Generalization, Discrimination & Stimulus Control

Question 1

What is generalization?

Answer 1

Generalization is the tendency for learned behaviour to “spread” to situations not involved in training.

Question 2

Why is generalization such an important topic?

Answer 2

It is important because we rarely experience the exact same situation twice, so we must be able to apply our learning to other instances for it to be useful.

Question 3

What is discrimination?

Answer 3

the tendency to behave differently in different situations

Question 4

Why is discrimination a valuable ability?

Answer 4

it prevents using a learned technique/method in an inappropriate situation.

Question 5

How can generalization be increased?

Answer 5

With reinforcement.

Question 6

Describe the results of Eisenberger et al. in inducing the generalization of desirable behavioural tendencies.

Answer 6

Rewarding a high level of effort on one task increased the effort on other tasks.

Question 7

What is learned industriousness?

Answer 7

Trying hard, if reinforced in one situation, may generalize to another situation.

Question 8

When is generalization is not helpful?

Answer 8

When a behaviour generalizes to situations in which it is not appropriate.
E.g., children playing violently and this behaviour generalizing to unplayful situations.

Question 9

Describe Dweck and Repucci’s study illustrating the negative side of generalization (students responding to teachers)

Answer 9

A teacher gave students difficult problems that they couldn’t solve, followed by easier problems that were at their level. The students had learned to give up and this generalized to the easier problems. When given by a different teacher, they were successful in the tasks given.

Question 10

What is a generalization gradient?

Answer 10

A figure depicting the relationship between stimulus similarity and generalization, showing the tendency for a behaviour to occur in situations that differ systematically from the training stimulus.

Question 11

Describe Guttman and Kalish’s study of stimulus generalization in pigeons.

Answer 11

Birds learned to peck a disk of a particular colour and later were able to peck disks of other colours, including the colour used for training. The closer the colour was to the training colour, the more often the birds pecked it.

Question 12

Define semantic generalization

Answer 12

When the learned behaviour generalizes on the basis of an abstract feature.

Question 13

Describe Razran’s study of the semantic generalization of Pavlovian conditioning.

Answer 13

They found that the CR to certain words would occur with similar sounding words, but would be even stronger to synonyms that did not sound like the original stimuli. This was an example of semantic generalization, since it was rooted in meaning and not physical characteristics of the words.

Question 14

Can generalization of the effects of extinction and punishment occur?

Answer 14

Yes. E.g., rats that learned to stop pressing a horizontal level were found to not push a vertical level either.

Question 15

Define excitatory stimulus generalization

Answer 15

when strengthening a response to a stimulus during training also strengthens responding to similar stimuli.

Question 16

Define inhibitory stimulus generalization

Answer 16

when weakening a response to a stimulus during training also weakens responding to similar stimuli

Question 17

Describe Honig and Slivka’s study of inhibitory stimulus generalization with pigeons.

Answer 17

They found that shocking pigeons when pecking disks of certain colours the overall tendency to peck all of the disks declined, despite reinforcing the pecking behaviour.

Question 18

What is discrimination?

Answer 18

Tendency for learned behaviour to occur in one situation but not other situations.

Question 19

Explain how discrimination and generalization are inversely related.

Answer 19

• the more discrimination, the less generalization.
• i.e., generalization gradients reflect the degree of discrimination, with a relatively flat gradient indicating little or no discrimination, and a steep gradient indicating much discrimination.

Question 20

What is discrimination training?

Answer 20

used to describe any procedure used to establish a discrimination.

Question 21

What is Pavlovian discrimination training?

Answer 21

One stimulus (CS+) is regularly paired with a US, and another stimulus (CS-) regularly appears without the US. For example, a dog will receive food at the sound of a buzzer (CS+) but not for a bell (CS-), thus learning the different consequences of each sound and only salivating for the buzzer.one stimulus (CS+) is regularly paired with a US, and another stimulus (CS-) regularly appears without the US. For example, a dog will receive food at the sound of a buzzer (CS+) but not for a bell (CS-), thus learning the different consequences of each sound and only salivating for the buzzer.

Question 22

What is operant discrimination training?

Answer 22

One stimulus (S+) typically indicates that a behaviour will receive a reinforcer, and another stimulus (S-) typically indicates that it does not receive a reinforcer.

Question 23

What are discriminative stimuli?

Answer 23

Stimuli in operant discrimination training that are associated with different consequences for behaviour (I.e., S+ and S-), with one being more reinforcing than the other.

Question 24

Define successive stimulus discrimination training.

Answer 24

S+ and S- alternate, usually randomly

Question 25

Define simultaneous discrimination training.

Answer 25

Discriminative stimuli are presented at the same time. This means that the subject must learn which stimuli to attend to.

Question 26

What is a matching to sample procedure (in relation to discrimination training)?

Answer 26

The task is to select from 2+ alternatives (I.e., comparison stimuli) and the stimuli that matches a standard (I.e., the sample). The comparison stimuli includes S+ and one or more S-. The subject use successfully discriminate between the stimuli that matches the sample (S+) and those that do not (S-).

Question 27

What is an oddity matching procedure?

Answer 27

Oddity matching is a variation of MTS that requires the subject to identify a stimulus different from the sample.

Question 28

What is errorless discrimination training?

Answer 28

procedure presents the S+, but presents the S- in a weaker form and for a short time. This makes it more difficult for the subject to identify/interact with the S-, meaning less likelihood of the subject committing an error. Errors are undesirable because they are demoralizing (punish the subject for making an effort). Gradually the S- is introduced for longer and/or in a truer form.

Question 29

What is the differential outcomes effect (DOE)?

Answer 29

Improved performance in discrimination training is a result of different consequences for different behaviours. Consistent differences in outcome for each behaviour results in faster discrimination and fewer errors.

Question 30

Define stimulus control.

Answer 30

When discrimination training brings behaviour under the influence of discriminative stimuli.

Question 31

Describe Pavlov’s theory of stimulus discrimination/generalization.

Answer 31

Discrimination training produced physiological changes in the brain, by exciting the area associated with CS+ and inhibiting the area associated with CS-. If a novel stimulus is similar to that of the CS+, it will excite an area close by, which will irradiate to the CS+ area and elicit the CR (response). Similarly with the CS- and similar stimuli. A similar explanation is used for operant learning. This, however, doesn’t have physiological evidence, only behavioural observations. This theory suffers from circular reasoning, so it isn’t popular.

Question 32

Describe Spence’s theory of stimulus discrimination/generalization.

Answer 32

A modification of Pavlov’s theory, by keeping the ideas of excitation and inhibition, but without physiology. Pairing CS+ with US results in increased tendency to respond to CS+ and similar stimuli. The inverse is true of CS- and US, resulting in a decreased tendency to respond to the CS- and similar stimuli. The generalization gradients that result is called an excitatory gradient/inhibitory gradient. He proposed the the response to novel stimuli would be accounted for by the net difference between excitatory and inhibitory tendencies, in that the tendency to respond to novel stimulus would be reduced by the tendency to not respond to that stimulus.

Question 33

Describe the Lashley-Wade theory of stimulus discrimination/generalization.

Answer 33

Generalization gradients depend on prior experience with stimuli similar to those used in testing. Discrimination training increases the steepness of the generalization gradient because it teaches the animal to tell the difference between the S+ and other stimuli. Some of this training occurs naturally in everyday life. This theory suggests that if an animal is prevented from having any experience with a certain stimulus, such as colour, its behaviour following training will be affected and the flatter the generalization gradient will be. Evidence regarding this matter, however, is unclear.

Question 34

What is a concept?

Answer 34

Any class in which the members share one or more defining features. The defining feature allows us to discriminate members of one class from members of another class.

Question 35

What does it mean to understand a concept?

Answer 35

To understand a concept means to be able to discriminate between stimuli that fall within the concept class and those that fall outside of it.

Question 36

What is transposition?

Answer 36

The subject learning to choose a relative option (via learning of a relational concept).
E.g., teaching chickens to choose the lightest of two grey squares, and then showing them another even lighter square which the chickens chose rather than the square that they were trained with

Question 37

Describe a typical mental rotation experiment.

Answer 37

This experiment involves showing participants an image, sometimes a letter, rotated in various ways. The greater the rotation, the longer it took for the people to answer/identify the object/letter. This suggests that people mentally rotate an internal representation of the object until it is in the correct position.

Question 38

How can mental rotation be interpreted as generalization?

Answer 38

These experiments form a graph similar to that of a generalization gradient, with shapes used in training taking the least amount of time to identify.

Question 39

Describe how Phelps and Reit flattened the mental rotation “generalization gradient.”

Answer 39

After continued training, the generalization gradients flattened because they had nearly perfected the task with practice and couldn’t improve anymore. This means that the mental rotation doesn’t explain the performance, but instead the covert behaviour involved. Reaction times are better explained by physical features of the situation and the learning history of the participant.

Question 40

Describe smoking and quitting smoking in terms of stimulus control relationships.

Answer 40

o People often relapse to smoking because of the associated stimuli in their environment that had previously been paired with smoking and have acquired a degree of stimulus control over cigarette use. Smoking is reinforced in a variety of situations, reinforcing the behaviour across a variety of places, times and circumstances.

Question 41

How can you use stimulus control principles to make it more likely that someone will quit smoking successfully?

Answer 41

The quitter might gradually expose themselves to situations/places while that they associate with smoking, preventing themselves from smoking. E.g., sitting in a non-smoking section of a restaurant if you associate smoking with having a cup of coffee.

Question 42

Describe the experimental neurosis in dogs that was initially discovered in Pavlov’s laboratory.

Answer 42

Experimental neurosis was used to describe a dog that had been tasked with discriminating near identical shapes. The dog had an intense outburst, similar to what Pavlov described as a human nervous breakdown, hence the name, by exhibiting behaviours that are uncharacteristic (or new) to the individual. This exists in humans.