Imagine yourself in a room surrounded by eleven objects arranged in a circle. You memorize the position of the objects, then you close your eyes, and rotate a third of the way around (120°). Keeping your eyes closed, can you point to the object that was behind you before? Most people can do this without much difficulty, and only take an instant longer than if they’d stayed in the same position.
Now imagine the objects are rotating on a turntable as you yourself rotate, so that the same object is still in front of you — in many respects, it’s as if you’ve never turned and the objects never moved. Yet for most people, it actually takes longer to point to the objects than when they had changed position relative to the objects. What’s going on here?
It shouldn’t be especially surprising that we’re quick to remember the location of objects in a room when we have rotated — after all, this is the kind of thing we must do all the time, like when we look from left to right to decide whether it’s safe to cross the road. But there are also times when an object or set of objects move along with us. If you’re driving a car and turn around a corner, you still remember that the groceries are right behind you in the trunk. So why is it difficult for us to imagine objects moving along with us?
The first two scenarios I described above correspond to an experiment conducted by Weimin Mou, Xiaoou Li, and Timothy McNamara. They paid students to enter a room configured like this:
Instead of symbols, the room contained real objects, like a candle, a hat, and a ball. Standing in the center of the room, the students first memorized the position of all the objects, then were blindfolded and told to point to the objects one at a time using a joystick (e.g. “point to the candle.”). Then they were told to rotate their bodies in place, imagining the objects rotating along with them. Again, still blindfolded, they were tested on the positions of each object. They were also asked to imagine the objects rotating while they remained in place, and to rotate while imagining the objects remained in place. Here are the results:
This graph doesn’t actually show you much. To really see what’s going on here you need to transform the data. We need to consider two possibilities: First, when the objects were in the same position relative to the viewer as when they were learned, like this:
When the position where the objects was learned is the same as where they are imagined, then the learned position minus the imagined position equals zero, like in the first data point on the graph. The second data point on the graph shows how long it took viewers to point to objects when they had rotated, and also imagined the objects rotating. So from the perspective of the viewer, the position of the objects didn’t change, yet it took nearly twice as long to point to the objects.
Now consider when the position of the viewer relative to the objects did change:
A second line is added to the graph. The original, blue line, represents when the actual position of the objects minus the imagined position of the objects equals zero — when the position of the objects relative to the viewer didn’t change. The new, green line, represents when the position of the objects relative to the viewer did change. The first point on the green line shows what happens when the viewer rotates but imagines the objects staying in the same place. The second point shows what happens when the viewer stays in the same place but imagines the objects rotating.
Next the researchers demonstrated to a new set of students that the objects were actually on a turntable that could rotate, while the viewers themselves were on an immobile platform in the center. In the first experiment, viewers hadn’t realized that the platform was able to rotate. Now the experiment was repeated in an identical fashion, except that while the viewers were imagining the objects rotating, the experimenter actually rotated the platform with the objects, so the objects actually moved just as much as the viewers. The viewers were told in advance this was what was happening, but from what they could actually perceive during the experiment, it was identical to the first experiment — remember, they were blindfolded the entire time.
This time, the results were decidedly different:
The two lines now overlap nearly exactly. Whether the objects were shifted relative to the viewer or in the same position relative to the viewer, it took the same amount of time to point to the objects. Simply knowing that the platform supporting the objects could turn drastically affected the results.
So it seems that there are indeed times when it’s just as easy to imagine an object moving along with us as it is to imagine it staying in the same place as we ourselves moves. Previously, many researchers had assumed that we are simply better at imagining objects staying in the same place, but this study demonstrates that when we truly believe an object is moving along with us, we can track it in our minds just as easily as if it was standing still.
Weimin Mou, Xiaoou Li, Timothy P. McNamara (2008). Body- and environmental-stabilized processing of spatial knowledge. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34 (2), 415-421 DOI: 10.1037/0278-73220.127.116.115