SHIFTING THE COGNITIVE BURDEN: AI-MEDIATED REDUCTION OF COGNITIVE LOAD IN FUNCTIONAL WRITTEN DISCOURSE

Abstract

In the linguistically demanding landscape of Pakistani higher education, the mastery of English-medium functional discourse often imposes a prohibitive mental burden on students. Applying the theoretical framework of Sweller's Cognitive Load Theory (CLT), this study utilises a quantitative, cross-sectional approach (N=385) to evaluate the role of generative AI as a cognitive intermediary in academic and professional composition. The research investigates the impact of AI tools on the "functional written discourse" of undergraduate and postgraduate students, specifically focusing on the optimisation of cognitive resources. The findings reveal a transformative reduction across three primary cognitive dimensions. First, addressing the functional utility and linguistic mechanics of writing, participants reported the highest reduction in Extraneous Load (Mean=4.41), with over 90% of students agreeing that AI effectively manages complex grammatical and structural mechanics. Second, the data indicates a significant reduction in perceived Intrinsic Load (Mean=4.26), facilitating a shift toward Germane Load, with 89% of participants reporting that complex writing tasks are substantially less exhausting when AI-supported. Third, the study identifies a substantial boost in Pragmatic Confidence (87% positive response rate), as AI ensures the formal tone essential to Pakistani academic conventions. Qualitative insights from regional hubs highlight a fundamental shift in student identity from "language-struggler" to "content-creator." The study concludes that AI acts as a vital linguistic equaliser by reallocating mental resources from surface-level mechanics to higher-order logical synthesis. It recommends that the Higher Education Commission (HEC) should adopt pedagogical models that integrate AI literacy to enhance student self-efficacy and professional readiness.