9/27/2010

Classical Conditioning Operant Conditioning Different Perspectives of learning

Behavioural Cognitive Ethological perspective

Psyc100, Chapter 4, Lecture 1

Learning - any process through which experience at one time can alter an individuals behavior at a future time
E.g. Habituation A reduction in response to a stimulus after repeated presentations e.g. startle response

The attempt to understand observable activity in terms of observable stimuli and observable responses John B. Watson (1913)
Give me a dozen healthy infants and my own specified world to bring them up in and Ill guarantee to take any one at random and produce teachers, lawyers

B.F. Skinner (1938)

Training procedure Neutral stimulus elicits reflexive response through being paired with another stimulus that already elicits this reflexive response Reflex = Stimulus-response sequence mediated by CNS Stimulus Response = action that automatically follows an event

NEUTRAL STIMULUS (Bell)

will elicit will elicit a

NO REACTION

UNCONDITIONED STIMULUS (Food)

REFLEX ACTION

UNCONDITIONED STIMULUS (Food) NEUTRAL STIMULUS (Bell)

will elicit a

REFLEX ACTION

CONDITIONED STIMULUS CONDITIONED STIMULUS (Bell)

will elicit a

CONDITIONED RESPONSE

9/27/2010

Neutral stimulus Unconditioned Stimulus

Does not normally elicit a response or reflex action by itself

A bell ringing A color A furry object

Always elicits a reflex action: an unconditioned response

Food Blast of air Noise

Unconditioned Response

A response to an unconditioned stimulus--naturally occurring

Salivation at smell of food Eye blinks at blast of air Startle reaction in babies Will eventually elicit the unconditioned response by itself

Conditioned Stimulus

The stimulus that was originally neutral becomes conditioned after it has been paired with the unconditioned stimulus

UCS (drug) UCR (nausea)

CS (waiting room)

UCS (drug) UCR (nausea) CR (nausea)

CS (waiting room)

Learning Experience
Stimulus A (The word ball) Stimulus B (Sight of a ball) Thought of B (Mental image of a ball)

After Learning
Stimulus A (The word ball) Thought of B (Mental image of a ball)

Conditioning Procedure
Neutral stimulus (Bell) Unconditioned stimulus (Food) Unconditioned response (Salivation) Conditioned response (Salivation)

Extinction Without food the bell elicited less and less saliva But, does not return animal to previous nave state Spontaneous recovery Passage of time following extinction partially renew the conditioned reflex

After Conditioning
Conditioned stimulus (Bell)

9/27/2010

Strong
Acquisition (CS+UCS) Extinction (CS alone) Spontaneous recovery of CR

Strength of CR

Extinction (CS alone)

Weak Pause Time

The conditioned stimulus is not truly lost during extinction, but is Inhibited. Eye-blink reflex studies (i.e. Tone + Puff of Air) in rabbits has shown that conditioning and extinction involve different sets of neurons Neurons involved in conditioning excite neurons that control eye-blinks. Neurons involved in extinction inhibit neurons that control eye-blinks

After conditioning, stimuli that resemble the CS will elicit the response even when they have never been paired before. Depends on the degree of similarity between new stimuli & conditioned stimuli. Further a new tone is away from the original tone, the less the dog salivated.

Generalisation can be abolished if the response to one is reinforced while the response to the other is extinguished E.g. Conditioning to black square is generalised to grey square. Grey square extinguished (no pairings) Eventually conditioned the dog to discriminate a black square from a grey that was (almost) imperceptibly different from the black. Allows investigations of sensory capacities

Generalisation occurs in both physically and semantically similar stimuli (Razran, 1939) Paired words with lemon juice squirts Style, urn, freeze, surf = Salivate Generalised to fashion, vase, chill and wave Did not generalise to homophones or orthographically similar words (e.g. Serf, stile, etc) These associations must be encoded deeply (i.e. Semantic rather than surface forms)

John B. Watson and Little Albert

first psychologist to explain human behavior in terms of Pavlovian conditioning Fear not seen as feeling but observable behavior, catching breath, stiffening body turning away This emotion can be thought of as a reflex and is therefore amenable to scientific investigation through CC

9/27/2010

UCS (loud noise)

UCR (fear)

CS (rat)

UCS (loud noise)

CS (rat)
Stimulus similar to rat (such as rabbit)

UCR (fear)

CR (fear)

Conditioned fear (generalization)

We dont just react to stimuli, we often behave in ways that produce certain changes Actions that result in a particular goal are known as operant responses

E.L. Thorndike (1898) Deprived cats of food Placed cats in puzzle box On first trial cat engaged in many different behaviors until accidentally opening box After 20 30 trails cat could open box almost immediately after entering it.

Learning = Trial and Error

After Many Trials in Box

Scratch at bars Push at ceiling Dig at floor Howl Etc. Etc. Press lever

Situation: stimuli inside of puzzle box

9/27/2010

Responses that produce satisfying effects in particular situations become more likely to occur again in that situation, and responses that produce discomforting effects become less likely to occur again Pavlovian condition: Animal = passive agent Thorndike: Animal = active agent that emits behaviour from its own accord

Did not like the term satisfying

Used term reinforcer for any event that follows a behavior AND strengthens the behavior

Invented better apparatus: the Skinner box

Animals could be kept in the box for the whole duration of the experimental session whilst multiple conditioning trials could take place

Consequences
positive and negative reinforcement positive and negative punishment

Punished behavior is not forgotten, it's suppressed--behavior returns when punishment is no longer present Causes increased aggression- shows that aggression is a way to cope with problems Creates fear that can generalize to undesirable behaviors, e.g., fear of school Does not necessarily guide toward desired behavior--reinforcement tells you what to do, punishment only tells you what not to do Punishment teaches how to avoid punishment

9/27/2010

Allows complex behaviours to be conditioned Process reinforces gradually more desired responses until final response is achieved
E.g. The grading system in musical instrument learning Reinforcement stays consistent (certificate etc.) required behaviour becomes more complex

Continuous Reinforcement
reinforcing the desired response each time it occurs learning occurs rapidly extinction occurs rapidly

Partial Reinforcement
reinforcing a response only part of the time learning occurs slowly resistance to extinction

Continuous: 1 to 1 ratio, a prize every time Ratio:

fixed: 1 to ?, a prize every ? time variable: ? to ?, maybe a prize, maybe not!

Interval:
fixed: announced examination variable: pop quiz

Ratio Schedules = higher response rates - the more you respond the faster you reach the set threshold and receive the UCS Variable Schedules = greater resistance to extinction as you are never sure when a reward will be presented so you dont know when is a good time to give up - it could always be just about to pay out (fruit machines)

Indicate when a reinforcer is potentially available E.g. a lever press will only result in a food pellet when a red light is illuminated
Red light = discriminative stimulus

E.g. waiting for the ready green light to flash on a camera before taking a picture E.g. waiting until people are in a good mood before asking them a favour

9/27/2010

Primary Reinforcer
An innate reinforcer Satisfies a biological need

Tokens are secondary reinforces

Reinforcing value related to previous learning experiences E.g. Work > Money > Buy Food

Secondary Reinforcer
A conditioned reinforcer An event that gains its reinforcing power through its association with a primary reinforcer

Chimpanzees will work for tokens and save them for use later when grape vending machine is removed from enclosure Clicker training

The most famous example was Azrin's token economy in psychiatric hospital Tokens could be exchanged for cosmetics, candy, cigarettes, clothing, bedside tables, use of the TV, stereo, & sleeping late Tokens were obtained by attending work and therapy as well as for good grooming, appropriate meal time behaviors, and minor housekeeping chores Reduction in bizarre behaviors and increased normal behaviors and social skills

Advantages of Tokens
Potent reinforcers big changes in behavior Bridge the delays between target responses and back up reinforcers Backed up by a variety of items and hence are less subject to satiating. Administration does little to disrupt on going behavior Can be used with many individuals all with different back up reinforcer preferences Can be accumulated towards valuable goals

Disadvantages of Tokens
Not usually seen in classrooms nor are food snacks or other unusual back up reinforcers Tokens, except for money and grades, are unavailable in the normal environment In some environments, people will use unauthorized means such as force/theft to obtain

Stimulus precedes the response and elicits it Elicited responses Learning as a result of association Pavlov

Stimulus follows the response and strengthens it Emitted responses Learning as a result of consequences Skinner

CLASSICAL

OPERANT

9/27/2010

In groups Think of some real world (human or nonhuman animal) examples