Experiments in psychophysics explore the limits of our senses.  Some of the earliest questions in psychology centered on the question of our relationship to the outside world.  How real is reality?  It often seems to depend on who and how you ask.

Are we influenced by things we don't know we see or hear?  Do subliminal stimuli really influence behavior?  Juggling the influence of the senses, transmission, processing, and recognition has created a wealth of research.  Is an item subliminal if the reticular formation edits the information as irrelevant?  What roles do attention and conscious process have on pure sensation?

There are several research methods which have been developed in psychophysics.  People's behavior changes depending on how they are asked to judge the stimuli, how the stimuli are presented to them, and how the responses are collected.  Here are three classic methods which approach the problem in slightly different ways.

Method of Adjustment:

The subject sees a standard line and is asked to adjust another line to match it.  The subject can lengthen or shorten the line until they are satisfied.

Psychologists are interested in the bias toward and against the original length of the adjustment line.  If the line was longer than the standard line, the subject is more likely to say the two lines are equal when the adjusted line is a little longer.  This is called the expectation error.  If the subject overshoots the mark and makes the adjustment line a little shorter most of the time, this is called the habituation error.

The context of the sensation influences the perception of the stimuli.  Does this hold true if the subject knows and understands the bias?  Is it a tactical error resulting from the eye movements used for comparing the two stimuli?  Are there subjects who habitually overshoot rather than undershoot the target?
 

Method of Limits:

The habituation and expectation errors are also found in the Method of Limits.  The subject sees a standard line and a comparison line.  They are asked to say if the comparison line is longer, shorter, or the same as the standard.  The comparison line adjusts accordingly.  If the comparison line is longer, it will shorten if the subject says it is longer.  It will continue to shorten if the subject again says it is either longer or equal to the standard stimulus.  If the subject says that it is shorter, the trial is over and the next trial begins.

In both the Method of Adjustment and the Method of Limits the habituation and expectation errors are counterbalanced by including equal numbers of trials with longer and shorter lines.
 

Method of Constant Stimuli:

The subject sees a comparison and a standard stimulus but cannot adjust them.  The subject simply indicates whether the comparison is longer or shorter.  This method is also influenced by a response bias; the tendancy to prefer to say shorter or longer when unsure.  The other two methods were developed to minimize this type of response bias but brought in other bias errors.

The method of constant stimuli requires many more trials to obtain an estimate of the JND and PSE than the other two.  Because the subject is making a forced choice, longer or shorter, you have to have a large number of trails to determine the level of chance.  The more trials you have for each comparison length, the more accurate the data.
 

 
Possible variations include:
 
1. Changing the length of the comparison and/or standard lines.
2. Changing the horizontal distance between the two lines.
3. Changing the vertical distance between the two lines.
4. Changing the number of trials.
5. Changing the conditions or environment -- instructions, setting, etc.

• Contrasting performance depending on the length of the line.
• Giving feedback concerning performance to see if it increased accuracy.
• Contrasting the different methods in a within subjects study to test for skill consistency.
• Testing Right versus Left handed subjects to test to see if there were differences in response bias.

We have both the computer based visual Psycholophysics program and a set of comparison weights which can be used -- or you can create a measure to test another modality.

You can find the Psychophysics program in a white ring binder next to the computer in my office and next to the computer on the desk on the landing.  The program is on a 5 1/4 inch disk.  Instructions are on the first page of the notebook.

The program includes complete instructions and a tutorial for you to use.  It will calculate your data for you and let you print the bar graphs.  To print a bar graph, simply hit the "Print Screen" button at the top right of the keyboard.  You should use the Epson dot matrix printer for best results.

There is an extensive literature on psychophysics and there has been increased interest in the area over the last five years.  I have a book by Dember which will give you some basic background.  The Psychological Abstracts and Psych Info will give you tons of references.  When you do a search, start with psychophysics and then limit it to your modality -- vision, touch, taste, weight.

© Marilyn Shea, November 1999
Department of Psychology, University of Maine at Farmington