how much he enjoys this kind of training. It’s much more fun than regular clicker training, and I use the same treats (and the clicker). It’s on par with nose work, which he enjoys very much.
You may find this news article interesting: http://www.sciencemag.org/news/2016/11/your-dog-remembers-more-you-think
The Do as I Do method was used in the experiment.
Thanks for your kind words about my blog. It sounds like the training you are doing with your dog is great fun, and I’m glad to hear you are using the clicker and treats when you do the emulation type work. I’ve read the Do As I Do studies, and they are actually one of the reasons I wrote about local enhancement. Imitation has a strict definition in learning theory (and has been argued about for decades), and it’s unlikely that it can be applied accurately between two species as physically different as dogs and humans. Many of the DAID behaviors have the same functional result, but the dog’s body is not used in the same way as the human’s. That would not qualify as imitation. Your nose touching behavior could be an exception: a rare instance of interacting with an object in a similar way with both species. But it’s not a novel behavior (another component of imitation) if your dog touches other things with his nose. I don’t know about your dog, but it’s practically a default behavior for my three; something that has been heavily reinforced and they offer readily.
Also I am concerned about the dichotomies raised in the language of the studies themselves between “operant conditioning” and “social learning” when they are not separate things. Social learning and social behavior acquisition can comprise operant conditioning and/or classical associations. Everything I have ever seen performed as DAID is definitely operant conditioning. It’s just that the antecedents are different, and sometimes the reinforcers are. That raises some completely unnecessary confusion about learning among many readers. This paragraph has nothing to do with you and your dog; I’m just explaining some concerns I have about the general use of the terminology.
So that’s why I was glad to hear that you are unabashedly using the clicker and food. You are treating it as another fun way to train and communicate with your dog.
Thanks again for the comment.
Thanks Eileen for taking the time to reply! I didn’t know what to expect, but this is fantastic! I’ve read your comment with great interest. Lucid as always.
I haven’t read the discussions about what to call the behaviour. I’ve accepted “social learning” and “imitation” without any further thought. The operant condtioning is clearly a part of it, and the DAID activity seems to be a reinforcer in itself. Simple obedience training with the clicker does not create the same level of enthusiasm in my dog. He is so eager to try! I find that very fascinating. And it’s effective, too. The other day I made up a new task: He should put both his front paws on the second step of a stepladder. This is something he has never done before. I showed him what to do by touching the step with both my hands. In his first attempt he used only one front leg, scratching the lower step. On the third or fourth attempt he put both his front legs on the second step. I’m not shure how to accomplish this in approximately five minuttes, as we did, using other metods.
In my mind there’s no alternative to positive reinforcement, and I promote it as often as I can. The clicker I use mostly when I need the extra precision, otherwise I have a special word I use. Also I try to avoid using a no reward marker, e.g. “No”. If he fails, I don’t say anything, I adjhust the criteria, or, in the case of DAID, I repeat my demonstration and we just make another attempt. Like you say, it’s a fun way to train for both of us. I feel sorry for those who use aversives, and the dogs; they miss out on all of the fun.
I have to stop now. It’s late here in Norway. Thanks again for your time.
Sorry for the delayed response! It sounds like your training is very enriching for both of you. Thanks for describing it!
Comments are closed.