Transcript behaviorism and operant conditioning

Operant Conditioning
What is it?
Learning from the consequences of
behavior
Depending on the consequences the learner
will learn to repeat or eliminate the
behavior
Law of Effect
Responses are
selected if they are
followed by
positive
consequences.
Thorndike argued
this was similar to
a process of natural
selection.
B.F. Skinner
Reinforcement
Stimulus/event that affects the likelihood
that a preceding behavior will be repeated
Skinner used shaping - reinforcers to guide
an animal’s action toward a desired
behavior. Immediate reinforcement is best.
Dolphins working for the
military?
Positive v. Negative
Reinforcement
 Adding something good
as a consequence
 Examples: chimps and
food; person working
hard for a bonus
 Removing something
negative
 Examples: students
who complain about
grades receive higher
grade; daydreaming
during a boring lecture
 2 types (escaping
something bad, or
avoiding it)
Aversive Conditioning
 Punishment - Adding
something negative
 Where reinforcement
increases behavior,
punishment will decrease a
behavior.
 But: fails to provide an
alternative and has
potentially negative byproducts
Other Principles
Feedback:
Essential to learning is that you need to
find out the response/results to your action
Reinforcement is a type of feedback from
your actions
Extinction:
Forget the learning if there is no longer a
response – it is distinct from punishment.
Contingency of
Reinforcement
A description of the relationship between a
response and a reinforcer
Example of scolding a child for undesired
behavior, but the behavior remains…what’s
happening?
Omission – reduce the probability of a
response by giving positive reinforcement
when the behavior does not occur…keep it
REAL!
SHAPING
Behaviorists view reinforcement as the
underlying principle of all behavior – but
how do we explain complex behavior?
Baby steps
Skinner argued that new behavior came
from random “behavioral drift.” Desired
behaviors are developed through shaping.
Shaping – guiding acquisition of a response
by providing successive reinforcement.
Reinforcement Schedules
 Continuous Reinforcement desired response is reinforced
everytime it occurs.
 Partial (intermittent)
Reinforcement
 Fixed Ratio/Interval - reinforce
after a set number of
responses/set amount of time.
 Variable Ratio/Interval reinforcement at unpredictable
response numbers/unpredictable
amount of time
Discriminative
Stimuli – being able
to determine which
specific stimulus
will bring a
consequence.
Non-contingent
Reinforcement –
when a reinforcer
is not related to a
response.
Consequences that
seem to come from
random events.
We develop
responses to these
reinforcers superstitious
Issues with Punishment
Depends on contiguity – need to be sure
that the consequence is associated with
the undesired behavior.
Avoidance – are you changing the behavior,
or learning to avoid the consequence?
Escape and Avoidance learning – learn new
behaviors to avoid something negative...how
does this complicate behaviorism?
Other Issues
Biofeedback – Autonomic conditioning
Biological constraints on learning
Critical Periods and Preparedness
Lorenz and the ducks
Garcia et al. and “bait shyness” – essentially taste
aversion
So why does any of this
matter?
 Animal training
 Child raising
 Reinforce good
behavior.
 Ignore whining.
 No harsh punishment,
explain misbehavior.
 Or YOU…. (token
economies)
LEARNED HELPLESSNESS
What if you keep trying to do something
and just can’t get it?
What if the reinforcement just doesn’t
matter that much to you?
Would you just give up?
Behavioral Neuroscience
Combining the behaviorist and biological
perspectives.
Learning is a process that involves changes in
neural activity.
Neural Plasticity – ability of the neural
system to change in response to experience
Habituation behavioral
response that
decreases over
successive
presentations of
the stimulus.
Sensitization –
behavioral
response increases
during
presentations of
Non-Associative
Learning
Associative Learning
Eye Blink
Learning resides in
the cerebellum
Cerebellum
normally inhibits
the eye blink, but is
conditioned to
“synaptically” not
respond.
Fear
Learning resides in
the amygdala.
Long Term
Potentiation –
persistent increase
in synaptic
transmission
Long Term Potentiation
Increased ability of
a neuron to receive
a message.
Involves activity of
NMDA receptors on
neurons.
Social Learning Theory
Social Learning Theory – focuses on
learning within a social context, that
people learn from one another. Includes
the concepts of observational learning,
imitation, and modeling.
An observer sees a
model – and repeats the
behavior of the model.
General Principles:
People learn by observing behavior of
others, and the outcomes of that behavior.
Learning can occur without a change in
behavior – just through observation.
Cognition (thinking) plays a role in learning.
Awareness of events and future thinking
determine what we learn.
Social Learning is a bridge between
behavioral learning and cognitive learning.
Role of the Environment in Reinforcing
Learning:
Observer can be reinforced by the model.
Observer reinforced by 3rd person.
Imitated behavior leads to reinforcing
consequences.
Consequences of the model’s behavior
affect the observer’s behavior vicariously.