Cross-ethnicity/race generalization failure of behavioral prediction from resting-state functional connectivity

Algorithmic biases that favor majority populations pose a key challenge to the application of machine learning for precision medicine. Here, we assessed such bias in prediction models of behavioral phenotypes from brain functional magnetic resonance imaging. We examined the prediction bias using two independent datasets (pre-adolescent versus adult) of mixed ethnic/racial composition. When predictive models were trained on data dominated by white Americans (WA), out-of-sample prediction errors were generally higher for African Americans (AA) than for WA. This bias toward WA corresponds to more WA-like brain-behavior association patterns learned by the models. When models were trained on AA only, compared to training only on WA or an equal number of AA and WA participants, AA prediction accuracy improved but stayed below that for WA. Overall, the results point to the need for caution and further research regarding the application of current brain-behavior prediction models in minority populations.

Automated summary

Algorithmic biases favoring majority populations pose a significant challenge to the use of machine learning for precision medicine. In this study, the researchers assessed such bias in prediction models of behavioral phenotypes using brain functional magnetic resonance imaging. The results showed a bias towards White Americans in the predictive models, resulting in higher prediction errors for African Americans. However, training the models on African Americans improved prediction accuracy but still fell below that for White Americans. Overall, this study highlights the need for caution and further research when applying current brain-behavior prediction models in minority populations.