This is the single most common question I get on a first session, and it usually comes with a little embarrassment attached — like using food is a shortcut somebody will catch them at.
Here is how I answer it. Nobody keeps going to a job that stopped paying. Your dog is the same, and so are you. What changes over time is not whether your dog gets paid, but what the paycheck is, how often it arrives, and how predictable it looks.
What reinforcement actually is
Reinforcement is anything that follows a behavior and makes it more likely next time. That is the whole definition. It is not a bribe, a treat, or a training gimmick — it is the mechanism behavior runs on, whether or not you are deliberately using it.
Your dog is being reinforced all day long without you. Barking at the window makes the delivery driver leave. Pulling on the leash makes the tree get closer. Counter surfing pays in sandwich. If you are not choosing what gets reinforced, the environment is choosing for you.
Bribery is a timing problem, not a food problem
A bribe comes out before the behavior: you show the cookie, then ask for the sit. A reward comes after: you ask, your dog does it, then the food appears from a pouch, a pocket, the counter. Same food, completely different learning. Get the food out of your hand within the first few sessions and the accusation of bribery goes away on its own.
Health and welfare notes before you start
Reinforcement fails when needs are unmet
Before you blame the schedule, look at the dog. A dog who has not slept, has not sniffed anything all week, or is over threshold in a loud place will not take food — and that is not a reinforcement problem, it is a needs problem.
Bark For Attention
Check these first
Sleep
Under-rested dogs get frantic and then get labeled stubborn. Most adult dogs need 12–16 hours of rest a day.
Sniffing
A daily decompression walk on a long line does more for training than any drill.
Hunger
Train before meals, not after. A full dog is a dog who has already been paid.
Set the game up so paying is easy
Build a currency list for your dog
Rank rewards from everyday to jackpot: kibble, training treats, cheese, chicken, tug, a thrown ball, permission to sniff a lamppost. You need at least three tiers, and the top tier is reserved for hard work.
Get food off your body
Treat jars by the door, on a shelf in each room, in the car. Your dog should never be able to predict a reward by looking at your hands.
Use a marker
A clear “yes” or a clicker buys you time. The marker ends the behavior and promises payment, so the food can arrive two seconds later from anywhere.
Keep the environment honest
If the trash can pays better than you do, put a lid on it. Management removes the competing reinforcement so your paycheck can win.
What you actually want instead
The goal is not a dog who works for nothing. The goal is a dog whose good behavior is maintained by the ordinary flow of life: doors, dinner, leashes, play, freedom, attention — with food dropped in unpredictably to keep the whole system strong.
The step-by-step plan
1. Acquisition: pay every single rep
While a behavior is brand new, one rep equals one reward, every time. This is a continuous schedule and it is the fastest way to build a behavior. Do not get clever here — skimping at this stage is why behaviors fall apart later.
2. Check fluency before you change anything
A behavior is ready to move on when your dog offers it quickly, correctly, on the first cue, in at least three different rooms. If you have to repeat the cue, you are still in step one.
3. Shift to variable reinforcement
Now pay roughly two out of three reps, chosen at random — never in a pattern your dog can count. Unpredictable payment is what makes behavior durable. Slot machines work on this principle and so do dogs.
4. Add differential reinforcement
Pay the good reps small and the great reps big. The fastest sit, the straightest heel, the recall away from a squirrel gets three pieces of chicken in a row. Now you are not just maintaining behavior, you are improving it.
5. Layer in real-life rewards
Sit before the door opens — the walk is the reward. Recall away from the fence, then release back to sniff. Loose leash for twenty feet, then permission to investigate. This is where the treat pouch quietly stops mattering.
6. Keep a maintenance schedule
Even a five-year-old behavior needs a paycheck sometimes. I tell clients: pay recall every single time, forever. Pay everything else often enough that your dog never has to wonder.
7. Repay whenever criteria go up
New environment, more distraction, longer duration, more distance — any of these means you are back near step one for that context. Go back to paying every rep for a few sessions, then thin out again.
8. Never quit cold turkey
Yanking reinforcement out suddenly produces an extinction burst: the behavior gets worse and more frantic before it disappears. Thin gradually, and if performance drops, pay more for a week and rebuild.
Where this fits in the Training Transformation Foundation™
Enrichment doubles as reinforcement
- Scatter feeding in grass after a good walk — pays the behavior and meets a foraging need at the same time.
- Snuffle mats and food puzzles fed as a reward for settling on a mat.
- A chew delivered after a calm greeting at the door.
- Permission to sniff a fascinating patch, given as a release cue after loose-leash walking.
- Two minutes of tug as the paycheck for a hard recall in the yard.
- Freedom on a long line — for most dogs, the biggest reward you own.
Common mistakes and troubleshooting
- Fading too fast. If the behavior is getting slower or needs a second cue, you thinned the schedule before it was fluent.
- Showing the food first. That is a bribe, and it builds a dog who checks your hands before deciding.
- Using the same boring treat for everything. Kibble does not buy a recall past a squirrel.
- Predictable patterns — paying every third rep exactly. Dogs count better than you think. Randomize.
- Feeding for free all day. If attention, food and access are unconditional, none of them work as rewards.
- Punishing the behavior you never paid. A dog cannot choose a behavior that was never built.
- “He only does it when I have food.” That is a fading problem with a clean fix: food off your body, marker first, pay from a jar across the room.
- “She won't take treats outside.” That is a threshold problem. Get further from the trigger, use higher-value food, and shorten the session.
When to get professional help
Go deeper
Print & keep
Your reinforcement checklist
- Currency list written out in three tiers
- Food stored off your body, in jars around the house
- A clear marker word used before every reward
- Every rep paid while a behavior is new
- Fluency confirmed in three rooms before thinning
- Random, unpredictable payment once fluent
- Bigger payment for better reps
- Real-life rewards named and used daily
- Recall paid every single time, for life
- Back to continuous payment whenever the picture gets harder
