Neuroscience Animations

John H. Krantz, Hanover College,

Using the Media






Skin Senses

Statistical Concepts

Hanover College
Psychology Department

Receiver Operating Characteristic

Brief description and instructions (DRAFT):

This applet illustrates the Receiver Operating Characteristic (ROC curve).  

The left hand graph is very similar to the Decisions in Signal Detection Theory applet showing hits and false alarms highlighted.  You can refer to that applet for instructions on how that part of this applet works.

The right hand graph is the Receiver Operating Characteristic graph.  The ROC curve is a plot of all the possible combinations of hits and false alarms for a given sensitivity to a signal.  False alarms are plotted on the x-axis and hits are plotted on the y-axis.  As the criterion is changed, the hits and false alarms will both change.  When the criterion is to the right hand side of the signal+noise curve it is called strict and both hits and false alarms will be low.  When the criterion is to to the left side of the noise curve, hits and false alarms will both be high.  As you move the criterion, the yellow dot on the ROC curve will move along the dimmer yellow curve which is the ROC curve.  Changing sensitivity determines how close the ROC curve is to the vertical diagonal line on the ROC graph.  The vertical diagonal line represents when there is no sensitivity to the signal.

Click here to open the applet.  It will open a new window that will fill your screen.