The Rescorla-Wagner model (Rescrola & Wagner, 1972) is a model of classical conditioning that was published in part to develop an associationistic theory that could account for Kamin's (1968) blocking effect. The model seeks a trial by trial explanation of how learning proceeds over a series of trials. The basis of the theory is the learning proceeds where there is a discrpancy between expectation of a unconditioned stimulus happening when a particular conditioned stimulus happens and what actually happens. If the expectation is matched, learning does not happen.
To get to deeper into the model, learning that occurs during classical conditioning is based upon several parameters:
the Salience of conditiones stimuli(S)m e.g., intensity,
the current strength of the US (A),
the
strength of the association between the various CS and the US (V) and the total learned
association between all of the present CS and the US (Vsum).
On each trial the change in the associative strength
between a CS and the US is determined by the following equations
ΔVi = Si x (A - Vsum)
where i indicates the particular CS. Salience is between 0 and 1 and the rest of the values are arbitrary.
In this simulation, you have two CSs that can be presented, a tone and a light. You control the salience of the CSs. You can also control the strength of the US (0 being a test trial).
Rescorla, R.A. & Wagner, A.R. (1972) A theory of Pavlovian conditioning: Variations in the effectiveness of
reinforcement and nonreinforcement. In A. H. Black & W. F. Prokasy (eds.), Classical Conditioning II:
Current Research and Theory,
pp. 64–99, Appleton-Century-Crofts.
Kamin, L.J. (1969). Predictability, surprise, attention and conditioning. In B.A. Campbell & R.M. Church
(eds.), Punishment and aversive behavior, 279–96, New York: Appleton-Century-Crofts
To see the illustration in full screen, which is recommended, press the Full Screen button, which appears at the top of the page.
Below is a list of the ways that you can alter the model and run the simulation. The settings include the following:
Test and Comparison Patches: at the top are two squares. They represent
the appearance of
the two stimuli to the simulated eye. Your job is to make the two
patches look the same in this simulated eye. The left-hand patch is the test stimulus, and the right
is the comparison. You will adjust the comparison to make it look like the test.
US: if selected you can run a trial with the US on. Delect and there will
not be a CS presented. The slider will control the intensity of the US.
CS: Light: select to turn on the light as a CS and the slider
controls the intensity or salience of the light CS.
CS: Tone: select to turn on the tone as a CS and the slider
controls the intensity or salience of the tone CS.
Number of Trials to Simulate: control the number of trials
that will be run when you press the Run Simulation
button.
Simulation Update Rate (ms): adjust the slider to control how fast the simuluate
runs when miltiple trials are run.
Run Simulate: run the simulation the number of trials indicated by the
Number of Trials to Simulate slider.
Run Trial, Run 5 Trials, Run 10 Trialz:
run the stimuluate the indicate number of trials.
Start Over: clear all of the trials and start the simulation from the beginning again.
Pressing this button restores the settings of the model to their default values.