Decision Making

Decision Making

Today we talking about decision making :

the signal : her own investment data ; the noise : the other data on the billboard 

Hit : she get the signal see the price up , then hold the positive response to sell her stock on the other hand and buy more stoke when the price is lowest . 

Miss : she get the signal see the price up , but she hold negative response miss the time  to sell her stock on the other hand and buy more stoke when the price is lowest . 

False alarm : she don`t get the signal but make a mistake thought the price up , then she make a positive response  to sell her stock on the other hand and buy more stoke when the price is lowest . 

Correct rejection : she don`t get the signal of the price up , and she have a negative response do nothing  to sell her stock on the other hand and buy more stoke when the price is lowest .

if she miss the buying and selling , it must be D-prime is small and Beta is high ; if she hold false alarm , it must be D-prime is small and Beta is small .

D-prime -- perceptual sensitivity 

large D-prime is the signal is more discriminable from the noise ,so more correct responses make hit and correct rejection .

If she want large D-prime , it`s better change her data color ,or maybe build a system on her phone when the data change to alarm her .

Beta--the decision criteria 

The degree to which the perciever is biased to detect or not to detect : 1. conservative ( large ) beta - minimal detection 2. liberal ( small ) beta - maximal detection 

Bias : if more correct rejection , more miss , say " no " all time ; if more hit , more false alarm , say " yes" all time .

Her bias is focus on the billboard all the time ,and want the price up .

Beta increase : the perceiver renders the criteria more strict , then the perceive is less open to detect the signal biased to reject it ( conservative ) , so more miss and correct rejection .

Beta decrease : the perceiver decreases the criteria ,then the perceiver is more open ( biased ) to detect the signal ( liberal ) , so more hit and false alarm .

p[Hit]+p[Miss]=1  ;  p[False Alarm]+p[Correct Rejection]=1

The expert said no loud music between 2-4 , in order to have a large D-prime to make more hits and more correct rejections .

For example : anticipating a phone call , then you will have a smaller beta-the decision criteria would become lower ( liberal ) , so will meet the false alarm and hits ;avoding a phone call while engaged in a different task , then the decision criteria increase-beta increase ( conservative ), so will meet more miss and correct rejections .