Aversive Control

1
Aversive Control

• Negative Reinforcement
• Avoidance Learning
• Escape Learning

2
Negative Reinforcement
Removes/Prevents
S e.g. operant chamber
R e.g. bar press
SAversive e.g. shock
Strengthens
Negative Contingency
p(SAversive/R) lt p(SAversive/noR)
Note if R removes SAversive Escape
if R prevents SAversive Avoidance
3
Discriminated or Signalled Avoidance
A warning stimulus signals a forthcoming SAversive
If the required response is made during the
warning stimulus, before the SAversive occurs,
the subject avoids the shock.
If a response is not made during the warning
stimulus, the SAversive occurs, and terminates
when the required response is made
e.g., one-way avoidance
two-way avoidance
4
The Two-Process Theory of Avoidance
Explains avoidance learning in terms of two
necessary processes
1. The subject learns to associate the warning
stimulus with the SAversive
2. The subject can then be negatively reinforced
during the warning stimulus
Thus the two-process theory reduces avoidance
learning to escape learning the organism learns
to escape from the CS and the fear that it
elicits.
5
The Two-Process Theory of Avoidance
Acquired-drive experiments support the
Two-Process Theory of Avoidance since animals do
learn to escape from the CS
6
Problems for the Two-Process Theory of Avoidance
1. Level of fear is not always positively
correlated with avoidance
2. Avoidance behavior should cycle at asymptote,
but it typically does not.
3. Avoidance behavior should not be learned if
the response does not terminate the CS, but it is.
4. Animals can learn Free-Operant (or Sidman)
avoidance
7
Alternative Theoretical Accounts of Avoidance
Behavior

• Positive reinforcement through Conditioned
  Inhibition
• of fear

• Conditioned Safety Signals

• performance of the avoidance response results in
• distinctive feedback stimuli (i.e., spatial cues,
  tactile
• cues, etc)

• the avoidance response produces a period of
  safety
• allowing the feedback cues to become signals for
  the
• absence of shock (i.e., safety signals)

• since a shock-free period is desirable, a
  conditioned
• inhibitory stimulus for shock could serve as a
  positive
• reinforcer

8
Alternative Theoretical Accounts of Avoidance
Behavior
2. Reinforcement of avoidance through reduction
of shock frequency

• with Two-Process theory, reduction in shock
• frequency was by-product of avoidance responses

• reduction in shock frequency is important

• rats will press lever to reduce frequency of
  shocks
• from 6/min to 3/min

9
Alternative Theoretical Accounts of Avoidance
Behavior
3. Avoidance and Species-Specific Defense
Reactions (SSDRs)

• more concerned with the actual response and what
• determines the animals response early in training

• aversive stimuli elicit strong innate responses
• (i.e., freezing, flight, fighting, seeking out
  dark areas)

• species typical responses are readily learned as
  
• avoidance responses (e.g., Jump out of box in two
  
• trials versus press a lever to avoid shock 1000s
  of trials)

• punishment is responsible for the selection of
  the
• avoidance response

10
Punishment
Produces
S e.g. operant chamber
R e.g. bar press
SAversive e.g. shock
Weakens
Positive Contingency
11
Punishment
Usually a response that must be punished is
maintained by a positive reinforcer, thus
experimentally SAversive is usually made
contingent on a response that has been or is
being positively reinforced.
Both Skinner and Thorndike claimed punishment was
not very effective in suppressing behavior
12
Skinners Experiment on Punishment
Phase 1
Rats were reinforced with food on a VI schedule
Phase 2
Extinction for 120 minutes on two successive days
During only the first 10 min of extinction on day
1 one group of rats was punished for each bar
press (paw slapped) the other, control, group
was not punished.
13
Results of Skinners Experiment on Punishment
Punishment suppressed responding while it was
being administered, but when punishment stopped,
the punished rats ended up making as many
responses overall in extinction as the unpunished
controls.
14
Skinner concluded that punishment was not an
effective way to control behavior.
15
Consideration of administration of punishing
stimulus Punishment effective if punishment

• is intense/prolonged from start

• is response contingent rather than response
  independent (fig 10.13)

• occurs immediately after response rather than
  delayed

• is on a continuous rather than partial
  reinforcement schedule (fig 10.14)

16
Consideration of response to be punished
Punishment effective if

• punished response is not being reinforced or
  motivation for reinforcer is reduced

• there is an alternative response to the punished
  response to acquire reinforcer (fig 10.15)

• the punished response is not a species-specific
  defence reaction

17
Consideration of situation in which punishment is
to be administered punishment effective if

• subject cannot discriminate when punishment will
  be administered and when not

• punishment does not signal SAppetitive

18
Problems that may arise with the use of
punishment to eliminate behavior

• undesirable CERs to the situation and/or person
  associated with punishment

• general suppression of responding

• difficulties in applying punishment
  systematically so that it is effective (e.g.,
  discriminative cues punishing every instance of
  the behavior

• imitation of the aggressive behavior involved in
  punishment

• escape/avoidance or aggressive responses in
  punishing situation

19
(No Transcript)
20
Practice Exams
Midterm
21

• According to the Rescorla-Wagner model, learning
  will only occur
• if an animal has experienced a(n)
  .
• A. CS
• B. US
• C. UR
• D. CR

• Suppose a 5-second tone is presented, then a
  5-second gap, then
• food. Now 60 seconds pass, and the tone-gap-food
  sequence is again
• presented. In this example, the CS-US intervalis
  seconds, and the
• intertrial interval is seconds.
• A. 10 60
• B. 10 70
• C. 5 60
• D. 5 70

22

• You have discovered a new species of creatures,
  the zorks. They
• eat birds and identify those birds that are
  edible by their call. Zorks are
• bothered by stinging insects who mark their
  territory with a sour fluid
• The zorks use this taste to avoid the insects
  territory. Suppose you
• perform the bright, noisy, and tasty water
  study, in which
• light/tone/saccharin CSs are paired with USs of
  shock or poison.
• If you assume the nonequivalence of
  associability, then the zorks who
• got USs should stop drinking only during
  a cue.
• A. Shock light/tone
• B. Shock saccharin
• C. Poison saccharin
• D. Poison light

23

• You have established a tone as a conditioned
  inhibitor (a CS-),
• using a shock US. Which of the following
  procedures would be
• most likely to cause the tone to lose its
  inhibitory power?
• A. Present the tone alone for 20 trials
• B. Present the tone alone for 200 trials
• C. Present 100 trials where the shock is
  followed by the tone
• D. Present 100 trials where shocks and tones
  are given randomly

• Suppose animals in Group 1 are exposed to a
  number of electric
• shocks, while animals in Group 2 are not. Next,
  all animals are given
• tone-shock pairings. What is the typical result?
• A. Both groups acquire a CR at the same rate
• B. Group 1 acquires a CR more quickly than
  Group 2
• C. Group 1 acquires a CR more slowly than Group
  2
• D. Group 1 develops an inhibitory CR

24
6. Which of the following produces the strongest
conditioning? A. Simultaneous conditioning B.
Backward conditioning C. Trace conditioning D.
Delayed conditioning
7. Which of the following is an example of a
CR? A. Salivating when lemon juice is put on
your tongue B. A pigeon pecking grain C.
Feeling nauseated when seeing moldy food. D.
Flinching when a tree limb falls near you.
25
8. Which of the following is an example of an
unconditioned response? A. You run when someone
yells, Fire! B. Your mouth waters when you
think about chocolate cake. C. You jump when a
balloon pops behind your head. D. Your dog wags
its tail when you open a can of dog food.

• When did Pavlov present the sound of a metronome
  and food
• powder?
• A. Whenever the dog was hungry
• B. When the dog was quiet and not reacting to
  other stimuli
• C. As soon as the dog salivated
• D. Independent of the dogs behavior

26

• According to contingency theory, inhibitory
  conditioning occurs
• to a CS only when
• A. CSs and USs never occur together.
• B. Unsignaled USs are more likely than signaled
  USs.
• C. Signaled USs are more likely than unsignaled
  Uss
• D. The likelihood of getting a signaled US is
  the same as the
• likelihood of getting an unsignaled US.

27
1 (4). Extinction entails the elimination of
an association. Evaluate this quote, and
provide evidence supporting your conclusions. 2
(3). Pavlovian conditioning is merely a way of
teaching conditioned reflexes. Evaluate this
quote. 3 (2). Describe an experiment that
supports the S-R theory of second-order
conditioning. 4 (4). What advantage does the
Pearce-Hall model have over the Rescorla-Wagner
model in explaining data from unblocking
experiments? 5 (2). Describe two common
procedures for measuring conditioned inhibition.
28
Practice Exams
Final
29

• Which of the following procedures results in a
  decrease in an
• instrumental response?
• A. Avoidance conditioning
• B. Escape conditioning
• C. Omission training
• D. Reward training

• Suppose a study is conducted with rats. In phase
  1, Group T-F is
• given tone-food pairings, while Group F is only
  presented with food.
• In phase 2, a lever is inserted, and each lever
  press is followed by the
• tone. What would convince you that the tone is a
  conditioned
• reinforcer?
• A. Either group pressed the lever in phase 2
• B. Group F failed to press the lever
• C. Group T-F pressed the lever in phase 2
• D. Group T-F pressed the lever more than Group
  F

30

• Punishment in operant conditioning is analogous
  to excitatory
• aversive conditioning in Pavlovian conditioning.
  What, in the
• punishment procedure, is analogous to the CS, in
  Pavlovian
• conditioning?
• A. The aversive stimulus
• B. The operant response
• C. The suppression of the response
• D. The reward

• Your friend is attempting, unsuccessfully, to
  teach her dog to shake
• hands using an operant conditioning procedure.
  You are concerned with
• contiguity, so you advise your friend
• A. To give the treat immediately after the dog
  responds
• B. Not to wait too long between saying shake
  and giving the treat
• C. To first reinforce any movement of the right
  paw
• D. To swat the dogs nose if it fails to
  quickly offer its paw

31

• Whenever a light comes on, a rats lever press is
  followed by a food
• pellet. To begin the extinction phase, the
  researcher should
• A. Stop turning on the light
• B. Stop delivering food after lever presses
• C. Remove the lever
• D. Both A and B

• The type of schedule that typically produces a
  pause and then an
• accelerating rate of responding is the
  schedule of
• reinforcement.
• A. Fixed interval
• B. Variable interval
• C. Fixed ratio
• D. Variable ratio

32
7. You have discovered a new species, which eats
mosquitos, and you have observed a number of its
behaviors in the wild, including jumping (to
catch mosquitos) and digging (before going to
sleep). You are now testing whether
instrumental conditioning occurs for the species
and try each of the following 1. Jumping is
followed by presenting two mosquitos 2.
Digging is followed by presenting two mosquitos
and 3. Pressing a lever is followed by
presenting two mosquitos From most to least, what
do you predict about how much each behavior will
increase with the above contingencies? A.
Jumping, lever pressing, digging B. Lever
pressing, jumping, digging C. Digging, lever
pressing, jumping D. All behaviors should
increase the same amount
33

• If Jason sets the table, he does not have to wash
  the dishes.
• Which procedure does this exemplify?
• A. Avoidance conditioning
• B. Escape conditioning
• C. Omission training
• D. Punishment

• If an animal experiences independent
  of its behavior,
• it later has trouble learning a response in a(n)
  situation.
• A. shocks avoidance training
• B. Food reward training
• C. Neither A nor B
• D. Both A and B

34
10. Two schedules which selectively reinforce
long IRTs are A. VR and VI B. VI and
DRH C. VI and DRL D. VR and DRL
35
1 (3). Explain why the term reinforcement is
not defined in terms of specific pleasurable
stimuli such as food. Explain how you would
know for sure that a given stimulus is indeed a
reinforcer. 2 (4). Describe the conditions
under which matching behavior does and does not
occur. Describe the two principal explanations
for matching. 3 (4). In an operant
conditioning experiment, associations can
develop between the discriminative stimulus and
the response (S-R association) and/or between
the discriminative stimulus and the outcome (S-O
association). Describe an experiment that
demonstrates each of these associations (i.e.
one experiment for each association). 4 (4).
Describe an experiment demonstrating the basic
learned helplessness effect. Does the learned
helplessness learning deficit result from lack
of control over outcomes? (Be sure to support
your answer with experimental evidence).
36
1 (3). In any experimental situation, it is
necessary to determine if the response you see
is due to conditioning, or is a by-product of
some other variable (i.e. pseudo-conditioning).
Describe three control procedures used in
Pavlovian conditioning. 2 (4). How does the
Two-factor theory explain avoidance
conditioning. Describe two problems for this
theory. 3 (4). A general process view of
learning suggests that any stimulus/response may
be conditioned using Pavlovian/operant
techniques. Discuss the extent to which such a
claim is justified, providing experimental
evidence to support your argument. 4 (4).
While both contiguity and contingency are
important in conditioning, it is thought that
contingency is more important. Describe one
operant and one Pavlovian experiment that
demonstrate the importance of contingency.