<![CDATA[This study aims to describe the mathematical problem-solving abilities of senior high school students with a visual learning style (GB) in solving Minimum Competency Assessment (AKM) numeracy tasks on quadratic functions, categorized by cognitive performance groups. This descriptive qualitative study involved research subjects selected through a combination of a learning style questionnaire and a cognitive baseline test (TG). The subjects were stratified into three distinct cognitive tiers based on Bloom's taxonomy criteria: a low-cognitive visual subject (mastering levels C1-C2), a medium-cognitive visual subject (mastering level C3), and a high-cognitive visual subject (mastering levels C4-C6). Data collection instruments comprised a customized AKM Numeracy test containing one contextual essay question and one complex multiple-choice item (true/false format requiring written logical validation), followed by task-based semi-structured interviews. The results demonstrated that the low-cognitive visual subject successfully performed initial stages of problem understanding by extracting text into clear geometric properties, yet failed to construct algebraic modeling or procedural executions. The medium-cognitive visual subject showed strong performance in identifying explicit data arrays and accurately executing mathematical algorithms, but struggled when translating complex non-linear spatial graphs into multi-variable expressions. Conversely, the high-cognitive visual subject exhibited a comprehensive and integrated problem-solving capacity, navigating fluidly across descriptive schemas, algebraic manipulation, optimization checks, and formal abstract justifications]]>
