Title
Agent-based modeling for investigating adaptivity of misperception
Abstract
Misperception is a term which is generally used in a negative sense. However, when information promoting a certain kind of behavior is obtained, it can happen that the diversity of the behavior is enhanced by the misperceptions of the individual in the population, resulting in a situation in which misperception works advantageously for the population. This paper presents this view in terms of four hypotheses, dealing with (1) the basic adaptivity of misperception, (2) basic properties of communication, (3) the adaptivity of misperception in communication, and (4) the behavioral specificity of information in the adaptivity of misperception. A simple agent model for the resource-searching problem is constructed. Direct misperception, which is misperception in the direct acquisition of information from the surrounding environment, and indirect misperception, which occurs when information is obtained through communication, are considered. Their effects are investigated by simulation experiments. It is shown that misperception enhances the diversity of agent behavior and can contribute to adaptivity. It is also shown that exact communication may decrease the diversity of agent behavior, and that adaptivity is decreased when false information is shared. A tendency for the adaptivity of misperception to decrease when the behavioral specificity in information terms is low is demonstrated. We believe that study of the adaptivity of misperception as a factor generating diversity will lead to new findings in cognitive science, memetics, and fundamental aspects of engineering. © 2006 Wiley Periodicals, Inc. Syst Comp Jpn, 37(12): 96–106, 2006; Published online in Wiley InterScience (). DOI 10.1002/scj.20266
Year
DOI
Venue
2006
10.1002/scj.v37:12
Systems and Computers in Japan
Field
DocType
Volume
Population,Agent-based model,Computer science,Cognitive psychology,Agent behavior,Artificial intelligence,Memetics,Machine learning
Journal
37
Issue
Citations 
PageRank 
12
0
0.34
References 
Authors
3
2
Name
Order
Citations
PageRank
Jin Akaishi1101.36
Takaya Arita213042.34