1.2 Applications of Learning Principles to Neurofeedback

Lesson: 1.2 Applications of Learning Principles to Neurofeedback

1.2 Applications of Learning Principles to Neurofeedback

Presented by Sevilay Dogan

Over the next few lessons in this module, we will be looking at the applications of learning principles to Neurofeedback, starting with classical conditioning, the law of effect and then operant conditioning.

The key figures involved in developing these paradigms and their experiments are:

Classical Conditioning
Law of Effect
Operant Conditioning
Classical Conditioning
IVAN PAVLOV

1887

Law of Effect
EDWARD THORNDIKE

1898

Operant Conditioning
B.F. SKINNER

1950s

We will also look at NFB and where NFB fits in with these paradigms.

Classical Conditioning

Let’s have a look at the classical conditioning paradigm using the below example:

(click on each image to learn more)

So that is classical conditioning, which is not used for training new behaviours so much, whereas operant conditioning is.

Thorndike’s Law of Effect

Now we are going to look at Thorndike’s experiments and his theory of Law of Effect.

Thorndike was interested in how learning takes place and how habits form. And his experiments came to be about a decade after Pavlov’s Classical Conditional experiments in 1898.

Thorndike used cats in his experiments along with a puzzle box. He would put hungry cats inside a box, and the cats had to learn to solve the puzzles within the box in order to access the food placed outside of the box.

The few ways in which the cats could solve the puzzle was pressing a level, pulling on a string or stepping on a platform that would open up the door and enable them to reach the food. What he observed was that initially, their behaviour and movements were at random – pulling, pushing and tugging – that is, until they stumbled across the solution.

Before we watch a short demonstration of Thorndike’s experiment, can you guess how many trials it took for the cat’s behaviour to become learned and habitual:

(hover your mouse over your guess)

Great try!
Guess again...

Well done!
This is correct.

Great try!
Guess again...

Thorndike was also tracking the time in which the cats would be able to stumble upon the solution and make it a learned and habitual behaviour. When he compared the average time it took for the average of the first and last 3 trials, what do you think the comparison was:

(hover your mouse over your guess)

Well done!
This is correct.

Great try!
Guess again...

Great try!
Guess again...

This led to the development of law of effect. Let’s take a look:

Post-reinforcement EEG synchronisation as an endogenous brain reward/consolidation.

Before conditioning takes place, we have the food as the unconditioned stimulus (it isn’t paired with anything and elicits a natural response on its’ own which is salivation) – and we have the unconditioned response of the dog salivating upon seeing and smelling the food.

Before conditioning occurs, we can take a neutral stimulus, such as the bell in Pavlov’s experiments– and on its own, it does not elicit a response so nil salivation therefore no conditioned response.

The neutral stimulus which is the bell and pair it with the unconditioned stimulus of food, and repeat this pairing enough times (so the ringing of the bell followed by production of food) which results in successful conditioning of salivation.

Result: now the bell becomes a conditioned stimulus which causes the dog to salivate at the sound of the bell as it was paired with the unconditioned stimulus of the food.