AI-Driven Mitigation of Cognitive Biases in Intelligent Personal Assistant Interactions: Evidence from African Contexts

This paper presents a rigorous investigation into how artificial intelligence-driven features in Intelligent Personal Assistants (IPAs) can mitigate cognitive biases within the culturally diverse landscapes of African societies. Positioned at the intersection of cognitive psychology, artificial intelligence, and African cultural studies, the research examines how traditional decision-making patterns in West, Southern, and Central African contexts interact with AI-powered debiasing mechanisms. Grounded in Dual-Process Theory and indigenous knowledge systems, the study explores how IPAs can be culturally calibrated to address confirmation bias, anchoring, and availability heuristics as they uniquely manifest within African socio-cultural frameworks. Employing a sequential explanatory mixed-methods design, the study integrates survey data from 528 participants across eight countries with 40 in-depth interviews. The findings reveal that while AI-driven interventions significantly reduce cognitive biases, their effectiveness is deeply moderated by cultural dimensions such as power distance, uncertainty avoidance, and collectivist orientations—each varying distinctly across regions. Culturally contextualized nudges and interventions aligned with local values and communication norms yielded the strongest debiasing outcomes. This research offers essential empirical insights into the emerging field of culturally responsive AI design, emphasizing the need to recalibrate debiasing techniques to reflect and respect African cultural perspectives rather than applying Western-centric models of cognitive optimization.