Lecture 14 - NAU jan.ucc.nau.edu web server

Download Report

Transcript Lecture 14 - NAU jan.ucc.nau.edu web server

4 Neobehaviorist Psychologists
Edward Tolman
Edwin Guthrie
Clark Hull
B. F. Skinner
1
Edward Tolman

Cognitive behaviorism

1911 - Graduated from MIT with a degree
in electrical chemistry

1915 – Graduated from Harvard with a
PhD in psychology

Most influenced by Robert Yerkes who
used Watson’s textbook and who rejected
the use of introspection
2
Edward Tolman


Taught at Northwestern 1915 –
1918 when he was fired for pacifist
activities
Moved to Cal – Berkley where he
taught for 40 years
3
Edward Tolman’s Psychology



Studied the behavior of rats running
in mazes
Determined that the rats’ behavior
was more than building S -> R
connections
Rats behaved with intelligence and
purpose
4
Purposive Behavior


Watson excluded purpose and
cognition from psychological study –
Tolman felt this was an error
Attempted to develop a behaviorism
based on objective behavior, but
included the purpose of behaviors
5
Purposive Behavior in Animals and
Men



Most of the book concerned rat behavior
in mazes
Completely rejected mentalistic
psychologies and endorsed behaviorism
Psychology should be the study of
objective behavior and it should include
behaviors that are purposive, goaldirected and cognitive
6
Example of a purposive behavior



When a rat learns to run a maze it not
only expects to get a reward, but learns
that a specific reward will be there
Different rewards have different values to
the animal
When reward was switched to a less
valued reward, rats ran slower and made
more errors
7
Latent learning and cognitive maps

1929 - Rats were placed on a maze with
no reward and then later given a reward




Control group – food in the goal box on all 7
days
Ex. group 1 – no food for 1st 6 days, food in
goal box on day 7
Ex. Group 2 – no food for 2 days, food
available on day 3 - 7
Results: experimental groups performed
like the control the day after the
transition from no food to food in the goal
box
8
Latent learning and cognitive maps



Conclusion: rats must have learned the
maze during the unrewarded trial, and
they developed a cognitive map of the
maze.
Tolman referred to this as latent learning
Learning without reinforcement a major
problem for more radical behaviorists
9
Law of least effort

Rats placed on an elevated maze with no walls and
allowed to explore

3 routes to the goal box that were of different lengths

Rats then made hungry and placed on the maze – they
chose the shortest route

Block the shortest route, they took the second
shortest, etc.

Conclusion: their cognitive map included the whole
maze not just one route
10
1937 – Place Learning


Presented in his APA presidential
address
Use of a plus maze to determine
how rat solve mazes
11
Response versus place learning




Modified plus maze
Response learning: start at either S1 or
S2 – food always found by turning left
Place learning: Start at either S1 or S2 –
food always at F1 or at F2
Results: response learning very slow;
place learning very rapid
12
Tolman’s model

Three variables influence behavior:




Independent variable – conditions of the
experiment – what is controlled
Intervening variables – subject variables, age,
skill, past experiences
Dependent variable – what is measured
Behavior as measured by the dependent
variable is a function of the independent
and intervening variables
13
Two major criticisms of Tolman


He did not develop a true theory of
learning showing a clear theoretical
position
Objection to his writings as being
subjective and mentalistic
14
2 important contributions of Tolman


Support of the rat as appropriate
subjects for psychological study
Most important was his realization
that Watson’s behaviorism had
“thrown out the baby with bath
water” in rejecting all mentalism
15
Edwin Guthrie and Clark Hull



2 very different people
Guthrie received a degree in
mathematics, but became more
interested in philosophy
He read Bertram Russell and
Whitehead’s Principia Mathematic
and decided the use of deduction in
philosophy would never lead to an
understanding of the mind
16
Clark Hull



A well-known psychologist who had
studied aptitude testing and hypnosis for
12 years when he read Principia
Mathematic
He felt this work could serve as a model
for a psychological system of behavior
Both became behaviorists, but for 2
different reasons
17
Edwin Guthrie



1912 - PhD in philosophy
Changed from an interest in the
association of ideas to the more
behavioral perspective of the association
of responses
Very similar opinions to those of Watson –
minimized the importance of
reinforcement maximized the importance
of contiguity
18
Contiguity Theory



Behavior is a function of the
environment
If food is available, an animal will
emit responses to obtain it –
successful responses will be learned
The S->R association is learned
through one trial learning
19
The role of reinforcement



Reinforcement (getting the food) is only
important because it ends the activity
The last act is the one that is learned and
that learning persists
Unsuccessful acts are not learned because
they are displaced by later successful acts
20
Studies supporting contiguity theory


Prevailing theory – animals would
learn a task to maintain
homeostasis
Guthrie showed that rats would
learn a task to get water sweetened
with saccharin. No nutritional or
survival value
21
Studies supporting contiguity theory


Male rats will learn to run a maze if
a receptive female is in the goal
box. Even if they are prevented
from copulating
Both presented as evidence of
animals learning without being
reinforced – indicating that
reinforcement was not important
22
Movements versus whole acts

Guthrie proposed:




Watson and others studied whole acts
because they were easy to measure
Movements that make up whole acts
should be studied
Individual movements are learned in
one trial learning
Putting these individual movements to
create whole acts requires practice
23
Criticisms of Guthrie's Contiguity theory



Early appeal of his theory was it’s
simplicity
This simplicity was later criticized for
ignoring or failing to address problems in
learning his theory couldn’t explain – he
had “mistaken incompleteness for
simplicity”
Theory based on very little experimental
data. Took a more philosophical approach
to theorizing
24
Clark Hull


His goal was to develop universal laws of
behavior
Major enduring contribution to psychology
was the application of formal logic to
psychological problems



Form theoretical suppositions or constructs
Develop predictions of the theory
Test predictions
25
Hull’s behaviorism

Greatest influences were Darwin’s theory
of evolution - Importance of adaptability,
natural selection, and continuity of
behavior in evolution



He did not study species specific behaviors
Looking for universal laws that could explain
both human and nonhuman behavior
Mathematics: attempted to develop very
complex mathematic formulas that could
predict behavior
26
The importance of intervening variables


Watson and others were interested
in describing the formation of the
connections between stimulus and
response
Hull more interested in the nature
of the connections (habit strength)
and the variables that influenced
their development
27
Hull’s concept of reinforcement



Watson – reinforcement was only
important in that it kept the animal on
task
Guthrie – it ended the behavior and the
last behavior was learned
Hull – reinforcement strengthened the
connection between a stimulus and a
response
28
Hull’s concept of reinforcement



Positive reinforcement – anything that
reduces tension
Negative reinforcement – creates tension
to be avoided
Two issues to be addressed:


Motivation had to be part of reinforcement
His approach had to hold for both humans and
nonhumans
29
The concept of drive



Reinforcement meant the reduction of a
drive state
If drive increased, then the level of
response would increase to decrease the
level of drive
Different species may have different
drives, but reduction of drive was the
basis of reinforcement
30
Basis of Theory


1. Habit strengths could form in a single
trial, but were strengthened through
repetition and reinforcement
Complex behaviors : different stimuli may
become associated with other stimuli and
with more than one response


The habit strength between the stimulus and
different responses differed
The response evoked by a stimulus depended
on which response had the greatest habit
strength
31
2 additional variables added to formula


Incentives – some objects were
more preferred – 2 objects may
both reduce drive, but one
increases the speed of learning
because it is more preferred
Inhibition


Reactive inhibition
Conditioned or learned inhibition
32
Criticisms of Hull’s drive reduction
theory



Behaviorists – too much reliance on
intervening variables – incentives, drives,
etc. not observable
Humanists – dehumanizing humans and
ignoring individual differences
His mathematic formulas called fantasies
because they assume learning is a
continuous process when it is not
33
Hull’s contributions


Previously mentioned – system of
formal logic used in psychology
His theories and ideas were so clear
and well formulated that they
became excellent targets for others
34
B. F. Skinner – the radical behaviorist



1930’s and 1940’s Guthrie and Hull were
the leading behaviorists in psychology
1950’s and 1960’s behaviorism dominated
by Skinner
Tolman’s works were neglected for this
period, but later became important for
the development of cognitive psychology
35
Influences on Skinner


Neurophysiologists Pavlov and
Sherrington – study of the reflex
Watson’s behaviorism
36
Operant conditioning



Classical conditioning – the animal
responds to the environment – learning
results from the environment
Operant conditioning – the animal
operates on the environment – the animal
performs arbitrary behaviors and if a
behavior is rewarded it will occur again
The animal controls the response rate not
the experimenter
37
Schedules of reinforcement


Accidental beginning – he was
studying the nature of
reinforcement and only had enough
food for a few trials
3 different schedules



Continuous
Ratio – fixed and variable
Interval – fixed and variable
38
Other contributions


System of behavioral training called
shaping – use of successive
approximations
Behavior modification training
39
Skinner and controversy

Skinner sought out controversy

Walden II – utopian society based upon
behavioral control

When asked if his house was on fire and he could
save his children or his books, he replied he
would save his books. His writings would make
greater contributions than his genes

Heir conditioner

Strongly attacked public education practices
40