Eye tracking as a powerful tool for exploring decision-making processes

Associate Professor Li-Sheng He is a psychologist specializing in behavioral economics and decision neuroscience, who leads the Behavioral Science Laboratory at the Shanghai University SILC Business School. Decision-making involves combining external information with internal mental schemas to formulate judgments.  

Visual information is our primary source of information during the decision-making process, and its efficiency is governed by attention allocation mechanisms. Eye tracking technology enables researchers to precisely quantify these attention patterns and visually map how individuals collect different types of information. Ultimately, it allows researchers to decode key mechanisms underlying human decision-making.  

Theoretical conflict in research on partner selection  

When selecting a partner, individuals evaluate multiple attributes such as physical attractiveness, financial resources, social status, and intelligence. But how do people process these diverse cues to reach a final decision? Existing research presents two theoretical models: non-compensatory heuristic strategy and compensatory strategies. 

Non-compensatory heuristic strategy suggests that people sequentially evaluate the value of different cues in potential partners, proceeding to evaluate further only when a cue's value surpasses the expected threshold. In contrast, compensatory strategies involve weighing the pros and cons of various options, where individuals make choices through comprehensive consideration. Additionally, it suggests that advantages in certain aspects could offset disadvantages in others. 

Dr. He noted that these two theories appear contradictory, and neither can fully explain the process of choosing a partner. He therefore decided to use eye tracking technology to dive deeper into the cognitive mechanisms underlying the choice between potential partners. 

Eye tracking decodes partner choice decisions  

To investigate the cognitive mechanisms explaining a person’s choice of partner, a total of 107 university students were recruited to participate in Dr. He’s study. Each participant was shown the facial image and income information of a potential partner and being asked to decide between these two attributes. While deciding, a Tobii screen-based eye tracker was used to collect eye movement data from the participants. The study specifically examines gender-based differences in visual attention allocation between income and facial attractiveness, while also analyzing sequential observation patterns during the decision-making process. 

The results indicated that male participants spent 75.7% of their total dwell time on facial areas of interest (AOIs), compared to 69.7% for females, with a significant difference suggesting males may prioritize facial information more than females. Regression models revealed that highly attractive faces led participants to pay more attention to income-related details. Higher income levels instead led to increased fixation duration on facial features. When accounting for gender differences, time-series analysis demonstrated that male participants turned from viewing faces to income cues. Females instead showed the reverse pattern, shifting from income to facial information. 

Eye tracking in idea-testing research 

In this study, Dr. He and colleagues confronted two conflicting theoretical accounts, turning to eye tracking technology to seek novel evidence. Traditional decision-making research has relied on participants' behavioral responses to suggest underlying cognitive processes. As a type of spatiotemporal signal, eye movement data can characterize the information collection process, thereby capturing the spatial and temporal details throughout the decision-making process. 

Building on conflicting prior theories and eye tracking experimental evidence, their research utilized eye tracking technology to provide evidence at the information-search level in support of non-compensatory strategy: rational decision-makers do not need to assess the value of all options. Instead, they should first identify the highest-value option, prioritize information collection from it, and then systematically explore lower-value alternatives. This decision-making model enables individuals to allocate their limited attentional resources more efficiently. 

References 

Wang, H., Ma, J., & He, L. (2022).
Value-directed information search in partner choice. Judgment and Decision Making 17(6), 1287–1312. 

* Images courtesy of Shanghai University SILC Business School