[Marker identification and mental rotation in automatic conditions]

A basic finding of mental rotation research is that the time required to discriminate an asymmetrical visual stimulus from its reflected or backward version typically varies systematically with the angular departure of the probe stimulus from its normal upright (or expected) position within the picture plane. This regular dependence of reaction time (RT) on the rotational position (orientation) of the probe stimulus was interpreted by assuming an internal process working as a kind of mental analog of an external physical rotation and being carried out to bring the internal representations of the stimuli into congruence with each other before a decision is made. The present study was designed to investigate whether the dependence of RT on orientation also holds true in the case of a probe that can be identified on the basis of a single distinctive feature. In addition, it was examined whether the efficiency of the underlying processes could be improved by training. Results showed that the dependence of RT on orientation (RT function) does not occur if the presented visual probes differ with regard to a simple feature as, for example, an easily detectable angular difference. A change of the RT function, which would have indicated a change in the speed of the underlying processes, did not occur despite an "automatization" training.