Identifying Neurodevelopmental Disorders through Social Media Algorithms

Abstract

In the context of the evolving digital landscape, this research paper explores the operations of social media algorithms and their potential to identify individuals with neurodevelopmental disorders. Neurodiversity is a significant aspect of contemporary mental health discussions in the digital age. The primary objective is to unravel the mechanisms of social media algorithms and assess their capacity to identify users diagnosed with neurodevelopmental disorders. Employing a dual-method approach, both quantitative and qualitative methodologies were utilized. A quantitative survey targeted general social media users, collecting 143 valid responses to gauge their interactions with algorithm-generated content. Simultaneously, a qualitative survey involved interviews with artificial intelligence specialists, providing expert insights into algorithmic functionality. The analysis revealed that social media algorithms operate on recommender systems, categorizing content based on users' historical preferences. However, these algorithms lack the inherent capability to identify neurodevelopmental disorders. Instead, user-interacted content influences subsequent algorithmic recommendations.