Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Chapter 6
Learning
Cognitive Functions of the Mind
Mediate adaptive behaviours
• Interactions between person and world
Form internal representations of the world
• Perception, memory
Reflect on this knowledge (transform in imagination)
• Reasoning, problem-solving, decision, choice
Use knowledge to guide behaviour
• Action
Communicate knowledge to others
• Language
Major function of the mind is to engage in these cognitive functions
and mediate adaptive behaviour to mediate interactions between
individual and the world.
Traditional View of Learning
Associationists – Mind forms associations between:
Stimulus
– events in environment
– correlates and consequences
Responses
– organism’s action
– correlates and consequences
Some Associations are Innate
Reflexes
– Involve individual muscles (patella reflex, eye blink)
Taxes
– Involve entire body (innate S-R patterns)
– Positive (move towards), Negative (move away)
Instincts
– Shaped by evolution to help organism adapt to a particular
environment
– Very discriminating
– Can be VERY complex patterns of behaviour
Tinbergen (1947)
Limitations of Innate Responses
Work fine as long as environment doesn’t change,
but can become maladaptive...
Need to have mechanism for
individual species to modify
responses to stimulation...
And THAT is what happens by
virture of learning.
Learning Defined
Relatively permanent change in current or potential
behaviour occurring as a result of experience.
– Not drugs
– Not injury
– Not maturation
All organisms with nervous system have some capacity to
learn.
Permits individual organisms ability to acquire new
behaviours under new circumstances – adding to the
repertoire already present as a result of evolution
Classical Conditioning
Ivan Pavlov (early 1900’s) Physiologist who did
Nobel prize winning research on digestion,
discovered (partly by accident) that neutral stimuli
had the capacity to elicit reflexive responses.
Classical Conditioning Terms
Unconditioned Stimulus (UCS) – reliably evokes a
reflex response on part of organism.
Unconditioned Response (UCR) – is the reflex
response reliably evoked by UCS.
Conditioned Stimulus (CS) – a previously neutral
stimulus that, once paired with UCS, has acquired
the capacity to elicit the same response as UCS.
Conditioned Response (CR) – a learned response
to a CS. (This happens after many pairings [i.e.
TRIALS] of CS with UCS).
Processes of Classical Conditioning
Acquisition – process by which a CS acquires the power to evoke a CR. – Reinforcement of CS by UCS
– Response gains strength (measured in terms of magnitude and probability) as trials continue.
Extinction – process by which the CS loses power to evoke CR by withdrawal of reinforcement.
– No reinforcement CR loses strength
Spontaneous Recovery – (itself subject to extinction) – Rest after extinction
– Retest CS alone (will see some degree of CR)
Re-Acquisition - relearn faster – CS reinforced by UCS
– Shows CR is not “lost” rather inhibited or suppressed
Generalization – CS0 vs. CS1…..CSn (baby Albert)
Discrimination – response to two CS for a time but eventually organism is able to discriminate as one CS is reinforced while the other is not.
XXX 6.8
XX 6.7
XX 6.10 Higher-Order Conditioning is another of the Processes of
Classical Conditioning.
Classical Conditioning: More Terminology
Stimulus contiguity = occurring together in time and space
3 types of Classical Conditioning
– Simultaneous conditioning: CS and UCS begin and end together
– Short-delayed conditioning: CS begins just before the UCS, end together
– Trace conditioning: CS begins and ends before UCS is presented
Too Extreme but...
Pavlov felt ALL forms of learning are basically variants
on Classical Conditioning. Now we know that’s too
extreme be we do understand that to a very large
extent, the laws of Classical Conditioning are laws of
emotional life.
By virtue of processes like Classical Conditioning, we
acquire our fears, aversions, joys, preferences, etc.
Some of these are innate but most we have learned to
respond to as a result of a process very similar to
Classical Conditioning.
Classical Conditioning in Everyday Life...
Conditioned Fears
Conditioned Emotional Responses
Conditioning and Physiological Responses
Conditioning in Advertising
Conditioning and drug effects
Thorndike’s Puzzle Box
Conclusion learning motivated by reward
F 6.11
Instrumental Learning ~ Thorndike’s Laws of Learning (1913)
Law of Readiness – learning is motivated by an organism’s internal state (which activates a whole sequence of behaviours.
Law of Effect – Responses that lead to reward, are strengthened while those not leading to reward are weakened.
Law of Exercise – Connections between stimuli and responses are strengthened by practice (repetition) and weakened by disuse.
Operant Conditioning
B.F. Skinner (1953) – principle of reinforcement
Operant Conditioning
B.F. Skinner (1904-1990)
"All we need to know in order to
describe and explain behaviour is
this: actions followed by good
outcomes are likely to recur, and
actions followed by bad outcomes
are less likely to recur." (Skinner, 1953)
Instrumental Learning/ Operant Conditioning
General Principles:.
Organisms acquire adaptive behaviour through the experience of success or failure.
Organisms OPERATE on their environment and their behaviour changes their environment. (Skinner,1953)
Behaviour is INSTRUMENTAL to obtaining a desired outcome. (Thorndike,1913)
Whereas in Classical Conditioning we see animals forming associations between CS and UCS, in Operant Conditioning, the associations are between behaviour(s) and outcome(s).
Operant Conditioning
Skinner (1938) revised: the Law of Effect
Used operant conditioning chamber
Two changes: – Specifically defined “reinforcement” as strengthening the
connection between response and a consequence: R Sr
– Also defined “increase” as a change in probability of occurrence of a response (more definable)
When a response is followed by a REINFORCER, that response increases in probability.
When a response is followed by a PUNISHER that response decreases in probability
Operant Conditioning
Operant Conditioning
type of learning in which behaviour is strengthened if
followed by reinforcement or diminished if followed by
punishment.
Type of learning in which the future probability of a
behaviour is affected by its consequences.
Law of Effect
Thorndike’s principle that behaviours followed by favorable
consequences become more likely, and behaviours
followed by unfavorable consequences become less likely.
Skinner box:
Pigeon pecks or rat presses
bar to receive reinforcers
Operant Chamber
Skinner Box
chamber with a bar or
key that an animal
manipulates to obtain a
food or water reinforcer
contains devices to
record responses
Basic Processes in Operant Conditioning
Acquisition – the initial stage of learning operant
responses
Shaping – a gradual process consisting of
reinforcement of closer and closer approximations of
a desired response (KEY in pet tricks)
Extinction – gradual weakening and disappearance
of a response (no longer reinforced).
Stimulus Control
– Generalization
– Discrimination
Operant Conditioning
B.F. Skinner (1953) – principle of reinforcement
– Operant chamber – “Skinner Box” – F 6.13a
– Emission of response (because Operant Responses tend
to be voluntary, they are said to be “emitted” rather than
“elicited”).
– Reinforcement contingencies – antecedents, behaviours,
and consequences (ABC)
– Cumulative recorder – F 6.13b
Figure 6.13 Skinner box and cumulative recorder
Operant Conditioning
Reinforcer
any event that strengthens the behaviour it
follows.
Reinforcement Contingencies
Circumstances or rules that determines whether
response leads to reinforcer.
Shaping
operant conditioning procedure in which
reinforcers guide behaviour toward closer and
closer approximations of a desired goal.
XX 6.12
Operant Conditioning
XX 6.18
Reinforcement: Consequences that Strengthen Responses
Delayed Reinforcement (immediate produces fastest
conditioning)
– Longer delay, slower conditioning
Primary Reinforcers - events that are inherently
reinforcing because they satisfy biological needs (food, water,
warmth, sex, and maybe affection expressed through hugging and close bodily
contact).
– Satisfy biological needs
Secondary Reinforcers - events that acquire reinforcing
qualities by being associated with primary reinforcers. (money, good
grades, attention, flattery, praise, and applause).
– Conditioned reinforcement
Schedules of Reinforcement
Continuous Reinforcement
reinforcing the desired response each time it occurs
Partial (Intermittent) Reinforcement reinforcing a response only part of the time
results in slower acquisition
greater resistance to extinction
Schedules of Reinforcement
Fixed Ratio (FR)
reinforces a response only after a specified number of
responses
faster you respond the more rewards you get
different ratios
very high rate of responding
like piecework pay
Variable Ratio (VR)
reinforces a response after an unpredictable number of
responses
average ratios
like gambling, fishing
very hard to extinguish because of unpredictability
Schedules of Reinforcement
Fixed Interval (FI)
reinforces a response only after a specified time has
elapsed
response occurs more frequently as the anticipated time
for reward draws near
Variable Interval (VI)
reinforces a response at unpredictable time intervals
produces slow steady responding
like pop quiz
XX 6.17
Schedules of Reinforcement
Variable Interval
Number of
responses
1000
750
500
250
0 10 20 30 40 50 60 70
Time (minutes)
Fixed Ratio
Variable Ratio
Fixed Interval
Steady responding
Rapid responding
near time for
reinforcement
80
Partial Reinforcement Schedules
Fixed Ratio: every nth response is reinforced
Fixed interval: the first response after x
amount of time is reinforced
Variable ratio: an average of every nth
response is reinforced (hardest to extinguish)
Variable interval: the first response after an
average of x amount of time is reinforced
Big Bang…
Reinforcers vs. Punishers Positive vs. Negative
Reinforcer = rate of response INCREASES
Punisher = rate of response DECREASES
Positive: something is ADDED to environment
Negative: something is TAKEN AWAY from
environment
NOTE that both reinforcement AND punishment can
be in positive and negative forms.
XX 6.20
Punishment
A consequence that decreases an organism’s tendency to make a particular response.
• aversive event that decreases the behaviour that it follows.
• powerful controller of unwanted behaviour
Punishment
Punishment
How to make it more effective while reducing its side effects:
• Apply swiftly (if delay too great, ineffective).
• Use punishment just severe enough to be effective.
• Make it consistent (if you want to eliminate a response, punish
the response every time).
• Explain the punishment (reason for punishment should be made very clear. Punishment combined with reasoning if more effective than either alone).
• Use noncorporal punishment, such as withdrawal of privileges.
Escape/Avoidance Learning Soloman & Wynne, 1953
Two kinds of association
• Lights shock (C.C)
• Vaulting barrier shock goes away (O.C.)
Response during shock ESCAPE
Response prior to shock AVOIDANCE
Two-Factor Theory (Mower, 1947)
• 1st condition fear to light
• 2nd reinforce escape/avoidance by
association of light to shock
Significance of Operant Conditioning
Voluntary behaviours come under control of environmental events.
behaviour outcome
Ubiquitous – all vertebrates and many non-vertebrates can show this type of learning.
Thorndike/Skinner felt ALL learning was a variant on Operant Conditioning. This view is also too extreme but it’s very clear that the laws of Operant Conditioning appear to account for much of the acquisition and display of a great deal of adaptive (and not so adaptive) behaviours.
• Habits
• Incentives (all behaviours that organisms learn to do under the
condition of incentives).
Operant vs Classical Conditioning
Table 6.1 Comparison of Basic Processes in Classical and Operant Conditioning
Assumptions of S-R Theory
Association by contiguity
– Co-occurrence in space and time
Arbitrariness
– Any stimuli any response
Empty Organism
– Understand animal’s behaviour solely in terms of S-R…
input and responding output.
– “Black Box” connects S’s and R’s (why? how? Don’t care).
Passive Organism (hidden assumption and metaphor of conditioning)
– Animals aren’t trying to figure anything out.
Assumptions of S-R Theory
No sense of expectations or free will on the part of
the organism.
Things happening “to” the organism.
Now know that these assumptions are all wrong…
But we need to talk about them because they
DOMINATED the psychology of learning for half a
century!
Still see residues in some psychology theories today.
Changes in Our Understanding of Conditioning
Biological Constraints on Conditioning
– Breland and Breland (1961) – misbehavior of organisms
– Instinctive Drift
– Conditioned Taste Aversion – Garcia & Koelling (1966) –
Figure 6.22
– Preparedness and Phobias
Cognitive Influences on Conditioning
– Signal relations
– Response-outcome relations
– Latent learning – F 6.23
Evolutionary Perspectives on learning
Experiments that Challenged…
Garcia and Koelling (1966)
research on conditioned taste
aversion. see p.239; Figure 6-23
Rats – Compound CS
Bright light
Loud noise
Sweet water
US 1) Foot shock (immediate)
2) X-Ray (delayed nausia)
Later given choice of water source
1) Bright & Noisy 2) Sweet
Shocked rats preferred sweet source
X-Rayed rats preferred bright and noisy
Association between sight and sound shock
Association between taste nausia
Seligman and Haager (1972)
taste aversion “Sauce bérnaise”
see p.238; Figure 6-22
Experiments that Challenged…
These landmark experiments showed that associations are NOT
arbitrary.
Avoidance learning capitalizes on species-specific repertoire of
defensive reactions built in by evolution.
Learning seems to be governed by what’s now known as
PREPAREDNESS PRINCIPLE (Seligman,1970)
By virtue of its evolutionary history, every species is
predisposed to form certain kinds of associations.
– Prepared (predisposed to acquire)
– Unprepared (not predisposed to acquire)
– Contraprepared (not possible to acquire)
Organism CANNOT be treated as if it is “empty”.
Have to know about the internal biological structures to know what
it can and can’t learn – how brain has been shaped by evolution.
RATS vs. BIRDS
Cognition and Learning
Latent learning (Tolman, 1932)
– Learning that occurs but is not apparent until there is some reason to demonstrate it
Cognitive map (Tolman,1932)
– A mental representation of the layout of one’s environment. For example, after exploring a maze, rats act as if they have learned a cognitive map of it.
Overjustification effect
– The effect of promising a reward for doing what one already likes to do. The person may now see the reward, rather than the intrinsic interest, as the motivation for performing the task.
Cognition and Learning
Intrinsic Motivation
desire to perform a behavior for its own
sake and to be effective
Extrinsic Motivation
desire to perform a behavior due to
promised rewards or threats of
punishments
Other Cognitive Processes in Learning
Learned Helplessness (Seligman, 1967)
– Occurs when an organism believes that behaviors are not related to consequences
– When people’s past experience leads them to believe that nothing they can do will change their lives, they tend to stop trying.
Insight (Kohler, 1947)
– The sudden grasp of new relationships that are necessary to solve a problem and that were not learned in the past.
– Kohler’s studies of chimpanzee problem-solving
Observational Learning: Basic Processes
Albert Bandura (1977, 1986) – Observational learning
– Vicarious conditioning
– Bandura, Ross, & Ross (1963) – featured study p. 245-246
4 key processes – attention
– retention
– reproduction
– motivation
acquisition vs. performance
Bandura, Ross, & Ross (1963) featured study - p. 245 – 246 – Figure 6.25
Observational Learning and the Media Violence Controversy
Studies demonstrate that exposure to TV and
movie violence increases the likelihood of
physical aggression, verbal aggression,
aggressive thoughts, and aggressive emotions
The association between media violence and
aggression is nearly as great as the correlation
between smoking and cancer – F 6.26 – third
variable problem
Figure 6.27. Comparison of the relationship between media violence and aggression to other
correlations.