Using neural data to forecast aggregate consumer behavior in neuromarketing: Theory, metrics, progress, and outlook

• Published in: Journal of consumer behaviour Vol. 23; no. 4; pp. 2142 - 2159
• Main Authors: Yao, Xiaoqiang, Wang, Yiwen
• Format: Journal Article
• Language: English
• Published: London Wiley Subscription Services, Inc 01.07.2024
• Subjects: Advertising campaigns; Cognition; Consumer behavior; Forecasting; Machine learning; Marketing; Motivation; Psychological processes
• ISSN: 1472-0817; 1479-1838
• DOI: 10.1002/cb.2324

The field of using neural data to forecast aggregate consumer choice has garnered attention in the past decade, holding substantial promise for both researchers and practitioners. However, a comprehensive understanding of this emerging field is lacking. This paper aims to bridge that gap by summarizing existing research, encompassing relevant theories, metrics, progress, and future directions. We begin by introducing the concept of neuroforecasting within the field of neuromarketing. We then delve into theories that leverage neural data for forecasting aggregate choice, including affect‐integration‐motivation framework, frontal asymmetry, and inter‐subject correlation. Subsequently, we review various metrics, including self‐reported, behavioral, and neural metrics employed to forecast market‐level behavior, presenting key findings from relevant studies. Furthermore, we examine the strengths and weaknesses of this field. Advantages of this approach include its ability to offer effective predictions of consumer behavior and provide enhanced insights into consumer preferences and choices, while its weaknesses encompass relatively high cost, sample size constraints, issues of ecological validity, and challenges related to reverse inference. In conclusion, future research should prioritize integrating diverse data types with machine learning techniques to forecast the outcomes of marketing campaigns in advance. Additionally, a deeper exploration of the psychological and cognitive processes underlying successful predictions can augment predictive accuracy and effectiveness. This review provides a systematic overview for researchers and practitioners in this field, offering valuable insights and guidance for future research endeavors and industry applications.