Code comprehension is a cognitively demanding task, becoming increasingly important as developers spend more time reviewing AI-generated code. Prior research has examined factors influencing comprehension, including source code characteristics, developer traits, practices such as refactoring, and different strategies developers use to comprehend code. Despite this work, there is no clear consensus on what effectively supports comprehension. Refactoring standards still rely largely on static code metrics, many best practices lack empirical validation, and findings are often contradictory. More fundamentally, given the cognitive nature of code comprehension, the cognitive reasons why developers adopt different comprehension strategies remain largely unexplored. In this study, we examine how individual cognitive differences impact code comprehension by looking at how refactoring affects developers’ cognitive processes. We performed a within-subjects eye-tracking experiment where student developers view four Java tasks, either a refactored or non-refactored version, allowing us to compare how code refactoring affects their cognitive behaviour. First, we use the GenderMag cognitive framework to characterize developers’ cognitive styles. Then, during code comprehension tasks, we use eye-tracking to observe their cognitive behavioural patterns and analyze whether these patterns differ or align across cognitive styles. Our preliminary results with four pilot participants showed clear variations in their code-tracing strategies across different cognitive styles. Addressing this gap in understanding developer cognitive behaviour in code comprehension provides insights into the needs of developers through empirical investigation. This research calls for a fundamental shift from generic best practices toward developer-centric best practices—ones that are aware and recognize individual differences across the software engineering workflow.