1. Learning
Chapter 6

2. General Concepts
Learning - a relatively permanent change in behavior due to experience
Types of associative learning
Classical Conditioning
Operant Conditioning
Laws of Conditioning
Association
Repetition
Reward

3. I. Classical Conditioning
A form of associative learning in which a stimulus (CS) that did not previously elicit the desired response (UR) is repeatedly paired with one that does (US) until the previously “neutral” stimulus (CS) elicit the response by itself (CR)
Ivan Pavlov
Russian physiologist in the early 1900’s
Original research: digestive process
Classical conditioning explains how a neutral stimulus can acquire the capacity to elicit (or draw forth) a response originally elicited by another stimulus.
Ivan Pavlov, a prominent Russian physiologist in the early 1900’s, who did Nobel prize winning research on digestion, discovered (partly by accident) that dogs will salivate in response to the sound of a tone. In doing so, he discovered classical, sometimes called Pavlovian, conditioning.
In classical conditioning, the UCS is a stimulus that elicits an unconditioned response without previous conditioning…Pavlov’s meat powder.
The UCR is an unlearned reaction to a UCS that occurs without previous conditioning…salivating.
The CS is a previously neutral stimulus that has acquired the capacity to elicit a conditioned response...the sound of a tone.
The CR is a learned reaction to a conditioned stimulus…salivating to the tone.

4. Classical Conditioning con’t
Acquisition Phase
Acquisition = initial stage in learning
Conditioned Phase CS US UR CS CR
Trial

5. Classical Conditioning: Important Terminology
Components
Conditioned Stimulus (CS) - previously neutral stimulus that has acquired the capacity to elicit a conditioned response...the sound of a tone
Unconditioned Stimulus (UCS) - a stimulus that elicits an unconditioned response without previous conditioning
Unconditioned Response (UCR) - an unlearned reaction to a UCS that occurs without previous conditioning…salivating
Conditioned Response (CR) - a learned reaction to a conditioned stimulus…salivating

6. Classical Conditioning: Important Terminology
Expansion of CC
Stimulus Generalization – tendency of a subject to respond with the CR to a stimulus similar to the original conditioned stimulus/CS
Discrimination – subject responds with the CR only to the original conditioned stimulus (CC)
Extinction - over presentation of the conditioned stimulus without the unconditioned stimulus causes the conditioned response to fade
Spontaneous Recovery – after extinction and a 24 hour period or longer, presentation of the CS will once again elicit the conditioned response (CR)
Higher-order conditioning – a second conditioning situation in which the CS of the first conditioning situation can be used as a US in a second situation
Extinction occurs when the CS and UCS are no longer paired and the response to the CS is weakened. We know that the response is still there, just not active, because of spontaneous recovery – when an extinguished response reappears after a period of non-pairing.
Generalization occurs when conditioning generalizes to additional stimuli that are similar to the CS; for example, Watson and Rayner’s study with Little Albert, who was conditioned to fear a white rat but later came to be afraid of many white, furry objects.
Discrimination is the opposite of generalization; that is, the response is to a specific stimulus… similar stimuli don’t work.
Higher order conditioning occurs when a CS functions as if it were a UCS to establish new conditioning…condition to respond to a tone with saliva, pair the tone with a light.

8. Classical Conditioning con’t
Variables Influencing Acquisition
(Include asymptote and conditioning paradigms)
Asymptote
a plateau of learning…for the given set of experimental conditioning, maximum learning has taken place

10. Classical Conditioning: Variables Influencing Acquisition con’t
Conditioning Paradigms
Simultaneous conditioning: classical conditioning procedure in which the CS and US are presented at the same time
Ex: Bell & food presented together
*minimal results
Trace conditioning: classical conditioning procedure in which the CS is presented and stops before the US starts
Ex: Bell starts & stops, then food is presented
*longer intervals btw. CS and US weaken conditioning
Short-delayed conditioning: classical conditioning procedure in which the CS is presented before the US, but the two events overlap (CS runs until US starts)
EX: Bell starts, ½ second later, food is presented, then both stop
** Best promotes acquisition of a CC response
Temporal Conditioning: classical conditioning procedure in which time functions as the CS
Ex: Food is presented after “X” amount of time
Backward Conditioning - classical conditioning procedure in which the US is presented before the CS
Difficult to obtain.
Sometimes work in intense situations. Ex: after a traumatic car crash, a person is now fearful of ambulance sounds
Graph on page 228

11. Classical Conditioning Applied
Watson
Baby Albert experiment
Mary Cover-Jones
The first behavioral therapist
Laid the foundation for desensitization

12. Problems with Pavlov: Contemporary Views of Classical Conditioning
Stimulus Contiguity vs. Contingency
Stimulus Contiguity - model of CC that states learning takes place when two events are paired together in time
Pavlov’s original model supports this idea
Stimulus Contingency – model of CC that states learning takes place when one event reliably predicts the other
Rescorla’s 1980s research supports this notion
In classical conditioning research, a trial is a pairing of the UCS and the CS. (How many times have the tone and the meat powder been paired?) Some behaviors are learned after only one trial or pairing, while others take many trials.
Acquisition refers to the initial stage of learning a response…acquiring the response.
Conditioning has been shown to depend on stimulus contiguity; that is, the occurring of stimuli together in time and space.
So when do you sound the tone in a classical conditioning task? What works best? Of the three types of conditioning (simultaneous, short-delayed, and trace), short-delayed conditioning appears to most promote acquisition of a classically conditioned response…ideally the delay should be very brief, about ½ a second.

13. Contingency Song Sung to “If You’re Happy and You Know It”
If the bell predicts the food then it’s a hit.
Contingency is best and Rescorla passed the test.
If the bell predicts the food then it’s a hit!

14. Problems with Pavlov: Contemporary Views of Classical Conditioning
Blocking Phenomenon - prior exposure to one CS prevents (blocks) learning of another CS (diagram next page)
Presented by Leon Kamin (1969), conflicts w/ Pavlov’s contiguity model
Rescorla & Wagner stated it may be due to all “available” learning being used on the first stimulus

15. - - - - - - Cut & Paste - - - - -
Blocking diagram
Blocking
Pre- training
Training
Test
Result
Did conditioning occur? Yes or No
Group 1
Pair tone (CS1) with shock (US)
Pair tone and light (CS1 and CS2) with shock (US)
Present tone (CS1) alone
Group 2
Present light (CS2) alone
Group 3
No Pre-training
Group 4
Benjamin, Hopkins, Nation 1994

16. Instrumental Learning & Operant Conditioning
A. Instrumental learning
Learning that takes place when a response is strengthened or weakened by its consequence
Like operant conditioning, deals w/ emitted behavior
Developed by Edward L. Thorndike (1913)
“Cat in the puzzle box” experiment
Law of effect
principle that the environmental consequences will determine whether or not a stimulus-response connection will be formed
Guiding idea: through a set of random responses, one will be reinforced, therefore strengthened
Thorndike’s law of effect stated that if a response in the presence of a stimulus leads to satisfying effects, the association between the stimulus and the response is strengthened.
This law became the cornerstone of Skinner’s theory.

17. B. Operant conditioning
B.F. Skinner’s (1953) version of Thorndike’s instrumental learning
Operant chamber (a.k.a. “Skinner Box”)
Skinner’s version of …
Skinner heavily favored reinforcement, especially positive, as opposed to aversive techniques such as punishment
Note: reinforcement is any event following a response that increases the organism’s tendency to make that response
Skinner’s principle of reinforcement holds that organisms tend to repeat those responses that are followed by favorable consequences, or reinforcement.
Skinner defined reinforcement as when an event following a response increases an organism’s tendency to make that response.
Skinner created a prototype experimental procedure, using animals and an operant chamber or “Skinner box.” This is a small enclosure in which an animal can make a specific response that is recorded, while the consequences of the response are systematically controlled. Rats, for example, press a lever.
Because operant responses tend to be voluntary, they are said to be emitted rather than elicited.
Reinforcement contingencies are the circumstances, or rules, that determine whether responses lead to the presentation of reinforcers.
The cumulative recorder creates a graphic record of responding and reinforcement in a Skinner box as a function of time.

18. Operant conditioning con’t
Basic Training Procedures (How to train)
Two Questions:
1. Does the response produce or remove a stimulus?
2. Is that stimulus positive (appetitive) or negative (aversive)?
* Chart on next slide

19. - - - - - - Cut & Paste - - - - -
Basic Training Procedures (How)
Types of Training
Emitted Behavior
Good or bad?
Environment Response
Gives or takes?
Stimulus Given
Appetitive or aversive?
Response Outcome
Behavior increases or decreases?
Positive Reinforcement
Negative Reinforcement
Omission Training
Punishment

20. Operant conditioning con’t
Skinner’s view of punishment…
“The responses to punishment are either and urge to counterattack, escape or a stubborn form of apathy.”
General (not necessarily Skinner’s) Problems with punishment
Tells what NOT to do, but not what TO do.
Subject may begin to fear the situation
Subject may begin to fear the punisher
*If used, punishment should be informative and swift.
Punishment occurs when an event following a response weakens the tendency to make that response. Punishment is much more than disciplinary procedures…wear a new outfit and friends laugh…punishing. Punishment may involve presentation of an aversive stimulus (spanking) or removal of a rewarding stimulus (taking away TV).
Some of the problems associated with punishment are that it can trigger strong emotional responses (anxiety, anger, resentment, hostility); physical punishment can lead to an increase in aggressive behavior.

22. Operant conditioning con’t
Reinforcers (What to use)
Primary reinforcers
Events that are inherently reinforcing, they usually have biological relevance (you don’t learn to like them)
Secondary reinforcers (a.k.a conditioned reinforcers)
Items/events that gain value by being associated with a primary reinforcers
Wolfe & the “Chimp-O-Mat”
Generalized reinforcers
Secondary reinforcers that have been associated with primary reinforcers for so long that they take on value of their own
Skinner said that reinforcement occurs whenever an outcome strengthens a response…this definition avoids the use of “experience of pleasure,” which is an unobservable feeling.
In operant conditioning, immediate reinforcement produces the fastest conditioning.
Operant theorists distinguish between primary reinforcers, which are events that are inherently reinforcing because they satisfy biological needs, and secondary reinforcers, which are events that acquire reinforcing qualities by being associated with primary reinforcers.
Primary reinforcers in humans include food, water, warmth, sex, and maybe affection expressed through hugging and close bodily contact.
Secondary reinforcers in humans include things like money, good grades, attention, flattery, praise, and applause.

23. Operant conditioning con’t
Premack’s Principle
States that any high-probability behavior can be used as a reinforcer for a lower-probability behavior
* Claims reinforcers are relative, and anything can be low or high probability depending on the environment

24. Operant conditioning con’t
Schedules of Reinforcement (When to reinforce)
Continuous reinforcement
Training schedule in which every desired response is reinforced
Leads to faster acquisition & extinction
Intermittent (partial) reinforcement
Training schedule in which only some of the desired responses are reinforced (which ones depends on the type of schedule)
Greater resistance to extinction
A schedule of reinforcement determines which occurrences of a specific response result in the presentation of a reinforcer.
Continuous reinforcement occurs when every instance of a designated response is reinforced (faster acquisition, faster extinction).
Intermittent reinforcement occurs when a designated response is reinforced only some of the time (greater resistance to extinction).
Ratio schedules require the organism to make the designated response a certain number of times to gain each reinforcer. A fixed-ratio schedule entails giving a reinforcer after a fixed number of non-reinforced responses. A variable ratio schedule entails giving a reinforcer after a variable number of non-reinforced responses.
Interval schedules require a time period to pass between the presentation of reinforcers. A fixed-interval schedule entails reinforcing the first response that occurs after a fixed time interval has elapsed. A variable-interval schedule entails giving the reinforcer for the first response after a variable time interval has elapsed.
More than 50 years of research on these schedules has yielded an enormous amount of information about how organisms respond to different schedules.
Figure 6.19, presented on the next slide, depicts some of what has been discovered.

25. Operant conditioning con’t
Four most common schedules
Ratio schedules (based on # of responses)
Fixed ratio
subject is reinforced after a specific number of responses have been made
Produces a start-stop pattern, easier to extinguish
Variable ratio
subject is reinforced after “X” number of responses, based on an average
Produces a high, steady rate; more resistant to extinction
Ratio schedules require the organism to make the designated response a certain number of times to gain each reinforcer. A fixed-ratio schedule entails giving a reinforcer after a fixed number of non-reinforced responses. A variable ratio schedule entails giving a reinforcer after a variable number of non-reinforced responses.
Interval schedules require a time period to pass between the presentation of reinforcers. A fixed-interval schedule entails reinforcing the first response that occurs after a fixed time interval has elapsed. A variable-interval schedule entails giving the reinforcer for the first response after a variable time interval has elapsed.
More than 50 years of research on these schedules has yielded an enormous amount of information about how organisms respond to different schedules.
Figure 6.19, presented on the next slide, depicts some of what has been discovered.

26. Operant conditioning con’t
Interval schedules (based on time)
Fixed interval
subject is reinforced for the first desired response after a specific amount of time has passed
Long pause btw. responses produces a “scalloping” effect, easier to extinguish
Variable interval
subject is reinforced for the first desired response after “X” amount of time has passed, based on an average
Produces a low, but steady rate of response; more resistant to extinction

27. Schedules of Reinforcement and Patterns of Response

28. Operant conditioning con’t
Teaching Complex Behavior
Shaping (a.k.a successive approximations)
Reinforcing emitted behavior in a step-by-step fashion as it begins to resemble the goal behavior
Chaining
Making elaborate sequences of shaped behaviors

29. Operant conditioning con’t
Additional Concepts
Escape learning
subject learns to engage in a behavior to decrease an aversive stimulus
Avoidance learning
Subject learns to respond to preliminary cues to prevent/avoid an aversive stimulus before it occurs
Learned Helplessness
Decrease in responding after repeated exposure to uncontrollable aversive events
As in classical conditioning, acquisition refers to the initial stage of learning.
Learning operant responses usually occurs through a gradual process called shaping, which consists of the reinforcement of closer and closer approximations of a desired response…key in pet tricks.
Extinction in operant conditioning refers to the gradual weakening and disappearance of a response tendency, because the response is no longer followed by a reinforcer…stop giving food when the rat presses the lever…results in a brief surge of responding followed by a gradual decline until it approaches zero.
Stimuli that precede a response can exert considerable influence over operant behavior, basically becoming “signals” that a reinforcer is coming (see Figure 6.17 re: roborats demonstrating stimulus control). Discriminative stimuli are cues that influence operant behavior by indicating the probable consequences of a response (ex. slow down when the highway is wet, ask Mom when she’s in a good mood, etc.).
Discrimination occurs when an organism responds to one stimulus, but not another one similar to it, while generalization occurs when a new stimulus is responded to as if it were the original. (ex. cat runs to the sound of a can-opener which signals food, but not to the sound of the mixer…discrimination…get a new blender, cat runs to it...generalization).
Responses can be strengthened either by presenting positive reinforcers or by removing negative reinforcers.
Negative reinforcement regulates escape and avoidance learning. In escape learning, an organism learns to perform a behavior that decreases or ends aversive stimulation (turning on the air conditioner). In avoidance learning, an organism learns to prevent or avoid some aversive stimulation (turn on the a/c before it gets too hot).
Punishment occurs when an event following a response weakens the tendency to make that response. Punishment is much more than disciplinary procedures…wear a new outfit and friends laugh…punishing. Punishment may involve presentation of an aversive stimulus (spanking) or removal of a rewarding stimulus (taking away TV).
Some of the problems associated with punishment are that it can trigger strong emotional responses (anxiety, anger, resentment, hostility); physical punishment can lead to an increase in aggressive behavior.

30. Operant conditioning con’t
Ideas on Changing YOUR Behaviors
1) Do a functional analysis – a behavioral assessment that determines…
a) antecedents - situational triggers
b) behaviors - frequency of behavior
c) consequences - reinforcement for that behavior
2) Create a behavior modification plan
Ex: avoid triggers, reinforce positive behavior

31. III. Changes in Our Understanding of Conditioning (CC and/or OC)
Biological Constraints on Conditioning
Instinctive Drift
when an animal’s innate response tendencies interfere with conditioning
Conditioned Taste Aversion
readily acquired, after only one trial and when the stimuli are not contiguous
i.e. becoming ill occurs hours after eating a food (sauce béarnaise syndrome)
Garcia solved the sauce problem
Taste & nausea is easy to condition, whereas taste & aversive stimuli is not
Preparedness and Phobias
some phobias are more easily conditioned than others, suggesting the concept of preparedness…that we are biologically wired to learn to fear objects or events that have inherent danger
Martin Seligman
New research has greatly changed the way we think about conditioning, with both biological and cognitive influences having been discovered.
Instinctive drift occurs when an animal’s innate response tendencies interfere with conditioning (the raccoon who would rather rub the coins together than obtain the reinforcer).
Conditioned taste aversions can be readily acquired, after only one trial and when the stimuli are not contiguous (i.e. becoming ill occurs hours after eating a food), suggesting that there is a biological mechanism at work.
Martin Seligman has outlined the fact that some phobias are more easily conditioned than others, suggesting the concept of preparedness…that we are biologically prepared to learn to fear objects or events that have inherent danger.

33. Changes in Our Understanding of Conditioning (CC and/or OC)
Cognitive Influences on Conditioning
Signal relations (Rescorla)
Idea that contingency plays a role in C.C.
Response-outcome relations
if a desired outcome follows a response, it is more easily strengthened if it seems that the response caused the outcome (predicts)
i.e. if you eat ice cream while studying and make an A…
*Signal relations and response-outcome research suggest that cognitive processes play a larger role in conditioning than once believed.
Signal relations theory (Rescorla) illustrates that the predictive value of a CS is an influential factor governing classical conditioning.
Response-outcome relations - when a response is followed by a desired outcome, it is more easily strengthened if it seems that it caused the outcome (predicts)…you study for an exam and listen to Smash Mouth…you make an A. What is strengthened, studying or listening to Smash Mouth?
Signal relations and response-outcome research suggest that cognitive processes play a larger role in conditioning than once believed.
The evolutionary perspective on learning assumes that an organism’s biological heritage places certain constraints on the learning process (which some theorists see as merely specialized mechanisms designed to solve particular types of adaptive problems for particular species).

34. Changes in Our Understanding of Conditioning (CC and/or OC)
Hey...this means you!
Evolutionary Perspectives on learning
While learning processes are somewhat universal, there may be species specific tendencies that aid adaptation
READ this in BOOK!!!

35. IV. Observational Learning: Basic Processes
“Social Cognitive Theory” (pre-80s = social learning theory)
Albert Bandura (1977, 1986)
Famous study: the ….?
Observational learning
Learning that takes place by watching a model perform the task, then repeating the behavior
Vicarious reinforcement
Strengthening/weakening of a response by watching the consequences a model receives
4 key processes: attention, retention, reproduction, motivation
* Bandura is the bridge btw. behaviorist & cognitive learning theories
Albert Bandura outlined the theory of observational learning. In observational learning, vicarious conditioning occurs by an organism watching another organism (a model) be conditioned. Observational learning can occur for both classical and operant conditioning.
In order for observational learning to take place, 4 key processes are at work. First the organism must pay attention to the model, retain the information observed, and be able to reproduce the behavior. Finally, an observed response is unlikely to be reproduced unless the organism is motivated to do so, i.e., they believe there will be a pay off.
Bandura distinguishes between acquisition (having the response in your repertoire) and performance (actually engaging in the behavior). Bandura asserts that reinforcement usually influences already acquired responses, more than the acquisition of new responses.

37. V. Cognitive Learning Insight Learning Latent Learning
Sudden awareness of a solution to a problem
Wolfgang Kohler (1920s)
Ex: Sultan the chimp
Case 1 – fruit outside cage
Case 2 – suspended banana
Latent Learning
Learning in the absence of apparent reward
Cognitive maps – mental representations, created from experience, to be used as future guides
Edward C. Tolman (1930s)
Dates back to Gestalt psychologist Kohler in the 1920s

38. THE END! Study “Beyond Recognition” Note cards are due TEST
25 minute free response
100 multiple choice

39. Why do we stop at a red light anyway?
"Clip art licensed from the Clip Art Gallery on DiscoverySchool.com"