Download LTFeb10

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Machine learning wikipedia , lookup

Reinforcement learning wikipedia , lookup

Transcript
PSY402
Theories of Learning
Monday
February 10, 2003
Partial Reinforcement Effect (PRE)



Extinction is slowest when behavior
was intermittently reinforced during
learning.
With humans, the lower the slot
machine payoff, the longer people
play (resistance to extinction).
But, if the percent of reinforced
trials is too low, rapid extinction
occurs (U-shaped relationship).
Explanations for PRE

Two explanations:



Amsel – frustration-based
Capaldi – sequential theory
Both provide good explanations for
observed data.
Amsel’s Frustration Theory


Frustration leads to rapid extinction
during continuous reinforcement.
During intermittent reinforcement,
frustration becomes associated with
responding.

Frustration then elicits not suppresses
responding.
Capaldi’s Sequential Theory


If reward follows a nonrewarded
trial, memory of the nonrewarded
trial is associated with responding.
During continuous reinforcement,
animals do not associate lack of
reward with responding.

When they encounter the first
nonrewarded trial, the state it produces
is not associated with responding.
Contingency Management



Assessment phase – determine the
frequency of behavior and the
situations in which it occurs.
Contracting phase – specifies the
relationship between responding
and reinforcement.
Management phase – implement
the contract and evaluate results.
Aversive Conditioning
Chapter 5
Aversive Events


Unpleasant, undesirable, bad for
survival.
Typically evoke strong negative
emotion:


Pain, fear, embarrassment or shame,
anxiety, frustration.
Strong emotions motivate escape
and avoidance behaviors.
Escape Conditioning

Escape response – behavior
motivated by an aversive event.


Rewarded by termination of the
aversive event.
Miller’s shuttlebox – rats escape
shock by turning a wheel that opens
a door so they can escape.
Factors Affecting Escape



Intensity of the aversive event – the
stronger the aversive event the
greater the escape response.
Amount of negative reward –
escape depends on receiving relief
from the aversive event.
Reward must be prompt – delayed
reward interferes with escape
learning.
Eliminating an Escape Response


Removal of negative reward –
escape response stops if the
aversive event continues despite it.
Removal of aversive event -- escape
response stops if the aversive event
no longer occurs.

Continues for a while due to
conditioned anticipatory pain
responses. This must be extinguished.
Vicious-Circle Behavior


Why did rats run into a pathway
with shock when staying still would
mean no shock?
Two explanations:


Fear motivates running and is
conditioned to the start box.
The animals do not realize that no
shock will occur if they don’t run.
Avoidance



Active avoidance response – an
action is necessary to avoid
aversive event.
Passive avoidance response – not
responding prevents aversive event.
Mowrer’s hurdle jumping paradigm.

CS causes animal to jump to other side
to avoid onset of shock.
Effects of Event Intensity



Except in two-way avoidance
learning, a stronger aversive event
leads to faster avoidance learning.
The greater the aversive event
intensity, the faster the passive
avoidance learning.
Greater delay between CS and UCS
interferes with avoidance learning.
One-Way vs Two-Way Avoidance


One-way – animal can avoid shock
by jumping to other side.
Two-way – animal can jump to
other side, but after a rest, it must
jump back again to avoid shock.



Animal avoids shock only by returning
to the place where it was first shocked.
The animal must ignore situational
cues.
Induces a conflict.
Flooding



Avoidance behaviors perpetuate
phobias.
Techniques for eliminating
avoidance are important to treating
phobias.
Flooding forces a person to
experience the conditioned feared
stimulus without an aversive
consequence.