AUTHOR=Marques dos Santos José Paulo , Marques dos Santos José Diogo 

TITLE=Explainable artificial intelligence (xAI) in neuromarketing/consumer neuroscience: an fMRI study on brand perception

JOURNAL=Frontiers in Human Neuroscience

VOLUME=Volume 18 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2024.1305164

DOI=10.3389/fnhum.2024.1305164

ISSN=1662-5161

ABSTRACT=The research in consumer neuroscience has identified computational methods, particularly artificial intelligence (AI) and machine learning, as a significant frontier for advancement. Previous endeavors, dating back a decade, utilized functional magnetic resonance imaging (fMRI) and artificial neural networks (ANNs) to model brain processes related to brand preferences. In the current study, we revisit this data, leveraging recent advancements in explainable artificial intelligence (xAI) to gain valuable insights in this domain. In this way, through the integration of fMRI data analysis, machine learning, and xAI, our study aims to provide answers to the fundamental question of "what" associations traverse consumers' minds during brand exposure.The application of AI to fMRI analysis has an intrinsic hurdle related to the data's high dimensionality. To overcome this, we applied ICA and extracted pertinent features from the returned ICs. An ANN is then trained on this data, followed by pruning and retraining processes. We then apply explanation techniques, based on path-weights and Shapley values, to obtain insights into the underlying brain processes.The fully connected ANN model obtained an accuracy of 54.6%, which dropped to 50.4% after pruning. However, the retraining process allowed it to, not only recover the lost accuracy, but to surpass the fully connected network, achieving an accuracy of 55.9%. The path-weights and Shapleybased analysis concludes that, regarding brand perception, the expected initial participation of the primary visual system is followed. Then, the processed stimulus arrives at other brain areas for further processing, sometimes integrating such processing for a complete stimulus description.