In the field of natural language processing (NLP), accurately understanding and processing complex linguistic structures remains a major challenge. This paper addresses the less-explored phenomenon of cataphora-where a pronoun or noun phrase points forward to a yet-tobe-mentioned entity in the discourse. While anaphora resolution has been extensively studied, cataphora detection and resolution have not received the same level of attention and remain underexplored. This paper seeks to bridge this gap by evaluating state-of-the-art techniques and identifying the obstacles that hinder effective cataphora resolution. We investigate the role of syntactic and semantic ambiguities, contextual influences, and the integration of world knowledge. Additionally, the potential of deep learning, neural network and hybrid models to advance cataphora resolution is explored.