TY - UNPB
T1 - Talking the talk, not walking the walk
T2 - The co-evolution of overconfidence and loss aversion
AU - Dawson, Chris
AU - de Meza, David
PY - 2023
Y1 - 2023
N2 - Perhaps the two most prominent human cognitive "biases" are unjustified self-belief, termed overconfidence, and loss aversion, a special form of reference dependence whereby an outcome is less attractive if perceived as a loss. As evolution tends to make adaptations that enhance fitness, the puzzle is why these biases arise, especially as they tend to offset each other. Specifically, overconfidence promotes risk-taking, such as starting a business, whilst loss aversion curbs it. Is evolution just burying its own mistakes? We propose an alternative reconciliation of the evolutionary basis of these seemingly conflicting biases. Following Robert Trivers, overconfidence arises to positively influence the behavior of others. The downside is that if acted on, overconfidence leads to mistaken decisions. Our point is that loss aversion limits how far these evaluation errors are implemented. Importantly, loss aversion corrects decision errors due to overconfidence more effectively than does augmented risk aversion.
AB - Perhaps the two most prominent human cognitive "biases" are unjustified self-belief, termed overconfidence, and loss aversion, a special form of reference dependence whereby an outcome is less attractive if perceived as a loss. As evolution tends to make adaptations that enhance fitness, the puzzle is why these biases arise, especially as they tend to offset each other. Specifically, overconfidence promotes risk-taking, such as starting a business, whilst loss aversion curbs it. Is evolution just burying its own mistakes? We propose an alternative reconciliation of the evolutionary basis of these seemingly conflicting biases. Following Robert Trivers, overconfidence arises to positively influence the behavior of others. The downside is that if acted on, overconfidence leads to mistaken decisions. Our point is that loss aversion limits how far these evaluation errors are implemented. Importantly, loss aversion corrects decision errors due to overconfidence more effectively than does augmented risk aversion.
KW - Overconfidence; Loss Aversion; Expected Utility Theory; Risk Aversion; Gender.
U2 - 10.31234/osf.io/gsdcv
DO - 10.31234/osf.io/gsdcv
M3 - Working paper
BT - Talking the talk, not walking the walk
PB - PsyArXiv Preprints
ER -