Signal Detection Theory

Sensitivity 

teenage girl with Misophonia, specifically related to eating sounds like chewing, swallowing, popping gum, etc., lives in a room which is across the hall from her family’s dining room. To drown out the eating sounds, she plays music on her speakers. The ogive figure below illustrates the noise (eating sounds) and signal + noise (eating sounds + background music) distributions. 

Signal + Noise 

Noise 

Criterion 

A liberal criterion generally produces more hits and correct rejections than misses and false alarms. Hence, using a liberal criterion is the best approach to this problem. The sensitivity depends upon the sensations or perceptions of the observer upon actually receiving the signal and in the case of the teenage girl the value is large. This means the girl is quite biased towards detecting the low frequency chewing sounds even as the music plays. 

However, to a random stranger who spends sometime in the girl’s bedroom while anybody in the dining room is eating, the criterion is neutral and the sensitivity towards detecting such sounds is much less than the girl’s as illustrated in the diagram below. 

A neutral criterion produces equal percentages of misses and false alarms as well as of the hits and correct rejections. This distribution reveals less sensitivity towards the low frequency eating sounds as the stranger is unbiased. Thus, the stranger is much less likely to detect the low frequency eating sounds than the girl.