Welcome to The Dog Trainer’s Quick and Dirty Tips for Teaching and Caring for Your Pet. I’m your host, Jolanta Benal. So, you come home after a long day at work to find your couch stuffing all over the floor, and ditto the kitchen garbage. “Rover!” you exclaim in dismay and anger. Sure enough, Rover slinks over to you, belly low, tail tucked, ears pinned back. Obviously, he knows he did wrong.

Or does he? This week’s topic is a nifty experiment and what it may teach us about doggy guilt.

Learning a new technology can be daunting.  Go To My PC, not so much. It’s built for non-technical people, so everyone can enjoy the freedom of working from anywhere, anytime. Setup is completely automatic. All you have to do is download the software, and you’re ready to go. Plus, free customer service is available 24/7 if you need it. Try Go To My PCFREE for 30 days! Visit GoToMyPC.com/podcast. That’s GoToMyPC.com/podcast podcast for a free 30-day trial.

Alexandra Horowitz, of the Barnard College psychology department, designed an experiment to test whether dogs act guilty because they’ve done something their owner disapproves of, or whether they’re responding to our voice and body language.

How the Experiment Worked

Owners first had to demonstrate that their dog would perform two behaviors: sit and stay for 10 seconds and-- when told not to eat a treat-- leave the treat alone for 10 seconds. Then, in the experiment, each owner asked his or her dog to sit and stay. They showed their dog a treat and told the dog to leave the treat alone. The owner then set the treat on the floor, in a spot where the dog could see it but not reach it. And then the owner left the room for 20 seconds (the experimenter stayed).

During that time, one of two things happened. The experimenter either picked up the treat and handed it to the dog -- who ate it; what a surprise! – or the experimenter took the treat away. Then the owner came back in and the experimenter told him whether his dog ate the treat. If the experimenter said that the dog ate the treat, the owner scolded the dog. If the experimenter said that the dog didn’t eat the treat, the owner just greeted the dog normally.

Now, here’s the punch line -- sometimes the experimenter lied. She told the owner that the dog had eaten the treat when, actually, the dog hadn’t eaten it. Or vice versa.

Scolding Made Dogs Act Guilty

The dogs offered significantly more “guilty” behaviors when they were scolded, regardless of whether they’d eaten the treat. And -- drumroll -- dogs who didn’t eat the treat showed as many guilty-looking behaviors as dogs who did eat it. In other words, how guilty the dogs acted had no connection with their actual “guilt.” In fact, the dogs who acted guiltiest of all were the “innocent” ones being scolded.

Guilty or Fearful?

You won’t be surprised by any of the actions on the list of guilty-looking behaviors. Among other things, dogs dropped to the ground and showed their bellies; they pinned their ears back and looked away from their owners; their tails dropped low; they avoided their owners. Dr. Horowitz, the researcher, points out in her paper that the list of “guilty” behaviors overlaps with the array of behaviors canid ethologists associate with fear and submission. Since the dogs acted “guiltier” when scolded, regardless of what they’d actually done, Dr. Horowitz suggests that they may have been offering submissive behaviors because the scolding made them expect a punishment. “What the guilty look may be,” she writes, “is a look of fearful anticipation of punishment.” And check this out. Three of the owners in the study had a history of using physical reprimands -- forcing the dog down to the ground, grabbing them, even hitting them. Their 3 dogs were among the 4 who offered the highest rate of “guilty” behavior.

Now, the study had only 14 subjects in all. (This is typical of behavioral studies, by the way – few of them include large numbers of dogs.) We certainly can’t draw firm conclusions from such a small sample, much less from the behavior of 3 dog-owner pairs. And the study doesn’t prove that dogs don’t feel guilty. But it does strongly suggest that actual “guilt” -- whatever “guilt” is if you have a forebrain as small as a dog’s -- and looking “guilty” aren’t linked. Fear might have more to do with it.

Anger’s a Time-Waster

As Dr. Horowitz points out, some trainers still advise us to communicate with our dogs as if they understood human moral codes. But probably dogs don’t have that understanding -- or, at least, it doesn’t look as if they do. What if we’re blaming our dogs for doing wrong when they don’t know right from wrong? We’re expecting too much of them and, probably, wasting our time and theirs in being angry.

So what’s your take-home for this week? Just this: next time you’re super double sure your dog knows that she did wrong, and knows what she did wrong, take a deep breath and think again. Maybe you need to train a little more, or change something about your dog’s daily schedule, or get professional help with a behavior problem. Whatever’s going on, a good first step in dealing with it is always going to be “Turn off your mad.”

The citation for Dr. Horowitz’s study appears on my site at dogtrainer.quickanddirtytips.com, along with some suggestions for further reading. Don’t forget to visit Go To My PC dot com for your free 30-day trial. And I would love to hear from you; your comments and questions help me shape future episodes. Email me at dogtrainer@quickanddirtytips.com, phone 206-600-5661, or visit me on Facebook – just search on The Dog Trainer. Bye, and thank you for listening!

References and Further Reading

Hauser, Marc. Wild Minds:What Animals Really Think. Holt, 2000. See especially chapter 9, “Moral Instincts.”

Horowitz, Alexandra. Disambiguating the “guilty look”: Salient prompts to a familiar dog behaviour. Behavioural Processes 81 (2009): 447-52.

McConnell, Patricia B. For the Love of a Dog: Understanding Emotion in You and Your Best Friend. Ballantine Books, 2006.

“What Really Prompts the Dog’s ‘Guilty Look.” ScienceDaily, June 14, 2009. Available at http://www.sciencedaily.com/releases/2009/06/090611065839.htm.