Decades of research indicate that humans are not rational decision-makers. Our decisions and assessments of situations we encounter and other individuals or groups are sometimes flawed because they are based on a limited acquisition and rational analysis of information, and strongly influenced by our past experiences. We develop in this paper mathematical models of human decision-making that incorporate the effect of cognitive biases. These models start from an optimal Bayesian decision making algorithm and modify it to account for cognitive biases and the effect of past information seen by the individual. Next, we show how it is possible to mitigate cognitive biases in binary hypothesis testing problems by properly selecting and sequencing information presented to an individual.
[1]
Ahmed H. Tewfik,et al.
Optimal ordering of observations for fast sequential detection
,
2012,
2012 Proceedings of the 20th European Signal Processing Conference (EUSIPCO).
[2]
Cliff C. J. Huang,et al.
A general model of starting point bias in double-bounded dichotomous contingent valuation surveys
,
2005
.
[3]
P. Weller,et al.
Quantifying Cognitive Biases in Analyst Earnings Forecasts
,
2002
.
[4]
Xiaojin Zhu,et al.
Cognitive Models of Test-Item Effects in Human Category Learning
,
2010,
ICML.
[5]
Thomas L. Griffiths,et al.
"Burn-in, bias, and the rationality of anchoring"
,
2012,
NIPS.