In this article
Your dog sits beautifully while you hold a biscuit, then appears to lose the instruction when the biscuit goes back in the tin. It is a familiar worry: have you taught a skill, or negotiated a snack contract? Dogs Trust recommends rewards as a way to make useful behaviour more likely to happen again.1
The practical issue is how food fits into the sequence. You can teach a dog to respond without first showing a treat, while continuing to reward the response. Those are different decisions, and separating them makes the training much easier to think about.
Watch the order of events
Dogs Trust's sit exercise illustrates a lure: food moves above the dog's nose to encourage the position. Once the dog understands, repeat the movement without food in that hand and still reward the sit. Add the spoken cue when the response is reliable. The visible food is a temporary teaching aid.2
Here is a useful way to describe what you see at home. These are everyday descriptions, not three scientific diagnoses. The question is what your hands and voice are doing before the dog responds, and what happens afterwards. Watch one familiar exercise without changing anything yet.
| Arrangement | What happens | Useful question |
|---|---|---|
| Lure | Visible food guides the movement.2 | Can the gesture later be empty-handed? |
| Reward | The response is followed by something the dog wants.1 | Was it delivered for the intended behaviour? |
| Last-minute negotiation | You repeatedly offer more to get a response. | What did the whole sequence look like? |
Calling everything bribery hides those differences. You might decide your next goal is just one successful response to the familiar empty-hand gesture. That is a more useful practice note than deciding your dog has become commercially minded. There is no need for a moral debate with the biscuit tin.
What the training research found
A UK study published in 2020 compared three groups of 21 dogs referred for off-lead problems. The authors reported better first-cue responses to sit and come in the group taught by reward-focused trainers. Preliminary observations estimated food rewards were used about five times more often in that group.3
This was a comparison of training approaches, not a test in which only treat delivery changed. Different trainers taught the reward-focused group in a different season, and dogs were usually on long leads near the trainer. The findings cannot establish that extra food alone caused the differences or guarantee off-lead reliability.3
A published commentary also challenged group comparability, statistical analysis and aspects of the final training day. That disagreement matters when judging how much weight one study can carry. It is reasonable to discuss the reported results while keeping the practical conclusion narrower than a claim that one feeding schedule has been proved best.4
Another experiment, published in 2023, studied ten pet dogs performing a muzzle-touch task for their preferred food or leisure item. Food generally supported more responding, including when progressively more touches were required. Leisure items also worked as rewards for many dogs compared with receiving nothing.5
The sample was small, the setting was familiar and quiet, and the task was one trained action. Food and leisure rewards were tested in separate conditions, not offered as a simultaneous choice. It would be a large leap from that result to saying your dog never wants a game, or values dinner more than your company.5
Choose something worth working for
Battersea describes food, toys, spoken praise and physical affection as possible training motivations. Preferences vary, and different foods or toys can have different value to the same dog. Watch the response to what you offer instead of assuming that a packet labelled training treats must automatically be the ideal currency.6
Its practical guidance recommends more valuable rewards for new skills and distracting places. It also suggests matching reward type to the aim: play may suit an energetic recall, while food can suit calm settling. These are guidance choices to try with the individual dog, not a ranking that applies in every situation.6
For your own notes, choose one easy behaviour and list what you currently offer afterwards. Does your dog engage with that reward? What changes between the kitchen and the garden? This small observation exercise is a suggested starting point for a conversation with your trainer, especially if a previously straightforward exercise has become a struggle.
Remove the display before removing the reward
Return to the sit example: Dogs Trust keeps the reward after the empty-hand gesture succeeds. Only later does the spoken cue come before the gesture. Changing what is in your signalling hand is different from stopping payment for a response the dog is still learning.2
Dogs Trust recommends rewarding every success at first, then varying food, praise and other rewards once the skill works reliably in different environments. If you use a trained marker word or click, follow that marker with its promised reward. The marker is information about success, not a substitute for everything the dog enjoys.1
You could put the following notes beside your practice area. They are prompts for observing your own session, rather than a test your dog has to pass. Keep the task modest enough that you can describe what happened without relying on words such as stubborn, greedy or disobedient.
Questions for the next practice
- What signal happened before the response?
- Was food visible in the signalling hand?
- What response did I reward?
- What was the place like compared with our usual practice area?
- What would I like to ask our trainer?
Keep the food budget visible too
PDSA advises weighing daily portions and making sure food is not being added unnoticed by other people. Reduce meals to account for treats. Setting some of the normal food aside for suitable practice is one way to organise that accounting, while keeping the complete daily amount in view.7
For example, a labelled container for the day's training portion can give everyone the same reference point. That is a household suggestion, not an individual nutrition prescription. It is particularly useful to discuss the plan with your vet when you need help tailoring feeding; PDSA recommends veterinary support for an appropriate feeding plan.7
Food rewards do not require a constant procession of rich extras. Think about what you are offering, how much has already been used and whether the reward is useful for this particular task. Good bookkeeping is less glamorous than buying another pouch, but it fits in a kitchen drawer.
Keep boundaries kind and get help when needed
The veterinary behaviour organisation AVSAB recommends reward-based methods for everyday training and behaviour problems. Its guidance includes structure, routines and boundaries, taught without fear, pain or intimidation. Using rewards does not mean abandoning rules or letting the dog rehearse every activity they happen to enjoy.8
If training has become confusing, ask a reward-based professional to watch the sequence. For aggression, worsening anxiety or a sudden change, start with a vet assessment and qualified behaviour support. AVSAB emphasises evaluating health as part of behaviour care and referring when the problem is beyond the trainer's expertise.8
You do not need to prove your dog's loyalty by withholding a reward. Give the exercise a clear signal, a worthwhile outcome and a difficulty you can both manage, then build from the response you actually see.
Sources and further reading8 sources
The small numbers in the article link to the source behind each fact. Sources open on their own websites.
- Positive reinforcement – training with rewards
- How to train your dog to sit
- Efficacy of Dog Training With and Without Remote Electronic Collars vs. a Focus on Positive Reinforcement
- Commentary: Efficacy of Dog Training With and Without Remote Electronic Collars vs. a Focus on Positive Reinforcement
- Efficacy of Edible and Leisure Reinforcers with Domestic Dogs
- Training motivations
- Your dog’s diet
- Position Statement on Humane Dog Training
