Echo chambers and filter bubbles remain deeply contested constructs across academic, industrial, and public discourse. This study conducts a systematic review of 391 peer-reviewed scholarly articles and 131 Google News articles to investigate how disciplinary field, data platform, methodological design, and institutional affiliation shape empirical claims regarding the existence and causes of these phenomena. Using weighted statistical analysis and structured content coding, we examine how epistemic knowledge is constructed, interpreted, and disseminated. Findings indicate that conclusions about the existence of echo chambers and filter bubbles are significantly influenced by disciplinary orientation, platform context, and methodological approach. In contrast, causal attributions, whether ascribed to user agency, algorithmic design, or their interplay, are associated with methodological choices, platform framing, and institutional positioning, but not with disciplinary background. Media narratives diverge sharply from academic accounts, consistently affirming the existence of echo chambers and filter bubbles while disproportionately attributing their formation to algorithmic systems. We argue that disciplinary silos, methodological specialization, and asymmetrical access to proprietary data have produced fragmented causal imaginaries, posing challenges for policy coherence and public accountability. These findings underscore the need for integrative analytical frameworks and improved science communication strategies to reconcile divergent understandings across research, industry, and public spheres.