Learning and adaptation under uncertainty and ambiguity

The first essay is about how high and moderate aspiration levels compare in terms of affecting the decision making and reinforcement learning in an uncertain environment. After developing a thought experiment and a computational model, I used lab experiments to test the model’s predictions: a high (...

Full description

Saved in:
Bibliographic Details
Main Author: ZHENG, Lei
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2020
Subjects:
Online Access:https://ink.library.smu.edu.sg/etd_coll/302
https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=1302&context=etd_coll
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
Description
Summary:The first essay is about how high and moderate aspiration levels compare in terms of affecting the decision making and reinforcement learning in an uncertain environment. After developing a thought experiment and a computational model, I used lab experiments to test the model’s predictions: a high (moderate) aspiration level reduces (increases) feedback ambiguity about the relative attractiveness of different options, thus increases the exploitation (exploration) tendency of the decision maker. The behavioural difference suggests that high aspirations lead to better performance in stable environments, but worse performance after disruptive shocks. The second essay investigates whether organizations should commit more (or less) to exploration in response to an increased environmental dynamism. Using a computational model, I address the literature contradictions by disentangle exploration intensity and width. I demonstrate that the phenomenon of “chasing a moving target” (Posen & Levinthal, 2012) – the decreasing optimal exploration level under increased environmental dynamism – is caused by the entanglement of exploration intensity and width. The third essay addresses the question about how ambiguous performance feedback across organizational levels affects resource allocation. Attribution theory suggests organizations and organizational members will attribute success internally while attributing failures externally, resulting different learning and response patterns following organizational success and failure. Using professional basketball data, I demonstrate the resources (minutes) allocated to players are subject to the players prior performance. Team performance (game win) positively moderates the relationship between allocated resource and a player’s performance. The moderating effect is the weakest when the team experience a loss with large point-deficit.