Opening a Window into the Visual and Conceptual Components of Visual Memory

Our visual memories are the tickets we use to re-experience and re-imagine our lives. But what is the true content of these memories? Are they a rich reinstatement of the visual aspects of the experience, with every item, color, and design from the original experience present? Or are they more conceptually rooted, informed by our knowledge of what an experience typically is, rather than the unique details from the original experience? My dissertation explores the visual and conceptual components of visual memory content using drawing, brain patterns, computational methods, and aphantasics without visual imagery as a “knockout” model of visual detail in memory. Across three chapters and six experiments, I find that (1) conceptual information is the first information to arrive in memory and the last to leave, (2) aphantasics have significantly less visual detail in memory than those with normal imagery even on the neural level, and that this detail may arrive in memory from frontoparietal and occipitotemporal connections in the brain, and (3) that the content of visual memory, such as the amount of visual detail, is changed by conceptual information. Taken together, these results reveal that the content of visual memory can vary wildly: ranging from largely conceptual with little detail unique to the original experience, to a vivid visual recollection of the experience with swaths of pixel-perfect memory. However, I find evidence that conceptual information may be more dominant in visual memory, remaining present across time, imagery abilities, and that it has the power to diminish visual detail and enhance the efficiency of memory recall.