Abstract: Intrinsic frame of reference system theory is a newly-developed theory in spatial memory researches. It suggests that when people learn locations of objects in a new environment, they are inclined to mentally represent a layout of objects according to a frame of reference system defined by the collection of objects. Many studies have proven that the dominant axis of intrinsic frame of reference may differ from their egocentric orientation. But there was few research in intrinsic frame of reference system theory on the effect of orientation information provided by the objects themselves in the scene. This research focused on the effect of orientation coincidence of objects in environment on people’s construction of intrinsic frame of reference system and hypothesized the coincident orientation of objects would be one of the choices of the dominant axis in intrinsic frame of reference system in symmetrical scene.
53 undergraduates took part in three experiments. In Experiment 1, participants stood at 0 degree facing the axis of symmetry and learned a symmetrical scene that was composed by balls with no apparent headings. In Experiment 2, participants stood at 0 degree and learned a symmetrical scene composed by toy bears with their faces orienting 315 degree, which was different from the axis of symmetry. In Experiment 3, participants stood at 0 and 315 degree and learned the same scene as that in Experiment 2. After learning, the participants needed to finish triplet recognition task in each experiment. In the task, they were required to judge if the directional relationship in these triplets were the same as the learned scene. The correct response latencies to the targets were analyzed by ANOVA.
In Experiment 1, when the objects in scenes had no apparent headings, the results of repeated measures showed the mean response latency to triplets including the direction of axis of symmetry was significantly shorter than that not including the direction. In Experiment 2, when the objects in scenes had apparent headings and they all oriented the direction different from the axis of symmetry, the difference between the mean response latencies to triplets including the axis of symmetry and including the coincident orientation of objects was not significant. This result was different from the significant difference found in Experiment 1. In Experiment 3, the difference between mean response latencies to triplets including the axis of symmetry and including the coincident orientation of objects was still not significant. The result was the same as that found in Experiment 2, even if the scene was learned from both the axis of symmetry and the coincident orientation of objects to eliminate the possible interference of learning viewpoint.
Results indicate that orientation coincidence of objects in symmetrical scenes affects the construction of intrinsic frame of reference system. However, there is no significant difference between the symmetry axis and the coincident orientation in the chances to be the dominant axis of intrinsic frame of reference system.

Key words: orientation coincidence, intrinsic frame of reference system, axis of symmetry, viewpoint, triplet recognition