Does Punishing a Dog After a Transgression Really Work?
Data shows that punishing a dog after it misbehaves is ineffective.
Posted May 28, 2013
This is a common scene where people attempt to prevent future bad behaviors on the part of their dogs by punishing them after they have found evidence that the dog has committed some unwanted act. It seems to be a natural human response, but unfortunately people seem to be unaware of the fact that there is scientific evidence that shows that this simply doesn't work. The research was done by Richard L. Solomon and his associates in the psychology department of the University of Pennsylvania way back in 1968 *. Perhaps these results are not often discussed because some people seem to have the idea that only the most recent research is valid, and anything more than a decade or two old must somehow be suspect. Or perhaps these results go unnoticed because they were published as a chapter in a book, rather than in a scientific journal, since most of the library search engines available today only seem to search journal articles. However the research is well controlled, the researchers are highly respected, and the data is important.
The question which Solomon's team was interested in was whether or not punishing a dog after the execution of some action was effective. The setup was really quite simple, and is sometimes called a "taboo training" technique. Two bowls were set in front of a dog (they were all Beagles), one containing meat and the other dried kibble. The dogs were allowed to freely eat the dried kibble but were punished for eating the meat by getting a whack on their snout with a tightly rolled newspaper. In this experiment there were two conditions which are of interest to us, one in which the dog is punished instantaneously when it initially hovers over the meat (call this the no delay punishment group) and one in which the dog gets to actually consume the meat and eat for a period of 15 seconds before it is punished (call this the delayed punishment group). Please notice that this delayed punishment is a rather short interval—only 15 seconds.
Both groups of dogs did acquire the food taboo although the group with the delayed punishment took longer. After both groups had successfully learned to avoid the meat and eat only the kibble the dogs were given a test. The experimenters made sure that the dogs were very hungry and then brought them into a room in which both bowls of food are available. However during the test there was no experimenter visible to enforce the meat eating ban. For the group where the punishment occurred directly before the dogs had the opportunity to complete the action of eating the meat, the taboo worked, and the dogs refrained from eating them meat during 30 days of testing. Contrast this to the group of dogs whose punishment was delayed until only 15 seconds after the act. They ignored the taboo and ate the meat after only two days of testing.
There were significant differences in the behavior of the dogs during the punishment based learning. Dogs in the no delay group quickly learned to avoid eating the meat, but were even a bit hesitant to eat the dry kibble at first. However they quickly got over this fear and later on showed "no obvious signs of fear during the approach to the dried chow and eating it". However the dogs which were trained using the delayed punishment showed lots of negative emotions and fear. According to the researchers they "crawled behind the experimenter or to the wall, urinated, defecated…crawled on their bellies to the experimenter" and so forth even while avoiding the bowl of meat. During the testing the dogs with the delayed punishment acted as if the experimenter were still there, but broke the taboo anyway. However their behavior still showed that they were experiencing conflict and emotional turmoil. According to the experimenters "they ate in brief intervals and appeared to be frightened."
So what can we conclude from this? These results suggest that if you can catch a dog just as he is beginning to initiate an unwanted act, then that immediate punishment may prevent the dog from performing that behavior in the future. However, if you punish a dog after he has already performed the unwanted behavior, or even while he is already committed much of his misbehavior and is in the middle of the act, it simply won't work. Instead the dog learns to be afraid of the entire situation, but will still give into temptation even though his own actions seem to evoke fear in him. Remember that we are really talking about short time intervals between the dog's action and the application of punishment in this study, so imagine how ineffective punishment is when applied to a dog a minute after his transgression, let alone 15 minutes or an hour later when you discover what naughtiness your dog has performed while out of your sight.
So if punishment doesn't work, how should my friend go about preventing his dog from snatching and eating the dinner rolls that he has left on his counter? The next day I gave him a breadbox with a hinged lid — problem solved!
* Solomon, R.L., Turner, L. H., and Lessac, M.S. (1968). Some effects of delay of punishment on resistance to temptation in dogs. In: Walters, R. H., Cheyne, J.A. and Banks, R.K., eds. Punishment, Penguin, London, pp. 124-135.
Stanley Coren is the author of many books including: Born to Bark; Do Dogs Dream? The Modern Dog; Why Do Dogs Have Wet Noses? The Pawprints of History; How Dogs Think; How To Speak Dog; Why We Love the Dogs We Do; What Do Dogs Know? The Intelligence of Dogs; Why Does My Dog Act That Way? Understanding Dogs for Dummies; Sleep Thieves; The Left-hander Syndrome
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission