Using neuroimaging (fNIRS) and choice experiments to explore female preferences for male attributes

Functional near-infrared spectroscopy (fNIRS) is increasingly applied to study cortical processes involved in decision-making due to its portability and non-invasive nature. However, the extent to which fNIRS-derived neural signals correspond to quantitative preference parameters obtained from discrete choice experiments remains insufficiently examined.

This study introduces an integrated framework combining choice-based conjoint analysis with fNIRS neuroimaging. Sixty female participants aged 20-30 viewed male faces and upper bodies while prefrontal cortex activity was recorded using fNIRS. Participants also completed a discrete choice experiment, from which individual-level part-worth utilities (PWUs) for male attributes were estimated using hierarchical Bayes modeling. fNIRS data were preprocessed using standard pipelines and analyzed via statistical parametric mapping. Stepwise regression was applied to relate PWUs to channel-wise cortical activation.

Preferred visual stimuli elicited significantly higher oxygenated hemoglobin responses in prefrontal regions associated with attention and decision-making. Regression analyses revealed significant associations between PWUs and activation in specific fNIRS channels, indicating that neural responses explained a meaningful proportion of variance in stated preferences. Psychographic variables, including social desirability and risk aversion, further modulated behavioral and neural outcomes.

Compared to traditional conjoint analysis or self-report measures alone, the proposed approach directly links quantitative preference estimates to neurophysiological markers. Unlike prior fNIRS studies without formal preference modeling, this method enables systematic integration of neural data with choice-based parameters.

The findings demonstrate the feasibility of combining fNIRS with discrete choice experiments to study preference formation and validate behavioral choice models using neurophysiological data.