Dog Steals Food from the Table: Why It Happens and What Really Helps

In brief

A dog that takes food from the table or the kitchen counter isn’t being naughty. He is doing what pays off for him. Four things matter. First: food is a very powerful reinforcer, and every successful grab rewards the behavior; occasional successes in particular make it persistent. Second: in experiments, dogs took forbidden food more often when the person wasn’t looking, when a barrier blocked her view or when the food was in the dark. Watching alone is therefore not enough. Third: the so-called guilty look doesn’t tell you whether your dog has stolen something; in studies it was mainly a response to the person’s scolding. Fourth: what helps most is management, meaning nothing within reach, combined with reward-based training.

This guide shows why dogs steal food, what research says about it, when to think of a health cause, how to make your kitchen safe and how to build a behavior that pays off more than grabbing the plate.

1. Why dogs steal food

1.1 Food is a very powerful reinforcer

A dog doesn’t have to learn to like food. It is what’s called a primary reinforcer. In one study, dogs could choose during short sessions between going to a person who petted them and going to a spot where food was available. As long as food was readily available, most dogs chose the food; when it became scarcer, many shifted their time to petting, with clear differences between individual dogs (Feuerbacher & Wynne, 2014).

Dogs differ here. In Labrador retrievers, Raffan et al. (2016) found a variant in the so-called POMC gene that was associated with higher body weight, more body fat and greater food motivation. Among the other breeds examined, it occurred only in the closely related flat-coated retriever. That doesn’t explain every Labrador that helps himself to the cake, but it shows that some dogs find it harder than others to leave food alone.

In practice: on top of that, dogs naturally search for food and take opportunities. From your dog’s point of view, a piece of cheese on the cutting board isn’t a forbidden object but a find.

1.2 Every success counts, and occasional successes count most

A dog who once snatched a sandwich from the counter was rewarded for jumping up. He will try again. Most of the time he finds nothing, but sometimes he does, and exactly this pattern makes behavior persistent. Learning research has long described that behavior reinforced only some of the time persists longer when reinforcement stops than behavior reinforced every time (Humphreys, 1939). Amsel (1962) explained this partly by suggesting that, in such a learning history, carrying on after disappointment is itself learned.

If your dog finds something only every tenth day, that’s no reason for him to stop. That’s why it matters so much that he stops finding anything at all.

👉 Reinforcement schedules in dogs

1.3 Opportunity makes the thief

In practice: most thefts happen at typical moments, for example when you go to the door while dinner is on the table, when the groceries are still on the floor or when the plate of meat sits at the edge of the counter while you cook. Many dogs learn very precisely which situations pay off.

2. Dogs steal when no one is watching

Many owners report that their dog never takes anything while they’re in the room but grabs it as soon as they turn around. This exact situation has been studied in experiments several times.

2.1 Eyes open, back turned, out of the room

Call et al. (2003) forbade twelve dogs, over a series of trials, to take a visible piece of food. In some trials the person looked at the dog throughout. In others she left the room, turned her back, busied herself with something else or closed her eyes. In most of the conditions in which the person wasn’t watching, the dogs behaved clearly differently than when she was. When the person looked at them, the dogs took less food, approached it in a more indirect way and sat more often instead of lying down.

2.2 When a barrier blocks the view

Bräuer et al. (2004) went a step further and placed barriers between the person and the food. Whether the dogs were more likely to take the forbidden food depended on whether the barrier actually blocked the person’s view, depending on its orientation, its size or whether it had a window. In another part of the same paper, however, the authors found no evidence that dogs know what they themselves have seen.

2.3 Stealing in the dark

In Kaminski et al. (2013), a person always forbade the dog to take a piece of food; what varied was whether the area around the person or around the food was lit. The dogs stole significantly more food when it was dark than when it was light. What mattered was the lighting around the food, not around the person. The authors consider it possible that dogs take into account whether the person can see the food. Whether real understanding or a simpler rule lies behind this has not been settled.

2.4 What this means for your kitchen

For everyday life, the interpretation is secondary. What matters is the pattern: dogs adjust their behavior to whether they are being watched. A dog who takes nothing while you’re there may simply have learned that it doesn’t pay off while you’re there. That’s why management matters more than watching: you can’t look every second, but you can make sure there is nothing to find.

3. The guilty look gives nothing away

You walk into the room, the butter is gone, and your dog crouches, flattens his ears and looks away. For many owners, that’s proof that he knows what he did. Research sees it differently.

Horowitz (2009) filmed fourteen dogs whose owner forbade them a treat before leaving the room. The owner was then told, correctly or incorrectly, whether the dog had obeyed. Guilty-look behavior did not differ according to whether the dog had eaten the treat, but it appeared markedly more often when the owner scolded, most strongly in dogs who had obeyed and were scolded anyway.

Two further studies fit this picture. In Hecht et al. (2012), sixty-four dogs could take food from a table while the owner was outside. Dogs that had eaten did not show guilty-look behavior more often afterward than dogs that had refrained, and owners could not tell from their dog’s greeting whether he had broken the rule. In Ostojić et al. (2015), owners behaved neutrally, and their judgments were not correct more often than chance.

Scolding after the fact therefore doesn’t help. What your dog is more likely to learn is that your return can turn unpleasant, not that the table is off limits.

👉 The guilty look in dogs

4. When interest in food suddenly increases

If a dog has always stolen, it’s usually a matter of learning history and opportunity. It’s different when a dog who never touched anything for years suddenly searches for food all the time or raids the trash can.

In practice: sometimes there is something simple behind it, such as too small a food ration, a change of diet or a weight-loss diet. A markedly increased appetite can also be related to illness or medication, though. If your dog suddenly searches for food much more than usual, drinks more, loses or gains weight, have it checked by your veterinarian before you think about training.

5. Why stealing can be dangerous

Some things on the table can be dangerous for dogs. Well-known hazards include, for example, chocolate, grapes and raisins, onions and garlic, and candy or chewing gum containing the sweetener xylitol.

In practice: there are also things that aren’t toxic but are dangerous, such as cooked bones, skewers, corn cobs or packaging that gets swallowed too. If you suspect your dog has gotten hold of something dangerous, don’t wait; call your veterinary practice or the emergency service right away, and bring the packaging if you can.

👉 Foods toxic to dogs

6. Management comes first

Because every successful grab reinforces stealing, and because dogs grab precisely when you aren’t looking, management is not a stopgap but the foundation. As long as training is under way, your dog must not keep succeeding on the side.

6.1 Nothing within reach

In practice: the most important rule is also the simplest: food is never left unattended where your dog can get to it. That applies to the table edge and the counter, but also to the coffee table, grocery bags and the trash can. Push food to the back or put it in the cupboard, and clear the table right after the meal. Many owners underestimate how high their dog can reach on his hind legs.

6.2 Baby gates and crates

In practice: when food can’t be put away, for example when cooking for guests or at family meals, physical separation helps. A baby gate at the kitchen door lets your dog see and hear what’s going on but keeps him away from the counter. A crate can help too, as long as your dog knows it as a pleasant place to retreat and not as punishment.

6.3 Plan the critical moments

In practice: think about when most stealing happens in your home, and plan exactly those moments. If the doorbell rings while food is on the table, take your dog with you or send him behind the gate beforehand. Children have fixed places where they eat, and with guests who like to feed from the table, separation is the rule for that evening.

7. Reward-based training

Management prevents stealing from being reinforced any further. Training makes sure your dog learns something else that pays off. The studies cited here do not directly test this training approach for counter-surfing in family dogs. The following steps are one possible practical approach from training.

7.1 Mat training while you cook and eat

In practice: the best alternative to stealing is a place where your dog lies relaxed while you cook or eat. Build mat training first in calm moments, with no food on the table. Reward your dog for lying down on the mat and staying there, briefly at first, then longer, then with you a little farther away. After that you practice in the kitchen, first while making coffee, then while making sandwiches, and finally while really cooking.

In practice: a chew or a stuffed food toy on the mat gives your dog something of his own to do during family meals. Count the rewards as part of his daily ration.

👉 Teaching your dog to settle

7.2 Rewarding four paws on the floor

In practice: many owners only react once the dog already has his front paws on the counter. Instead, watch for the moments when he stands next to you with all four paws on the floor, and reward exactly then, ideally with a treat placed on the floor. That way he learns that down there pays off more than up there. Later you reward less often, but not never.

7.3 A reward-based “leave it”

In practice: a cue like “leave it” is often used as a threat. It becomes more reliable when built with reward. Put a treat in your closed hand. As soon as your dog stops sniffing and turns away or looks at you, he gets a better treat from your other hand. Once that works reliably, you say the cue just beforehand. Then the treat lies openly on the floor, later on a stool, eventually on the table. The reward always comes from you, never from the forbidden food.

In practice: use the cue in everyday life only once you’re sure it works. A “leave it” that is followed by the sandwich after all weakens the cue.

👉 Dog eats everything outside

7.4 Reinforcing alternatives instead of just forbidding

Behind these exercises lies one principle: you deliberately reward a behavior that takes the place of the unwanted one. The technical term is differential reinforcement of alternative behavior. A review of 116 studies rated this approach as well established (Petscher et al., 2009). That work comes from people, though, mostly people with developmental disabilities; no comparable body of research exists in dogs.

In practice: lying on the mat, four paws on the floor or a glance at you should pay off more than any attempt at the table. That only works if stealing stops paying off at the same time.

7.5 Stay realistic

In practice: a dog with a long history of stealing won’t pass up a roast left unattended on the table after two weeks. The goal is an everyday life in which he lies calmly on his spot while you cook and eat, and a kitchen in which there is nothing to take. For very food-motivated dogs, management remains part of the solution for good.

8. What about punishment?

A loud shout, slapping the table or so-called booby traps such as rattling cans at the table edge can stop stealing in the moment. The question is what the dog learns from it, and at what cost.

The studies on stealing under observation (Call et al., 2003; Bräuer et al., 2004; Kaminski et al., 2013) suggest that dogs pay very close attention to whether a person is looking. If the dog is punished only in your presence, he may mainly learn that stealing is unpleasant when you’re around, and steal when you’re not. Rattling traps can also trigger fear of the kitchen, of noises or of you, and that then affects everything your dog experiences there.

The research on training methods points in the same direction. In a study of 92 dogs from seven training schools, dogs from schools using aversive methods showed more stress behavior, larger cortisol increases and more pessimistic judgment (Vieira de Castro et al., 2020). In an owner survey, more frequent use of punishment went along with more reported problem behavior (Hiby et al., 2004). A review found no evidence that aversive methods are more effective than reward-based training, but did find indications of risks to welfare (Ziv, 2017).

In practice: if you catch your dog in the act, calmly interrupting is fine: his name, a recall, a reward for coming. Then put the food away and ask yourself how it came to this.

👉 Punishment fallout in dogs

9. A note on the evidence

There is hardly any targeted research on stealing from the table in everyday life. The studies on stealing under observation (Call et al., 2003; Bräuer et al., 2004; Kaminski et al., 2013) are laboratory experiments on social cognition with small groups, twelve dogs in Call et al. (2003). They show that dogs adjust their behavior to whether they can be seen, but not how to stop stealing at home, and their interpretation is debated. The guilty-look studies are also small (fourteen dogs in Horowitz, 2009) but reach a consistent picture with different designs (Hecht et al., 2012; Ostojić et al., 2015).

The choice study (Feuerbacher & Wynne, 2014) concerns one particular laboratory situation, the gene variant (Raffan et al., 2016) only Labrador and flat-coated retrievers, and it shows an association, not an automatic outcome. The foundations on intermittent reinforcement come from classic learning research (Humphreys, 1939, studied an eyelid reflex in humans), and the review on differential reinforcement (Petscher et al., 2009) concerns people. The studies on training methods (Vieira de Castro et al., 2020; Hiby et al., 2004; Ziv, 2017) concern training in general and can only be transferred with caution; the survey by Hiby et al. (2004) shows associations, not causes. The advice on management, health and training steps is therefore marked as practical experience.

10. Conclusion

A dog who steals food from the table follows a simple logic: food pays off, and every success, even a rare one, keeps the behavior alive. Because dogs pay close attention to whether you’re looking, catching him in the act isn’t enough. It works better in the long run to remove the opportunities and build a behavior that pays off more.

The three things that make the difference:

Nothing within reach. No food left unattended at the table edge, on the counter or in the trash can, and a baby gate if needed.

Reward an alternative. Mat training while you cook and eat, four paws on the floor and a reward-based “leave it.”

Take causes seriously. A sudden increase in appetite belongs at the veterinarian, and scolding after the fact doesn’t help.

Key takeaways on dogs stealing food from the table

Food is a very powerful reinforcer. As long as food was readily available, most dogs chose food over petting (Feuerbacher & Wynne, 2014), and in Labrador and flat-coated retrievers a gene variant was associated with greater food motivation (Raffan et al., 2016).

Occasional successes make it persistent. Behavior reinforced only some of the time persists longer when reinforcement stops (Humphreys, 1939; Amsel, 1962).

Dogs tend to steal when unobserved. Dogs took forbidden food more often when the person wasn’t looking, a barrier blocked her view or the food was in the dark (Call et al., 2003; Bräuer et al., 2004; Kaminski et al., 2013).

The guilty look is not a confession. It depended on the owner’s scolding, not on whether the dog had eaten, and owners could not detect rule-breaking (Horowitz, 2009; Hecht et al., 2012; Ostojić et al., 2015).

Reinforcing alternatives is well supported, in people. A review of 116 studies rated differential reinforcement of alternative behavior as well established (Petscher et al., 2009).

Punishment has its price. Dogs from training schools using aversive methods showed more stress behavior and larger cortisol increases (Vieira de Castro et al., 2020).

Working on a behavior problem with your dog? Our step-by-step dog behavior workbooks cover leash reactivity, being left alone, resource guarding, two-dog households, barking at passersby and noise fear.

References

• Amsel, A. (1962). Frustrative nonreward in partial reinforcement and discrimination learning: Some recent history and a theoretical extension. Psychological Review, 69(4), 306–328. https://doi.org/10.1037/h0046200
• Bräuer, J., Call, J., & Tomasello, M. (2004). Visual perspective taking in dogs (Canis familiaris) in the presence of barriers. Applied Animal Behaviour Science, 88(3–4), 299–317. https://doi.org/10.1016/j.applanim.2004.03.004
• Call, J., Bräuer, J., Kaminski, J., & Tomasello, M. (2003). Domestic dogs (Canis familiaris) are sensitive to the attentional state of humans. Journal of Comparative Psychology, 117(3), 257–263. https://doi.org/10.1037/0735-7036.117.3.257
• Feuerbacher, E. N., & Wynne, C. D. L. (2014). Most domestic dogs (Canis lupus familiaris) prefer food to petting: Population, context, and schedule effects in concurrent choice. Journal of the Experimental Analysis of Behavior, 101(3), 385–405. https://doi.org/10.1002/jeab.81
• Hecht, J., Miklósi, Á., & Gácsi, M. (2012). Behavioral assessment and owner perceptions of behaviors associated with guilt in dogs. Applied Animal Behaviour Science, 139(1–2), 134–142. https://doi.org/10.1016/j.applanim.2012.02.015
• Hiby, E. F., Rooney, N. J., & Bradshaw, J. W. S. (2004). Dog training methods: Their use, effectiveness and interaction with behaviour and welfare. Animal Welfare, 13(1), 63–69. https://doi.org/10.1017/S0962728600026683
• Horowitz, A. (2009). Disambiguating the “guilty look”: Salient prompts to a familiar dog behaviour. Behavioural Processes, 81(3), 447–452. https://doi.org/10.1016/j.beproc.2009.03.014
• Humphreys, L. G. (1939). The effect of random alternation of reinforcement on the acquisition and extinction of conditioned eyelid reactions. Journal of Experimental Psychology, 25(2), 141–158. https://doi.org/10.1037/h0058138
• Kaminski, J., Pitsch, A., & Tomasello, M. (2013). Dogs steal in the dark. Animal Cognition, 16(3), 385–394. https://doi.org/10.1007/s10071-012-0579-6
• Ostojić, L., Tkalčić, M., & Clayton, N. S. (2015). Are owners’ reports of their dogs’ ‘guilty look’ influenced by the dogs’ action and evidence of the misdeed? Behavioural Processes, 111, 97–100. https://doi.org/10.1016/j.beproc.2014.12.010
• Petscher, E. S., Rey, C., & Bailey, J. S. (2009). A review of empirical support for differential reinforcement of alternative behavior. Research in Developmental Disabilities, 30(3), 409–425. https://doi.org/10.1016/j.ridd.2008.08.008
• Raffan, E., Dennis, R. J., O’Donovan, C. J., Becker, J. M., Scott, R. A., Smith, S. P., Withers, D. J., Wood, C. J., Conci, E., Clements, D. N., Summers, K. M., German, A. J., Mellersh, C. S., Arendt, M. L., Iyemere, V. P., Withers, E., Söder, J., Wernersson, S., Andersson, G., … O’Rahilly, S. (2016). A deletion in the canine POMC gene is associated with weight and appetite in obesity-prone Labrador retriever dogs. Cell Metabolism, 23(5), 893–900. https://doi.org/10.1016/j.cmet.2016.04.012
• Vieira de Castro, A. C., Fuchs, D., Morello, G. M., Pastur, S., de Sousa, L., & Olsson, I. A. S. (2020). Does training method matter? Evidence for the negative impact of aversive-based methods on companion dog welfare. PLOS ONE, 15(12), Article e0225023. https://doi.org/10.1371/journal.pone.0225023
• Ziv, G. (2017). The effects of using aversive training methods in dogs – A review. Journal of Veterinary Behavior, 19, 50–60. https://doi.org/10.1016/j.jveb.2017.02.004

Note: This guide does not replace individual advice. If your dog suddenly has a much bigger appetite, his weight or drinking changes, or he may have eaten something dangerous, contact your veterinarian.

Frequently asked questions about dogs stealing food from the table

Causes, management and training