Abstract: The human visual system can represent an object’s spatial structure using either first-person perspective or third-person perspective. It can also utilize either first-person or third-person processing mode to perform mental rotation. Since Shepard and colleagues’ classic study, experiments over next 30 years did much to elucidate the basic nature of mental rotation. However, most of the studies were only the ones of first-person rotation, and third-person rotation has been neglected in the field of mental exploration. So, this study’s specific aim was to compare the relative difficulty of first- and third-person rotation of imagined environment by the experiments. The hypothesis was that third-person rotation would be made easier than first-person rotation in the transverse and coronal planes.
Two experiments were conducted, each comparing first- and third-person performance in the transverse (Experiment 1) or coronal (Experiment 2) plane, respectively. Seventy-six right-handed undergraduates took part in the experiments at the age between 19 and 21. Each experiment was a 2 (character task: first- and third-person rotation) × 4 (degree) mixed-design. Each participant performed either first-person rotation or third-person rotation. In first-person task, participants were asked to imagine themselves rotating in the center of a four-object array a given number of degrees and were then asked to name the object at a particular position (experiment 1: right, left, front, or back; experiment 2: top, bottom, right, left) relative to their imagined orientation. In third-person task, participants were asked to imagine another person rotating in the center of a four-object array ahead of them a given number of degrees and were then asked to name the object at a particular position (experiment 1: right, left, front, or back; experiment 2: top, bottom, right, left) relative to the person’s orientation after rotation. Rts and number of errors were recorded and analyzed. RT was measured from the end of the experimenter’s question to the beginning of the participant’s response.
The results confirmed the hypothesis. In experiment 1, the third-person task was performed faster and more accurately than the first-person task in the transverse plane. The ANOVA on reaction time revealed main effects of character task and degree, as well as a significant character task × degree interaction. More specifically, the first-person and third-person RTs were not different at 0°, but the overall latency of first-person responses was greater than that of third-person responses for 90°, 180°, and 270° of rotation. The ANOVA on errors also indicated main effects of character task and degree. The results of Experiment 2 were similar to the findings in Experiment 1.It was found reduced RTs and errors in the third-person rotation compared to the first-person rotation in the coronal plane in which people cannot naturally locomote.
The present experiments revealed that there is an advantage of third-person rotation over first-person rotation. Third-person rotation is easier than first-person rotation in the transverse plane and coronal plane

Key words: mental rotation, first-person, third-person, transverse plane, coronal plane