Date of Award

Spring 2022

Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Keil, Frank


Mental representations are the essence of cognition. Yet, to understand how the mind works, we must understand not just the content of mental representations (i.e., what information is stored), but also the format of those representations (i.e., how that information is stored). If we want to understand how sensory information is translated into symbolic representations, if we want to know how the mind forms ‘cognitive maps’, if we want to know how the firing of neurons can lead to the emergent phenomenon of human cognition — all of these things require us to understand how information is organized in the mind. In this thesis, I describe three ‘case studies’ of representational format in the domain of spatial cognition. I focus on spatial cognition for several reasons. First, spatial cognition is ubiquitous in the animal kingdom; thus, understanding spatial cognition in the human mind has the potential to reveal insights that generalize to all minds. Second, spatial cognition may be the single domain for which we know the most about the format of representations; indeed, the field was essentially founded on the premise that there exists a discernable ‘cognitive map’ within the mind. As such, it serves as an apt domain to study representational format. Finally, spatial representations (location representations in particular) may serve as the format of other higher-level information (e.g., numerical information, social information, etc.). Understanding the formats of spatial representation, therefore, may shed light on how other kinds of information are represented and organized in the mind. The first case study I describe pertains to the format of location representations. I show that, using a simple ‘error correlation’ analysis, we can uncover from simple spatial tasks the coordinate systems underlying spatial behavior. Using this approach, I argue that locations are spontaneously represented in polar coordinates, but flexibly in other coordinate systems (e.g., Cartesian coordinates) as needed. The second case study I describe pertains to the format of size representations. It has been known for many decades that the perception of size is illusory; for example, larger objects are perceived as being relatively less large. However, these illusions are typically explained by vague, unfalsifiable theories of size perception. I offer a simpler (and falsifiable) explanation of size illusions: that perceived size is equal to the sum of an objects’ dimensions rather than the product. Here, I focus primarily on the perception of area in adults, but this phenomenon appears to be highly general: I briefly allude to similar illusions that children experience, as well as similar illusions of volume. The final case study I describe pertains to how spatial information is used as a format to represent other information. I show that task-irrelevant ‘spatial structure’ spontaneously improves working memory. This effect is specific to spatial information; color information and audio information produce no such benefit. I discuss how these findings relate to existing models of working memory, and help us to understand the relationship between space and memory more broadly. I conclude with some final remarks about how understanding spatial behavior in light of the formats of representations can help us to understand the building blocks of cognition.