Extinction - Holoendemic?
As a child I was bilingual, with 2 accents.
As a child I also learned to ride a skate board. Nearly 20years later I went back to the place of my childhood. It took me 2 years to speak German again, this time in 3 accents, at an educated, professional adult level. Over 20yrs after having a break from riding a skateboard, my commute saw me having to park further away than I wanted to walk. I bought a skateboard. Now I can ride the board with better stability, efficiency and precision than before. I still cant ollie though... In the world of animal training Extinction is the be all and end all of commercial behavioural reduction. This is also firmly established into the domestic market. With dire consequences.
Extinction certainly has its place. There are some behaviours that can be ignored and they do go away. These are behaviours still in their operant phase, ones that are not yet respondent in nature. Once the behaviour is respondent, you cannot use extinction alone to eliminate the behaviour.
Operant behaviours are ones that require a conscious thought by the subject. Respondent behaviours are ones that occur unconsciously. When learning to catch a ball a toddler must learn to cup their hands and grasp the ball being thrown whilst watching the ball fly through the air at them, which they normally avoid. These are conscious efforts. The kid playing catch in the yard with friends is performing this unconsciously, they simply react to the thrown ball in order to catch it. A well trained subject, be that animal, athlete, academic, pilot, surgeon, anesthetist or any other such specialist is able to make decisions much faster than a lesser trained fellow. Because they understand what is required and can apply it to what is happening. They are also able to pay attention to the novel stimulus that are apart from the standard schema of operation. Meaning that if there is something unusual they can adapt to it and still make the behaviour happen. A Winger can sidestep a would be tackler and still score, rather than simply running down the line without deviation. Now that we understand the inherent strength of a respondent behaviour, let us see how resilient it is. Extinction has the ability to reduce the occurrence and intensity of a behaviour to 0. But, 0 still has a magnitude and as such can still be modified. Take my personal examples above. I had not spoken a language for 20 years. A behavioural output of 20years maintained at 0 through dormancy. But it came back, with ungraduated exposure. As my mind raced to make sense of the input, specific centers where excited and "keys" "unlocked" "doors". Within a day I could order simple food and beer. I could survive! More than 2 decades after not standing on a deck, I was riding my skateboard through traffic to get to work. I could survive! That is the nature of extinction, extinction suppresses superfluous behaviours and recalls them later when they are needed. Now, those behaviours lack fluency to start, but the resurgence is as efficient as it is reliable. Think of it like watching your favourite show, pressing pause and coming back to it a day later. You are a little disjointed since your not "in the flow" of the show, but you get it and soon your back in the swing of the emotional roller coaster of whatever it is you are watching. Now imagine you where studying a documentary and you pressed pause, had a break for a couple of days and came back to it. Would you remember the details you needed to learn? Could you pick up halfway through that scene or do you need to rewind some to catch up? Like the rest of us, you would need a quick catch up.
A part of extinction is that when the behaviour does not yield the expected response, we try and try again. My car wont start? I turn the ignition a couple more times...maybe that will help. My dog wont sit? I say sit a bunch of times...maybe that will work. As the subject is put through extinction, those behaviours that previously where maintained by reinforcement, are no longer maintained by that or allied reinforcement. The behavioural output is ignored, as such no feedback is given. The subject soon learns that this is not worth the effort, until through frustration, the behaviour ramps back up again in frequency and/or intensity before finally entering its extinguished state of dormancy. Lying exhausted on the field of behavioural battle until like a respondent vampire it rises again to feed on the intermittent reinforcement until regaining full strength and complementing the behavioural repertoire once more. Wait, I said that Extinction has its place! It sure does! If the behaviour is operant then you can use Extinction in these early phases to shape the behaviour to the criteria you desire. For example, you can train your dog to sit faster by reinforcing the faster sits and ignoring the slower ones. You can train your dog to look at you instead of the food in your hand by reinforcing the eye-eye contact only. What you cannot do with Extinction is eliminate problem behaviours. If a bird is prone to pecking your toes, extinction will not stop it. If a dog is prone to jumping to your face, extinction will not stop it. The behaviour will spontaneously recover. Because somewhere along the line subtle environmental cues regenerate the behaviour and the behaviour is suddenly worth it again. Management does allow for Extinction to be practicable, Shelter, Zoo and Circus animal behaviour can be affected with Extinction with some degree of reliability. Think of a dolphin kept in a large empty holding tank or a tiger in an adequate concrete and steel enclosure or a prisoner in a holding cell. The level of management required in commercial operations such as Zoos is tremendous and the quality of management is largely due to the interactive restrictions presented upon the environment, attendant and the animal. This is not practicable in the domestic setting where the environmental stimulus is out of your control. If a dog is in estrus a couple of blocks away, it does not matter if your male has ever had sex before. That behaviour is spontaneously called into action. With all this mind, why is it that the domestic animal training industry calls for extinction to be the panacea to cure all evils? The flaw is in the proof of the science. Scientific research must place a metric upon all inputs, as such each case must be measured in its totality. That indicates the level of management required to study animal behaviour with even contemporary technology. To turn of the electric current coursing over the floor of its experimental box; the pigeon - which is restricted in using its feet for motion - only has the option of pecking at a switch to affect its now unpleasant environment. Thereby reducing the behaviroural outputs of the subject. This is a famous example of behavioural research. This serves well to illustrate the restricted nature of variability. If you own a dog, throw a lead on them and go for a walk at the same time, through the same path, every day for 3 weeks. Record the amount of new and novel occurrences that appear each day, however subtle. The amount of variability is staggering. What is the management plan supposed to be that will enrich your dogs life? The only way to eliminate a behaviour is to reduce its frequency and intensity to less than 0. The only way to achieve that is by teaching the subject to internally inhibit the behaviour. Once that is done, there is very little maintenance required of the eliminated behaviour, because it is gone.