Training Framework

Fading Without Losing Behavior

You can't reinforce every sit forever — and you shouldn't. Intermittent schedules produce behavior that lasts. But jump from every-time to random and you'll watch the behavior fall apart in a week. Thinning is the bridge, and the steps matter.

Two Dimensions at Once

Schedule thinning reduces how often reinforcement occurs: CRF → FR2 → FR3 → VR3 → VR5 → real-world variable. Reinforcer thinning changes what the reinforcement is: food → food paired with praise → praise alone → intermittent praise → life rewards.

Effective thinning adjusts both dimensions gradually — but never both at once.

Why Variable Beats Fixed

Variable-ratio (VR) schedules produce higher, steadier response rates and substantially greater resistance to extinction than fixed-ratio schedules. A dog on VR5 doesn't know which repetition will pay, so it responds consistently on every trial. This is why the progression moves from fixed to variable before thinning further.

A Practical Thinning Progression

CRF

Every sit produces a treat. Continue until fluent across 3+ sessions.

FR2 → FR3

Treat every second sit (with a verbal marker on the unreinforced one), then every third.

VR3 → VR5

Average every third, but vary — sometimes second, sometimes fourth. Then average every fifth. The unpredictability is the mechanism.

Real-world variable

Life rewards replace the treat pouch: door opens, leash clips on, food bowl goes down — plus occasional food.

Recognizing Ratio Strain

Ratio strain is the behavioral breakdown that occurs when the schedule gets too lean too fast. Observable signs: increased latency after the cue, emotional behavior (whining, barking, pawing), incomplete responses, displacement behaviors (ground-sniffing, body shaking), or leaving the training area entirely.

The 80% Rule

If accuracy drops below 80%, the schedule is too lean. Return to the previous step, rebuild for 2-3 sessions, then thin again in a smaller increment. Going backward is not failure — it's correct schedule management.

The "Random Reinforcement" Trap

When trainers jump from CRF to "I'll just reinforce randomly," the result is not a variable schedule — it's functionally an extinction schedule with occasional reinforcement. The dog experiences long, unpredictable stretches without reinforcement and develops frustration-based behaviors. True VR thinning is systematic and gradual, not random.

Every Dog Is Different

The right approach depends on why your dog does this — and that varies by temperament, history, and environment. The Synchrony coach can tailor these principles to your dog's specific behavior profile.

Sources: Ferster & Skinner (1957), Cooper, Heron & Heward (2020), Vicars et al. (2014). From the Data Dogs research brief: Reinforcement in Practice.