Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


ChenThesis.pdf (1.32 MB)

ETD Abstract Container

Abstract Header

Essays on Learning, Decision-making and Attention

Chen, Wei
http://orcid.org/0000-0001-6360-9348
http://rave.ohiolink.edu/etdc/view?acc_num=osu1491925104416652

Abstract Details

2017, Doctor of Philosophy, Ohio State University, Economics.
My dissertation consists of three chapters. In chapter 1, I study epiphany learning (EL). EL is characterized by an unexpected moment of insight, which contrasts with the gradual learning process in the traditional reinforcement learning (RL). I develop a computational model of EL in a behavioral experiment where subjects learn the optimal strategy in the 2-person beauty contest game. I show that subjects' learning curve can be better fitted by my EL model than the RL model. In chapter 2, I investigate further into EL and test my model with an eye-tracking experiment again using the 2-person beauty contest game. In this experiment, I add the feature that subjects could commit to a strategy whenever they want, namely to lock into that strategy for the rest of the experiment. The idea is that subjects should only commit to a strategy if they have an epiphany. I find that subjects who committed to the optimal strategy --just before they committed-- had a very different gaze dwell time distribution, saccade pattern and pupil diameter, compared to those who committed to other strategies or never committed. Moreover, these subjects' gaze data were aligned with a key latent process in my EL model. In the last chapter, I study how pupil dilation is modulating the simple choice decision-making process in the context of the attentional drift-diffusion model (aDDM). In an incentivized binary food-choice experiment, I find that the size of the pupil diameter is negatively correlated with the response time and accuracy (i.e., the probability of choosing the food item that the subject previously claimed to have a higher value). The latter model fitting exercise show that this was largely because that pupil dilation was modulating the attentional bias parameter and the noise in the evidence accumulation process in the aDDM.
Ian Krajbich (Advisor)
James Peck (Committee Member)
Paul Healy (Committee Member)
98 p.
Economics
epiphany learning; decision-making; eye-tracking; neuroeconomics; pupil-dilation; drift-diffusion model

Recommended Citations

Citations

  • Chen, W. (2017). Essays on Learning, Decision-making and Attention [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1491925104416652

    APA Style (7th edition)

  • Chen, Wei. Essays on Learning, Decision-making and Attention. 2017. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1491925104416652.

    MLA Style (8th edition)

  • Chen, Wei. "Essays on Learning, Decision-making and Attention." Doctoral dissertation, Ohio State University, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=osu1491925104416652

    Chicago Manual of Style (17th edition)

osu1491925104416652
208
© 2017, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.