How do humans detect the signal?
we all send and receive the signal every day
Signal processing theory
A theory built on the premise that it is difficult to distinguish between reliable information (Signal) and false information (Noise) in this world.
A theory that expresses statistically and logically the psychological state of choosing the content of information to be taken for decision-making
Radar signal processing
HIT: Bandit Appears Actually, Radar Detects Bandit Appearance (O-O)
Miss: Bandit appeared, bandit not detected (O-X)
False Alarm: Bandit not appearing, judged to have appeared (X-O)
Correct Rejection: Bandit Not Appearing, Bandit Not Appearing (X-X)
What is optimal signal detection?
If the appearance of the bandit is not detected, it is a case of neglecting the duty of vigilance, and if the enemy is bombarded, it is a fatal loss because life is lost.
For this reason, Radar Detectives will overuse “False Alarms” (if they detect even though they don’t appear) to reduce “Miss” (if they show up but don’t detect them). In this case, you must continue to mobilize even when the bandit does not appear. This can also increase the new cost and reduce the reliability of detection, resulting in a “shepherd boy” effect.
Even computers can misdetect signals and noise. If so, how is it most reasonable to accept external signals? Is our brain accepting the other person’s signals (speech, writing, display, image, etc.) as an optimal judgment?
Optimal Error Reduction Strategies
Ultimately, reducing Type I error (False Alarm) and Type II error (Miss) is the optimal signal detection strategy.
Type I error ( α): False Alarm
Falsely judging a fact that does not exist, judging a false fact as true, judging a fake as real
Type II error (β): Miss
Wrong judgment of facts that exist, the judgment of facts as false, the judgment of truths as fakes
Signal processing theory
Since a signal and noise come in with a specific probability of occurrence, this is expressed as a probability distribution curve.
Signal processing theory is significant in providing a scientific basis for human decision-making. In particular, the example of radar signal processing can help us reflect on our biases and how they affect decision-making. Concerns about how to handle information in interfaces are also an issue that can lead to problems.
