Do compasses or distant mountains help people learn a new place?
Giving people a compass or a visible mountain range while they learned a virtual campus did not improve their memory for its layout.
Source
This is not the way: global directional cues do not improve spatial learning in an immersive virtual environment
Study at a glance
- Design
- Human experiment — Two between-subjects experiments in immersive VR on an omnidirectional treadmill: compass vs no cue (Exp 1), mountain range vs compass (Exp 2), with Bayesian tests.
- N
- N=123 · Pointing-task samples: 56 in Experiment 1 (29 compass, 27 control) and 67 in Experiment 2 (36 mountain, 31 compass); model-building samples were slightly smaller (54 and 61).
- Population
- Undergraduate students at the University of Florida.
- Outcome
- Pointing error between buildings and map accuracy (bidimensional regression R squared) after learning a virtual campus.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
Students wearing a VR headset walked on an omnidirectional treadmill through Virtual Silcton, a virtual campus, learning the locations of eight buildings along set routes. In Experiment 1 some had a virtual compass on their controller and others had no cue; in Experiment 2 some saw a distant mountain range and others had a compass pointing the same way. Afterwards they pointed from each building to the others and built a map by dragging buildings into place.
What they found
In Experiment 1 Bayesian tests gave moderate evidence that the compass made no difference to pointing error or map accuracy. In Experiment 2 the mountain range was no better than the compass, and combining both experiments still showed no benefit of any cue. People oriented their maps to the direction of the first route they walked rather than to the cue. Men and video-game players performed better overall, but neither group benefited more from the cues.
The limits
What it doesn't show
The samples were young college students from a flat part of the U.S., who may not be used to using mountains for orientation. Experiment 2 had no no-cue control, so the combined analysis compares groups from different semesters and is exploratory. The mountain range was not very visible from route starting points and participants were not made to look around, so some may have noticed it late. The environment may already have offered enough landmarks and body-based cues that a global cue was unnecessary, so the results may not apply to sparser settings.
Key terms
- Reference frame
- An orienting direction relative to which a person represents the positions of landmarks and themselves.
- Construction hypothesis
- The idea that navigational aids help by encouraging people to build new, more accurate spatial representations aligned to the aid.
- Access hypothesis
- The idea that navigational aids help only by giving access to spatial knowledge already encoded relative to that direction.
- Salience hypothesis
- The idea that a directional cue will only be used if it is highly visible and embedded in the environment.
- Bayes factor
- A ratio of how likely the data are under one hypothesis versus another; values below one favour the null.
- Bidimensional regression
- A method that scores how well a drawn map matches the true layout after allowing for differences in scale, rotation and position.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Which hypothesis did the results most clearly undermine?
Common questions
Does this mean compasses are useless?
Not in general. The results suggest people don't spontaneously use a global cue to build a map of a new place when other cues are available; a compass can still help when knowledge was already encoded relative to north.
Why use Bayesian statistics here?
Because the key claim is a null effect, and Bayes factors can quantify evidence for no difference rather than just failing to reject it.
Did participants ignore the cue entirely?
No. They could point toward the cued direction better than chance, and people who did so more accurately also learned the layout better, but the cue did not shape how their maps were oriented.
More on Spatial cognition