Discrimination & Generalization Flashcards
1
Q
Operant antecedents
A
- Discriminative stimuli (SD) – signal that when present
responses are reinforced; when absent responses are
not reinforced; occasion setters - Still don’t elicit behavior, just change probability
- Light (SD) : Press Lever (R) →Food (SR
) - Discriminative stimuli for punishment (SP
) – signal that
when present responses are punished; when absent
responses are not punishment - Light (SDP) : Press Lever (R) → Shock (SP
) - Discriminative stimulus (antecedent), operant behavior
(behavior), & consequence = three- term contingency
2
Q
Discriminative stimulus
A
- Individual is trained to give the target behavior only
when a particular stimulus is presented - Spray bottle: Barking →Sprayed
SDP R SP - Cop car: Speeding → Ticket
SDP R SP
3
Q
- Signal extinction (S∆)
A
- Behavior is reinforced in the presence of the
desired stimulus but not in the presence of other
stimuli - Kids at school:swearing → classmate approval
SD R SR
Grandparents:swearing → no approval
SD/∆ R
4
Q
CC – OC overlap
A
- Tone (NS): Food (US)→ Salivation (UR)
- Tone (CS)→ Salivation (CR)
- Tone: Lever press → Food
SD R SR - Response matters
- Tone is CS to salivation, but SD for lever pressing
5
Q
Stimulus control
A
- Behavior consistently reinforced in the presence of a
particular stimulus, that stimulus affects the probability of
the behavior occurring - These are essentially cues to behavior
- Cueing what consequences would be
- Presence of the discriminative stimulus (SD) affects
probability of the behavior - Does not elicit the behavior itself, just signals that
reinforcement is available. - Examples
- 2000 Hz Tone : Lever Press → Food
- Bar : Smoking Urge → Smoking
6
Q
Generalization
A
- Tendency for an operant response to be emitted in the
presence of a stimulus that is similar to the
discriminative stimulus (SD) for that response - Example: If you have learned to beg for candy in a
grocery store, you might also do it in a convenience
store - More similar the stimulus to the SD the more likely the
response - Example: Rat is rewarded with food for lever pressing
in the presence of a 2000Hz tone. More likely press
the lever in presence of 1800 Hz tone than 1000 Hz
tone
7
Q
Generalization gradient
A
- Depiction of strength of
responding in the
presence of stimuli
similar to the SD - Strongest responding to
the SD - Steeper gradients = less
generalization and more
discrimination - Shallow gradients = more
generalization and less
discrimination
8
Q
Stimulus Generalization
A
- Physical Similarity
- Similar context
- But what if learning context differs?
- Stimulus class or concept
9
Q
Stimulus class
A
- Common-element stimulus class
- Set of stimuli, all of which have some physical
characteristic in common
*All 4 legged animals are “dog”
*All vehicles are “truck” - Conceptual behavior – emitting behavior to all
members of a common-element class, but not
those that don’t belong to the class - Stimulus Equivalence Class
- Set of completely dissimilar stimuli learn to
respond to in the same way (learned to group)
10
Q
Discrimination
A
- Tendency for an operant response to be emitted
more in the presence of one stimulus than another - Opposite of generalization
- Example: If you have learned to beg for candy in a
grocery store, but you don’t do it in a convenience
store - Example: Rat is rewarded with food for lever
pressing in the presence of a 2000Hz tone. More
likely press the lever in presence of 1800 Hz tone
than 1000 Hz tone
11
Q
Discrimination training
A
- While the discriminative stimulus signals that
reinforcement is available, other stimuli become
signals that reinforcement will NOT be available. - These other stimuli are called “discriminative
stimuli for extinction”
12
Q
- Discrimination training (operant procedures)
A
- When responding is NOT reinforced in the
presence of certain stimuli those stimuli become
discriminative stimuli for extinction (symbol = S∆
) - Example
- 2000 Hz Tone (SD) : Lever Press (R) → Food (SR)
- 1200 Hz Tone (S∆) : Lever Press (R) → No Food (-)
13
Q
Effective Stimulus Discrimination Training
A
- Choose distinct signals
- Minimize opportunities for errors
- Maximize the number of trials
- Need multiple trials to learn the behavior
- Arrange for several reinforced responses in the
presence of the SD - When the S ∆ is presented, make the change from
the SD very obvious and follow the rules for
extinction for the behavior of concern - Make use of rules: describe the
contingencies - Rules can speed up learning
- Wean the individual from the program
- Plan natural reinforcers
- Plan periodic assessments of behavior to make sure it is
occasionally being reinforced and that the desired frequency of
the behavior is being maintained in the presence of the SD
14
Q
Limitations of discrimination
training
A
- When discriminating SD from SΔ
, Ps will make
mistakes - Can result in frustration, emotional behavior etc.
- Fading & errorless discrimination learning
- Can reduced number of errors to SΔ
if: - SΔ
is introduced early in training, soon after SD
- SΔ
- SΔ
is started weak then gradually strengthened
(fading)
- SΔ
- Example
- Pigeons trained to peck at red key (VI 60-sec)
- Once behavior established 5 sec extinction intervals
implemented (light key switched off) - Reinforcement and extinction sessions gradually
increased to 3-mins each - Dark key then gradually illuminated with green tinge
- Almost no errors made to S Δ
relative to control group
15
Q
Discrimination & Errorless training
A
- Practical application
- In humans
- Enhance learning of math skills (Haupt et al., 1975)
- Setting up study areas
- Helping with insomnia
- In animals
