Janis and Feshbach conducted a ground breaking experiment into the effect of fear or anxiety producing appeals within a message. Their study presented four groups of students with different messages regarding dental health. The control group received no fear arousal, while the other groups received mild, moderate, and strong fear appeals.
The content of these messages varied from the minimal fear inducing techniques where the students were warned of the results of not using the proper toothbrush, to the strong fear appeals where the students viewed a film and witnessed horrifying tooth decay.
The results in the following table reveal how fear appeals are most effective in gaining the desired attitude when the fear appeals are minimal.
|
Strong (%)
|
Moderate (%)
|
Minimal (%)
|
Control (%)
|
|
| Increased Conformity |
28
|
44
|
50
|
22
|
| Decreased Conformity |
20
|
22
|
14
|
22
|
| Net change in Conformity |
+8
|
+22
|
+36
|
0
|
By comparing stated habits prior to and after the communication, the students attitudes were most effected by the minimal amount of fear induction.
Hovland points out that these results may not be inclusive of all of the possibilities of fear arousal. For instance, if the communicator has a great deal of power over the receivers such as the ability to punish them, then clearly and strong fear appeal may be most persuasive. The "optimal dosage" of fear, however, does seem to indicate a low amount of anxiety arousal.