Imagery and Action • Maps
People are able to use culturally designed representations, like maps, to simplify tasks. Imagery research has typically assumed that imagery depends on perceptual representations that are unmediated. We have been examining how the use of maps can facilitate imagery for large-scale tasks, even when people’s eyes are closed.
Imagine you are sitting in a sculpture garden. You are told to memorize the positions of sculptures A, B, and C. You then close your eyes. You are told to imagine that you are at sculpture A facing B. When you have successfully completed that task, you tell the experimenter. You then hear that your job is to point so that your arm makes a parallel to the line from C to B (as if you were at station A facing B). The figure provides a rough schematic of the task. We measure the time it takes to complete 1, then 2, and finally the accuracy of 3.
That is the No Map condition. In the Map + Action condition, you hold a map of the sculpture garden in your hand. It is correctly oriented so that north on the map is the same as north in the sculpture garden. You complete the same sequence of tasks, only once you close your eyes, you rotate the map so that A lines up with B in front of you. You then point to the parallel made by C to B. Finally, there is a Map + No-Action condition in which you hold the map, but you are not allowed to turn it to help with the task. You complete all three conditions.
A pilot study (recently replicated with 25 subjects) had 6 subjects who each completed four trials for each condition. The trials sampled a variety of angles and sculptures. Every subject was both much faster and more accurate when s/he turned the map. The study permits many different data analyses using the angles and measures. For brevity, the table below simply displays the average total latencies and the average absolute errors in the judgments of relative direction. The parentheses hold the average within-subject, standard deviation to indicate the stability of the subjects’ performances in the Map + Action condition.
| Avg. Solution Time (sec) | Avg. Absolute Error (deg) | |
| Map + Action | 9.9 (2.4) | 20.1 deg. (19.5) |
| Map + No-Action | 17.8 (6.0) | 53.4 deg (33.5) |
| No-Map + No-Action | 19.2 (7.0) | 48.2 deg. (38.4) |
The results show that actions mediated by representations can facilitate imagery for spatial updates. (A second study showed that turning a blank sheet of paper did not facilitate imagery compared to turning the map.) The results also show that manipulations of small-scale space (the map) can support inferences about large-scale space (the sculpture garden), though one might debate if using the map is a fair comparison. There are also interesting finer results; for example, everyone performs about the same when using action, but there appear to be individual differences in the no-action conditions. Regardless, for now, it is sufficient to note that we have a robust task that we can use to explore the effects of actions mediated by maps. Subsequent studies will examine whether having the experimenter turn the map in the subject's hands will also facilitate imagery, or whether the movement of the map has to come from their own actions.
