I heard an interesting statement that I want to tease out a little, because I’ve heard variations of this statement many times, particularly from those who are against training with Positive Reinforcement, without pressure or force.
Giving a R+ trained cue and then most importantly, HOW they respond, gives us information. A cue is not a command, it’s not a “do it or else” and it’s not an ultimatum without choice. A cue is a polite ask and it’s their choice if they want to oblige. A cue in R+ training *ideally* has choice built in.
The amount of times I’ve started off posts saying, “what I love about Positive Reinforcement training” is a bit embarrassing.
But there is so much depth and nuance to the training and sometimes it makes me sad that people are so dismissive of something that is so complex, nuanced and organic. It’s also extremely empowering and pleasurable for the animal being trained and it develops a communication between species that no other training can replicate.
What I also like is that you don’t have to be an expert ethologist or reader of body language to train with R+ and know what your animal is telling you in a training session. All behaviour ie. responses that have been observed in an R+ training context, have been analysed and have had a name given to them that we can discuss. It’s not new or magical to observe an animal’s behaviour and identify how they are responding and interpreting our behaviour and our cues in R+ training. Latency is a good example of this.
What I find interesting is that detractors of R+ training claim it is manipulative, coercive, controls the animal and their mind and that there’s no choice. Do it for the cookie or else! R+ training is quite the opposite actually, if done well.
A behaviour trained with R+ and put on cue is not an ultimatum. In the hands of a good trainer, it is anything but.
There is a famous quote attributed to Viktor Frankl which I have since learnt, was a quote describing some of his work, but not actually directly quoted from him, “Between stimulus and response there is a space. In that space is our power to choose our response. In our response lies our growth and our freedom.”. What I like about this quote is that I relate it to choice in R+ training. If the ‘stimulus’ is the cue for the behaviour (Discriminative Stimulus (SD) ) then the space refers to how the animal *chooses* to respond to the SD/cue at that moment in time. HOW they respond, gives us information. A cue is not a command, it’s not a “do it or else” and it’s not a given. A cue is a polite ask and it’s their choice if they want to oblige. This is the opposite of pressure trained cues, worse, the command or demand is the way that the behaviour is elicited as well. There’s not much choice in that – discomfort or what?
But when we train a behaviour with R+ and put it on cue, if we’ve trained it in a way that met the animal’s needs, provided alternative sources of reinforcement, be it positive and possibly negative as well and watched and listened to their behaviour and body language and responded appropriately, then we’ve set the stage correctly. The most important thing in my mind is responding to their behaviour and body language, because that’s where the communication starts and where we can tell them that we are listening, we are responding and we are offering them alternatives and choices.
The other important thing in providing choice in training is to approach the R+ training with an ‘errorless mindset’. The animal is never wrong, what did we do? When we avoid withholding positive reinforcement (food) for “incorrect responses” and look at all behaviour as communication and that we want to encourage and even reinforce that communication, then we not only open the doors of communication, we tell our animal they have choices.
In practical terms, if I’ve trained a behaviour and I know it’s on a reliable cue and my animal does not do the behaviour when cued, or is slow to do it, or offers a different behaviour or sadly, shuts down, then that’s their choice in that moment, in response to the stimulus. What I do is give them food anyway and then either cue a super easy behaviour, click and feed and end the session, or end the session straight away (possible Negative Reinforcement or Negative Punishment) leaving a generous amount of end of session food, or I might cue the behaviour one more time, in case they just didn’t hear, see or recognise it. But I’ll only cue it one more time. Any more than that and it can turn into pressure or a command or a nag, and that’s not fun. I’ve already done a post on this, focussed on discussing errorless learning.
Then it’s up to me to figure out why the animal didn’t perform the cued behaviour. I often find there is something urgent and important happening for them that over-rides the cue, something like pain. This is another thing that detractors like to say, but good R+ training means we don’t train away fear and we don’t ignore pain. High latency, lack of alacricity in performing behaviours, performing other behaviours or even calming signals, are all signs there is a problem and we don’t keep giving the cue and ignoring what the animal is saying to us. This is where choice comes in.
What happens when the animal has choices in training is that we give them a level of control, IF we listen to them and their behaviour. Choosing to respond in a certain way is controlling their environment.
Choice and control in R+ training!