Modern code reviews, typically associated with pull requests in version control systems such as Git, are a common practice in the software industry. However, individual reviewers may encounter challenges such as code comprehension difficulties, stress, fatigue, or distraction that can negatively impact the quality of their review task, particularly in what concerns the detection of possible bugs in the code under review. In this paper, we present a new technique for evaluating the quality of code reviews by assessing the reviewer’s cognitive load. Our technique utilizes biometric information collected from non-intrusive sensors and wearable measures, such as Heart Rate Variability (HRV) and task-evoked pupillary response, to estimate the cognitive load and code comprehension level of the reviewer. We also use an inexpensive desktop eye-tracker to identify code regions (i.e., snippets of code under review) that are not well-reviewed. To predict the quality of code review, we employ explainable machine learning models that consider both biometric and non-biometric features such as the reviewer’s expertise, code complexity, number of code region revisits, and review time. Our experimental results show that our approach can accurately predict review quality, with a prediction accuracy of 87.77% and a Spearman correlation coefficient of 0.85, indicating a strong correlation between the predicted and actual review performance. This evaluation also demonstrates the accuracy of our cognitive load measurement using EEG signals as a reference point for the HRV and pupil signals. The proposed technique has the potential to improve the quality of code review in the software industry by automatically identifying code regions that were not thoroughly reviewed and enabling reviewers to take corrective actions.