<p>Given the importance of memory to our daily functioning, humans have made use of many strategies to improve their encoding and later retrieval. Recently, drawing has emerged as a particularly effective strategy that robustly improves memory compared with traditional encoding strategies, and in the current study, we investigate the cognitive features that underlie this effect. Participants (<i>N</i> = 60) engaged in four distinct encoding tasks—Drawing, No Ink Drawing, Tracing, and Visualizing—each designed to engage one or more of the multisensory ‘components’ thought to underlie the impact of drawing on memory. We evaluated memory performance using both a classic old/new recognition task and source memory judgment, providing insights into both general recognition performance, and memory for specific encoding contexts. Consistent with prior work, we found Drawing led to the best memory performance in both recognition and source memory. Notably however, the No Ink Drawing condition, which lacked the visual component offered by Drawing, produced the second best old/new recognition performance, but the worst source memory. Unpacking this result, we found participants were frequently confusing No Ink Drawing with Drawing during source memory judgments, perhaps reflecting a hierarchy to the components that enhance memory, which seems to favor motoric engagement and elaboration over visual information. Together, our results extend our current understanding of the cognitive processes underlying successful encoding and provide new insights into how various memory tests can provide information about the specific information encoded alongside memories.</p>