Silvia Shiwei Zhou¹ (), Hala Rahman¹, Keanna Rowchan¹, Tasha Ignatius¹, Jeffrey D. Wammes¹; ¹Queen's University

One’s ability to generate mental imagery can influence a broad spectrum of cognitive processes. However, imagery is often measured as a single item or construct using trait-level questionnaires despite the multifaceted nature of imagery both within and across individuals. This makes it difficult to capture whether not only the ability to generate imagery but also the particular properties of that imagery (e.g., sharpness, colorfulness, prototypicality) contribute to learning and memory. Multidimensional experience sampling (mDES) offers a promising method for capturing the nuanced qualities of ongoing thoughts (e.g., visual vs. verbal, detailedness). Inspired by mDES, we developed a novel set of 15 imagery-mDES items that ask participants to rate the experiential properties of their mental images on continuous scales. These items assess perceptual (e.g., colorfulness, sharpness), subjective (e.g., difficulty, immersive), and dynamic properties (e.g., broader context) of imagery. Participants (N = 82) encoded by viewing an object, generating a mental image for 4–8 seconds, making a judgment of learning (JOL), and then completing the imagery-mDES items. Visual memory was later tested using a difficult 4-alternative forced-choice remember/know/guess task (RKG). We identified three dominant principal components that describe the dimensions along which people’s imagery varied: “vivid detailed,” “dynamic contextual,” and “difficult prototypical.” These patterns were meaningfully related to memory: remembered objects were associated with more “vivid detailed” imagery, and less “difficult prototypical” imagery. Among correct trials, “vivid detailed” imagery was associated with a dramatic RKG gradient, whereas “dynamic contextual” imagery led to fewer guesses. Participants’ JOLs paralleled these effects, suggesting metacognitive sensitivity to imagery properties. By characterizing imagery at the item level, our approach helps quantify how different imagery themes contribute to encoding and retention. This multidimensional approach opens new possibilities for understanding how visual cognition shapes learning, and revealing individual differences in how imagery is deployed.