The advent of the COVID-19 pandemic accelerated the need to conduct synchronous online instruction in which students and faculty were distributed geographically but aggregated digitally onto a single screen on a computer. This approach will likely persist into the future. Thus, a better understanding of how engagement approaches can be adapted to the synchronous online teaching environment is critical for future success. This research presents an experiment-based investigation into the effectiveness of two often-debated engagement approaches, namely, the camera-on and the vote/chat approaches in a synchronous online learning environment. We found that using individual engagement methods, such as requiring cameras to be on or using the vote and chat functions to elicit student responses, improves learning outcomes. Surprisingly, the combined use of these methods hinders student learning, and they perform no better than the control group where no specific engagement methods were required. We posit that the reason for this result is that too much engagement increases total cognitive load to a point where it is detrimental to learning.