Netlist Reverse Engineering Practice: Effects of Prior Knowledge, Cognitive Abilities, and Personality Traits

Abstract

Netlist Reverse Engineering (NRE) is a critical human-computer interaction task for identifying security threats at the microchip level. Yet, little is known about how individual factors shape NRE performance and practice effects. We conducted a mixed experimental and quasi-experimental study with N = 155 participants, randomly assigned to a practice or a no-practice control condition, with practice difficulty adapted to baseline performance. After a baseline assessment, practice groups completed a 10-minute NRE practice phase, while the control group engaged in an unrelated task. Performance was reassessed following a one- to two-night consolidation phase. Multilevel modeling revealed a significant practice effect on both speed and accuracy; adjusting for regression toward the mean, we found no evidence that its size depends on baseline performance. Self-rated prior knowledge predicted baseline performance but not learning gains. Our findings are consistent with perceptual learning as one plausible mechanism, and suggest that benefiting from practice does not require prior NRE experience.