Objective Spatial locations of objects in the surrounding environment are not only perceived based on observer-centered (egocentric) coordinate, but also on object-centered (allocentric) coordinate in which the referenced center is an object instead of being the observer.For example, when a basketball player is doing defense rebound he needs to know where the ball is related to his hands (egocentric location) as well as the balis position related to the basketry (allocentric location).In addition to spatial perception, allocentric coordinate has also been used in other cognitive functions, such as spatial long-term memory, in which the locations of objects are mapped in allocentrie coordinate.In this case, the memorized spatial relationships among objects never change regardless of the alteration of observers location in space.However, when interacting with the surrounding environment, one need first to know where the objects are related to his body, then the navigation system will utilize the egocentric spatial information to guide ones behavior.In other worlds, the perceived or memorized allocentric spatial information needs to be transformed to egocentric coordinate prior to the formation of motor command.It has been known for long time that information could be transformed between coordinates.Byrne and Becker proposed a model to explain the transformation from egocentric coordinate to allocentric coordinate based on studies of single neuron recording.Andersen and his colleagues found that the visual, memory and saccadic activities in monkeys posterior parietal cortex (PPC) were strongly modulated by the orbital position of the eyes during fixation.Such phenomenon has been named as gain field modulation and linked to information transformation from retinotopic coordinate to body-centered coordinate.However, these studies did not solve the problem of how spatial information from allocentric coordinate converse to egocentric coordinate.Methods In the present study, we measured the manual reaction time (RT) while six human subjects were performing a position discrimination task in which the behavior response purely relay on the allocentric information of a target related to a referenced object (allocentric left or right), but regardless of the egocentric positions of the target (egocentric left or right).Results Our results indicate that the conversion of allocentric coordinate to egocentric coordinate has the following characteristics: (1) The conversion is significantly faster when targets allocentric position is compatible with egocentric position (SAME) comparing with incompatible condition (OPPO); (2) The conversion difference between SAME and OPPO is increased following the increase of the visual angle between targets location and the observers sagittal middle line in egocentric coordinate; (3) The conversion difference between SAME and OPPO is larger, in most right hand subjects (5 out of 6), when target is in egocentric left than in egocentric right.Conclusion We conclude that each hemisphere prefers to integrate the contralateral allocentric information with contralateral egocentric information.The greater RT in OPPO condition accompanies with greater visual angle indicates that each hemisphere preferentially receives and progresses egocentric contralateral information and has week input from ipsilateral side.Moreover, the greater difference between SAME and OPPO when target appears in egocentric left than in egocentric right, is consistent with findings in previous studies, i.e.fight hemisphere is preferentially dealing with spatial information.