Is Rotation of Visual Mental Images a Motor Act?

Abstract

The relationship between the mechanisms of vision and of visual mental imagery, such as mental rotation, has been well established. The relations between mental rotation and motor action, on the other hand, have hardly been studied, despite the fact that, ecologically, most non-mental rotation is the result of motor actions such as manual manipulation of medium-sized objects. I propose the following motor/imagery hypothesis: transformations of visual mental images are functionally closely related to the planning stages of the motor system. There is a certain amount of indirect support for this hypothesis in the literature. In the present work the motor/imagery hypothesis was tested directly, by means of a dual task paradigm. Subjects performed two tasks simultaneously: the Shepard - Cooper visual imagery task, which involves mental rotation; and a motor rotation (which could not be seen), turning a joystick handle in the plane of the visual image at a previously learned angular speed and direction. The motor/imagery hypothesis predicts a correlation between corresponding features of the two rotations.

The results strongly confirm the motor/imagery hypothesis. The concurrent motor task shifts the classic V-shaped mental rotation RT curve: mental rotation is faster and less error-prone when it is in the same direction as the motor rotation than when it is in the opposite direction. Moreover, there is a strong correlation between the speeds of the two rotations: all else being equal, subjects' mental rotations were slower when their manual rotations were slower, and vice versa.