Transcript and the Shuttle Box Experiment The Shuttle Box Experiment

Operant Conditioning – Chapter 9
Theories of Learning
October 19, 2005
Class #25
Lecture Outline



I. Shuttle Box Experiment
II. Avoidance paradox
III. Two-process Theory of Avoidance
Solomon and Waynne (1953) and the
Shuttle Box Experiment
The Shuttle Box Experiment

Shuttle box consists of:





Experimental Session – 10 trials each





2 chambers separated by a barrier several inches high
Could freely move between chambers
Separate lights
Metal floor that could be electrified to deliver shock
Dog placed in one compartment of a shuttle box
After a time, light in the compartment that dog was in went off, while other
light stayed on
10 seconds later, the dog received a shock through the floor until jumped
over barrier
Measured response latency (how long it took dog to jump over barrier once
light went out)
What they found:


Escape first: First few responses were usually longer than 10 seconds. So
dog was getting shocked and then escaping
Avoidance response: By roughly the 5th trial, dogs response latency was
less than 10 seconds. So dog never received shock in these cases.
Avoidance Paradox

Sidman (1989)

“Successful avoidance meant that something – the shock – did not
happen, but how could something that did not happen be a
reinforcer?”
 Sidman goes on to say that things are not happening all the time
and seems illogical to say that things that did not happen explain
things that did
 Escape makes sense because stimulus present
 Avoidance involves no change in shock condition
 No shock pre- and post- behavior
Two-Process Theory








There is no paradox: No such thing as avoidance, just escape. In both
cases, fear is being escaped
Two kinds of learning involved with avoidance: Pavlovian and operant
Dog learns to jump hurdle to escape the shock – Jumping is negatively
reinforcing
What reinforces jumping when there is no shock to escape?
Classical conditioning:
Shock (US) ---- Fear (UR)
Extinguished Light (CS) ---- Fear (CR)
Removal of fear evoking CS is observable change in the stimulus
environment that acts as negative reinforcer
Miller (1948)



Escape from an aversive stimulus can reinforce behavior
Through association with shock, the white compartment had become a
CS for fear, and escaping from it was reinforcing
First part:




Rats in white chamber shocked
Chamber had door that rat could jump through to get to black chamber
where no shock
Put rats back in white chamber and did not shock them; Rats continued to
go to black chamber
Second part:



Miller then put rats into white compartment and closed door
Rats could open door by running on a wheel
Rats soon learned to run on wheel to open door, even though no shocks
presented
Discriminated or Signalled Avoidance
SAversive

A warning stimulus signals a forthcoming

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
The Two-Process Theory of Avoidance
• Mowrer (1947, 1960)
• 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
Problems with Two-Factor Theory


Avoidance with no sign of fear
Problem: Fear should be greatest when avoidance
responses the strongest



Signs of fear may include heart rate increase, changes in breathing,
sweating
If theory correct, should be able to observe
 1.) Increase in fear when signal for shock presented
 2.) Decrease in fear once avoidance response made
Evidence that observable signs of fears begin to disappear
as subjects become more experienced with the task


Example: Soloman and Wynne (1953) noted that dogs initially show
signs of fear (urination) when light turned off
These signs disappeared once they making frequent avoidance
response
Problems with Two-Factor Theory




Failure of avoidance behaviors to extinguish
Two-Factor model uses classical conditioning.
From a Pavlovian perspective, each trial that shock is avoided is a
extinction trial
Extinction: CR gets weaker and disappears the more that CS not paired
with US
 Avoidance consists of presentation of CS without US
 Example: CS (Darkness) without US (Shock)
Problems with Two-Factor Theory


Two-Factor model:
 Predicts that avoidance responding should
eventually deteriorate after series of trial
without shock. We would expect cycles of
conditioning.
Problem:
 Avoidance behavior is extremely resistant to
extinction
Anxiety Conservation Hypothesis

Solomon & Wynne (1954)


Proposed a modification of two-process theory
to account for why extinction does not occur
In avoidance conditioning:

The organism avoids the stimulus so quickly
that there is very little exposure to the CS, so
extinction does not occur, and in fact,
incubation may occur (an increase in fear due
to the brief exposure to the fear-eliciting CS)
Avoidance Conditioning and
Phobias

Phobia:

A strong, persistent , and unwarranted fear of
some specific object or situation
Avoidance Conditioning and
Phobias

Phobias are thought to be a classically
conditioned response


Extinction doesn’t ever seem to occur…
Why???
Lab experiments compared to real world…

Mineka (1985)

Illustrates some differences:


In the lab the animal avoids the UCS (the shock)
In human phobic conditioning the person avoids the
CS (the dog – not the attack)
But…

Stampfl (1987)



Fear response in phobias is often
established whith only one brief pairing of
Cs/UCS
The organism must therefore avoid both
CS and UCS
The organism will then likely show
successful avoidance on 100% of these
trials
Credits:



www.olemiss.edu/courses/psy309/Negative.ppt
http://dogsbody.psych.mun.ca
www.radford.edu/~pjackson/avoidancephobias2.pdf