Home » No we don’t need to feed every step with Positive Reinforcement!

No we don’t need to feed every step with Positive Reinforcement!

Behaviour Chains:

A few really common questions around this subject are: Do I have to feed every step, for every R+ behaviour? Or how do I use Positive Reinforcement during continuous work when I can’t stop and feed every step? Or How do I avoid extinction without feeding every step? How will this ever work when I’m riding? I just want to go for a ride without stopping and reinforcing every behaviour…. My quick response is Behaviour Chaining.

First we need to feed often and for little…

When first teaching behaviour it is imperative to feed lots and often. This avoids extinction, food anxiety etc… When new R+ horses don’t understand the game the difference between frustration and relaxation can be seconds. To compound that, humans that are new to using R+ usually tend to lean towards not feeding enough versus feeding too much. They often want to start feeding less for behaviour too early as well. So, to avoid the impending unwanted classical conditioning (frustration, anxiety etc.), they hear experienced R+ trainers repeating “feed more, for less” a whole lot!! This can give the impression that every behaviour will need to be clicked and fed every-time forever! Which would obviously be a big problem for any kind of continuous work.

Once the basics are in place, then we can begin to chain it….

What happens once the R+ behaviour is well established and on cue? The cue for R+ behaviour and the behaviour itself becomes a secondary reinforcer, similar to the clicker. Which means we can use behaviour to reinforce behaviour (there are other ways to use behaviour to reinforce behaviour, but we will stick with this one for now). This results in a behavioural chain. A click and primary reinforcer (food) will reinforce the entire chain at the end.

So here’s what we need in place to just go for a ride with our R+ horses without having to stop and reinforce after every cue to behaviour:

  1. Each behaviour needed is well trained with positive reinforcement (has been paired with a primary reinforcer), and is acting as a secondary reinforcer for the previous behaviour.
  2. Each behaviour is well established, on cue and has been trained with duration.
  3. Each behaviour is maintained on a variable schedule.
  4. A click and primary reinforcer will reinforce the entire sequence.

To break down what these mean:

#1

Each behaviour correctly trained with Positive Reinforcement (R+) will have a positive association to it (a strong association with a primary reinforcer). So essentially R+ trained cues for R+ behaviour become secondary reinforcers, just like the clicker. The cue for walk-on applied after the standing behaviour, is essentially like clicking for standing. A behavioural chain could look like, the walk-on cue reinforces the previous standing behaviour, a trot cue will reinforce the walking behaviour, another woah cue will reinforce the trot behaviour, a click and reinforce with a primary reinforcer (food) reinforces the entire chain of behaviour. So instead of stopping and clicking each behaviour we can keep adding cues, and then stop and click the whole chain.

#2.

Before chaining behaviour, it is important that each behaviour is well established with R+ and is on cue. The behaviour is well beyond the initial learning where each small instance must be reinforced for the learner to understand. At this point our learner knows exactly what behaviour to perform after the cue, they are used to doing the behaviour for a period of time, and they have a good association with the behaviour.

Chaining… half pass left, to renver turn… Walking is reinforced by cavesson target, cavesson target is reinforced by shoulder following cue, which is reinforced by haunches-in cue, which are all held for duration, reinforced by new shoulder cue, reinforced by new haunches-in cue, followed by a click, followed by the primary reinforcer food. (the first parts of the clip before it gets to half pass left are also behaviour chains) https://www.instagram.com/reel/CtMm3a_pUd3/?igsh=djRkb2ExbDFpYm8w

#3.

Each behaviour usually needs to be on a variable schedule before being added to the chain. A variable schedule of reinforcement is where a behaviour is reinforced after a varied amount. A variable schedule produces and maintains really strong behaviour that won’t break down after too many steps, which is important for it to function well in a chain. This keeps each behaviour strong, and keeps the chance of behavioural extinction and the frustration that goes with it quite low. So loosely if I cue woah after 1 minute of walk, and then 30 seconds of walk I’m reinforcing walk on a variable schedule.

#4.

Finally a click can be given which should cue a standing reinforcement receiving behaviour, then food is delivered. The food is a primary reinforcer. Which means it does not need to be followed by something else to hold its strength as a reinforcer. The click and primary reinforcer reinforces the entire chain. Note the click must always be followed by the primary reinforcer (food), as we want it to maintain its position as an extremely strong secondary reinforcer. The horse needs to understand the meaning of the click and reinforcement, and what he can do to cause it.

This is how I work continuously and still maintain behaviour throughout. I don’t have to click, stop and feed for every single behaviour. Behaviour chains are very useful, and make R+ horses look similar to R- horses. The only difference is the motivation behind the behaviour…

This is also how I am able to go out with one treat pouch, and don’t run out of food. I’m all about food efficiency especially when I’m out with Fox. I want him to burn more calories than I am giving him!

Some things to be mindful of:

Behaviours can lose their strength as secondary reinforcers if they aren’t sometimes reinforced by the primary reinforcers (food). So when riding sometimes I will click and feed directly after I say “walk-on” and my horse takes a few steps, or after giving a rein cue, or after a varied duration of walk, trot, or canter etc. This maintains the strength of those behaviours both as behaviours and as secondary reinforcers. Sometimes I will just follow them with another cue (secondary reinforcer) and click and reinforce somewhere else.

Imagine if one of my horse’s behaviours was motivated by negative reinforcement…. Say my horse is walking motivated by positive reinforcement, and then I apply a negative reinforcement woah cue (even a light cue). Will that negative reinforcement cue work as a secondary reinforcer? Or will it break the chain? Think about what may happen with the walk behaviour if it isn’t reinforced either by a primary reinforcer (food) or by a secondary reinforce (the next R+ behaviour) but is instead followed by an R- cue… chances are we will see less walking behaviour. Possibly extinction as well.

There are ways to see if all of the behaviours in a behavioural chain are functioning correctly… I will explain why this is important in more detail in a future post….

Have a great day!

Courtney

Leave a Reply

Your email address will not be published. Required fields are marked *

Follow Me

Share via
Copy link