Information acquisition during online decision-making

A model-based exploration using eye-tracking data

W. Shi, M. Wedel, R. Pieters

Research output: Contribution to journalArticleScientificpeer-review

Abstract

We propose a model of eye-tracking data to understand information acquisition patterns on attribute-by-product matrices, which are common in online choice environments such as comparison websites. The objective is to investigate how consumers gather product and attribute information from moment to moment. We propose a hierarchical hidden Markov model that consists of three connected layers: a lower layer that describes the eye movements, a middle layer that identifies information acquisition processes, and an upper layer that captures strategy switching. The proposed model accounts for the data better than several alternative models. The results show that consumers switch frequently between acquisition strategies, and they obtain information on only two or three attributes or products in a particular acquisition strategy before switching. Horizontal and contiguous eye movements play an important role in information acquisition. Furthermore, our results shed new light on the phenomenon of gaze cascades during choice. We discuss the implications for Web design, online retailing, and new directions for research on online choice.
Original languageEnglish
Pages (from-to)1009-1026
JournalManagement Science
Volume59
Issue number5
DOIs
Publication statusPublished - 2013

Fingerprint

Information acquisition
Decision making
Eye movements
Online retailing
Alternative models
Consumer products
Web design
Hidden Markov model
Web sites
Cascade

Cite this

@article{b756af931a6f4c8b9b3ac7d15ea9c607,
title = "Information acquisition during online decision-making: A model-based exploration using eye-tracking data",
abstract = "We propose a model of eye-tracking data to understand information acquisition patterns on attribute-by-product matrices, which are common in online choice environments such as comparison websites. The objective is to investigate how consumers gather product and attribute information from moment to moment. We propose a hierarchical hidden Markov model that consists of three connected layers: a lower layer that describes the eye movements, a middle layer that identifies information acquisition processes, and an upper layer that captures strategy switching. The proposed model accounts for the data better than several alternative models. The results show that consumers switch frequently between acquisition strategies, and they obtain information on only two or three attributes or products in a particular acquisition strategy before switching. Horizontal and contiguous eye movements play an important role in information acquisition. Furthermore, our results shed new light on the phenomenon of gaze cascades during choice. We discuss the implications for Web design, online retailing, and new directions for research on online choice.",
author = "W. Shi and M. Wedel and R. Pieters",
year = "2013",
doi = "10.1287/mnsc.1120.1625",
language = "English",
volume = "59",
pages = "1009--1026",
journal = "Management Science",
issn = "0025-1909",
publisher = "INFORMS Inst.for Operations Res.and the Management Sciences",
number = "5",

}

Information acquisition during online decision-making : A model-based exploration using eye-tracking data. / Shi, W.; Wedel, M.; Pieters, R.

In: Management Science, Vol. 59, No. 5, 2013, p. 1009-1026.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Information acquisition during online decision-making

T2 - A model-based exploration using eye-tracking data

AU - Shi, W.

AU - Wedel, M.

AU - Pieters, R.

PY - 2013

Y1 - 2013

N2 - We propose a model of eye-tracking data to understand information acquisition patterns on attribute-by-product matrices, which are common in online choice environments such as comparison websites. The objective is to investigate how consumers gather product and attribute information from moment to moment. We propose a hierarchical hidden Markov model that consists of three connected layers: a lower layer that describes the eye movements, a middle layer that identifies information acquisition processes, and an upper layer that captures strategy switching. The proposed model accounts for the data better than several alternative models. The results show that consumers switch frequently between acquisition strategies, and they obtain information on only two or three attributes or products in a particular acquisition strategy before switching. Horizontal and contiguous eye movements play an important role in information acquisition. Furthermore, our results shed new light on the phenomenon of gaze cascades during choice. We discuss the implications for Web design, online retailing, and new directions for research on online choice.

AB - We propose a model of eye-tracking data to understand information acquisition patterns on attribute-by-product matrices, which are common in online choice environments such as comparison websites. The objective is to investigate how consumers gather product and attribute information from moment to moment. We propose a hierarchical hidden Markov model that consists of three connected layers: a lower layer that describes the eye movements, a middle layer that identifies information acquisition processes, and an upper layer that captures strategy switching. The proposed model accounts for the data better than several alternative models. The results show that consumers switch frequently between acquisition strategies, and they obtain information on only two or three attributes or products in a particular acquisition strategy before switching. Horizontal and contiguous eye movements play an important role in information acquisition. Furthermore, our results shed new light on the phenomenon of gaze cascades during choice. We discuss the implications for Web design, online retailing, and new directions for research on online choice.

U2 - 10.1287/mnsc.1120.1625

DO - 10.1287/mnsc.1120.1625

M3 - Article

VL - 59

SP - 1009

EP - 1026

JO - Management Science

JF - Management Science

SN - 0025-1909

IS - 5

ER -