We have two different reference systems for spatial memory

We have two different reference systems to use to understand spaces: one that tracks objects relative to our own bodies in space (egocentric) and one that tracks objects relative to other objects in space (allocentric). This was controversial for some time, but has been widely accepted since the 2000s.

Roughly, we can say in experiences where we’re tracking objects relative to ourselves, we understand that experience as a tabletop. In experience where we are tracking objects relative to other objects, we understand that experience as a landscape. Reference systems are intrinsic to an experience and in the physical world, do not change.

In the digital world, it’s very easy for intrinsic cues, extrinsic cues, and egocentric cues¹ ⊕ ¹ I have a note that experiences that can be assigned an easy “top” are likelier to be understood as a tabletop, but have lost track of where it came from. There are likely to be lots of other egocentric clues. to change or conflict, causing us to switch the reference system that we’re using and become confused. Many difficult experiences are difficult because of poor use of scale. It seems to be more common to try to use metaphors of tabletops for landscape experiences than the other way around, but neither can work.

Whether we perceive an experience as a tabletop or a landscape profoundly affects how our brains make sense of it.

Spatial memory is the memory of landscapes, and it is extremely powerful in humans, as traditions like the Method of loci and other technologies that require our excellent Mapping skills show. Spatial knowledge is phenomenal, and we can often use it more readily than we can reason against it. Inputs to spatial memory are functionally equivalent, we can form spatial memories of landscapes in several ways.

Landscapes are most relevant for information architecture, because most large-scale content-based experiences follow their conventions. understanding naive geography and the corresponding principles of wayfinding are essential for legible landscapes.

The ability to form spatial memories and navigate well seems to be independent of visual ability. People who do not have visual memories can still often navigate well, and congenitally blind people have been shown to have spatial memory. (There are many different neurological causes for a lack of visualization ability or visual memory; some of these causes preserve navigational ability, others don’t.)

See: Spatial scale

References

Avraamides, Marios N., et al. “Functional Equivalence of Spatial Representations Derived from Vision and Language: Evidence from Allocentric Judgments.” Journal of Experimental Psychology: Learning, Memory & Cognition, 2004, pp. 801–14.

Congenitally blind study subjects have been shown to create spatial memory, update it, and use it functionally identically to sighted participants (p. 802)

Avraamides, Marios N., and Jonathan W. Kelly. “Multiple Systems of Spatial Memory and Action.” Cogn Process, 2008

We need to do two kinds of spatial processing (p. 93) - Online, to make decisions in the moment - This is unconsciously dynamically updated as you move through space, but decays quickly (p. 94, 97) - This is how you keep from bumping into things as you walk (both when you can see and when you’re moving through a dark room) - It’s very hard to notice yourself doing this kind of thing. - Offline, to deliberately remember spaces to communicate or reason about them - Planning how you’ll get somewhere, giving someone directions, or deliberately thinking about where something is are all off-line. (p 93) - This spatial reasoning is harder to update, but lasts much longer (p. 97, 99) - You can do it on purpose, but it’s at a coarser level of detail

Some researchers also include viewpoint-dependent visual snapshots of the environment. Some animals can use these to navigate, with a kind of dead-reckoning approach, but humans can’t. (p. 13)

Mcnamara, Timothy P. “How Are the Locations of Objects in the Environment Represented in Memory.” In, 2003.

We have two different reference systems to use to understand spaces: one that tracks objects relative to our own bodies in space and one that tracks objects relative to other objects in space. (p. 1)