This paper reports on an exploratory study designed to quantitatively investigate the effects of three code reading methods. We conducted an experiment with 22 university students with programming experience, who were tasked with finding and fixing bugs in C language source code using three different reading methods: "reading aloud," "subvocalization," and "silent reading." The experiment was designed as a within-subjects study, where each participant experienced all three conditions. An analysis of the experimental data using paired t-tests and other methods yielded the following findings. Overall, subvocalization showed statistically significant higher debugging performance compared to reading aloud (p < 0.05). However, this effect was not uniform, varying with participants' programming proficiency and their conscious ability to distinguish between the reading methods. Notably, among the group of participants who could clearly differentiate between silent reading and subvocalization, subvocalization demonstrated a statistically significant higher performance than silent reading as well (p < 0.05). These preliminary findings suggest that consciously controlled subvocalization may be an effective strategy for managing cognitive load and directing attention to the details of the code during debugging tasks. This paper reports these findings in detail and discusses their potential applications in programming education and practice. [For the complete proceedings, "Proceedings of the International Association for Development of the Information Society (IADIS) International Conference on Cognition and Exploratory Learning in the Digital Age (CELDA) (22nd, Porto, Portugal, November 1-3, 2025)," see ED677812.]