A reliable cognitive map forms only after repeated travel through the same scene

What it is

The first pass through a campus yields two or three turns; the third pass along the same roads is when a plan can be drawn in the head. A cognitive map (survey knowledge) is a survey representation from which any point can be indicated from any other, and it does not grow on the first trip. The first trip is usually landmarks and one route. Survey needs repeated travel, through the same patch, from different directions.

This is not whether there is a distinctive tower, and not whether a minimap can insert survey from the outside. It asks how many walks the person’s own map needs before it is reliable.

Why it happens

Siegel and White’s classic layering is landmark knowledge → route knowledge → survey knowledge. A landmark can be met once; a route strings landmarks into “then left”; survey weaves the route net into shortcuts, detours, and pointing in from the side. Weaving needs the same node to be met from different edges — walk only a fixed tour and the route can be fluent while survey is still missing. In virtual scenes the first pass is also often full of novelty and control load; working memory goes to “how to grab, how to jump,” not to spatial structure.

“Reliable” means being able to point, unseen, toward a shortcut never walked, not being able to retrace the original route. Retracing only proves the route is still there. Shortcuts and sideways pointing are what prove survey has grown.

Studying it

Have people walk the same scene once, three times, or in from different entrances, then run a shortcut task, off-site pointing, and a sketch map. Compare pass count and route diversity, not landmark count. Sketches after pass one are often a line; after pass three, rooms start to sit in the right relative places. Pointing at a never-visited room is harsher than pointing at the exit just used.

Seeing a minimap and growing survey by walking are not the same object. People who saw a map will shortcut better; that is the symbolic layer. Whether they can still point with the map gone is the cognitive map. A retest days later is consolidation, not working memory from the same session.

Where it stops holding

Tiny or strongly axial spaces (one corridor, a symmetric court) may be enough on pass one, and repetition gains little. Professionals and the environmentally fluent encode differently; a lab “three passes” will not fit a firefighter walking their own station. If the scene changes every entry (procedural, multi-user rebuilds), repetition has no same map to build. A guided, no-choice tour can add passes without growing survey — deciding which way to turn is part of the weave. Severe distance compression can wrong the metric of a sketch while leaving topology intact; do not use metre accuracy to deny survey that has already grown.

Applying it

• For spaces people must later find their own way through, let them pass key nodes from at least two directions. Do not ship only a forced tour.
• Leave load on pass one for arrival and safety; put shortcut-dependent tasks on pass two and three.
• If the product promises “once you have been here you will know,” accept with shortcuts or next-day pointing, not with “followed the arrows to the end.”
• How to check: after one pass of the same campus, point at an unvisited side door; retest after two more passes. Only once pass count is up and routes have diversified should pointing stabilise, and only then has a cognitive map been built. If pass one is already correct, the space may be too small to generalise to a large campus.

Related

• Same group: N4.09.1 Without stable landmarks, virtual space is easier to get lost in than real space · N4.09.2 Unique, identifiable landmarks help more than stylistically uniform scenery · N4.09.3 A minimap or compass gives global position but cannot replace environmental direction cues · N4.09.5 Teleport skips path information and weakens route-memory formation
• Nearby: N4.01 Teleport Locomotion · N4.05 Redirected Walking
• Search terms: cognitive map · survey knowledge · shortcut task