摘要：The ability to mentally navigate through time - thinking about the past, present and future - has been coined Mental Time Travel (MTT; Suddendorf & Corballis, 2007). MTT can be understood as a fusion of episodic (Tulving, 1984) and prospective memories and is thought to necessitate high-level cognitive processes such as recursion and analogy that operate on the mental representations of events in time. The theoretical landscape for MTT is closely related to our current understanding of space and number processing in the brain. For instance, a common reference frame for the representation of time, space and number has recently been proposed (Bueti & Walsh, 2009; Dehaene & Brannon, 2011) and evidence in favor of the existence of interferences across magnitudes is on the rise (Lambrechts et al., 2013). In order to gain a better understanding of the possible interfacing between the mental representations of time and space, we directly compared MTT with Mental Space Travel (MST). For this, we developed a psychophysical paradigm which constrained participants to produce a concurrent mapping in mental time and space. A set of real-world events was carefully designed using controlled locations, dates, and contexts for each single event; all events were near-homogeneously distributed across time and spatial longitude and centered on the year 2013 in Paris. Given a set of reference historical events taking place at a particular moment in time or at a particular geographical spot in the world, participants were asked to either project themselves mentally in time or in space in order to perform the task. Specifically, participants had to answer in a 2-alternatives forced-choice (2-AFC) whether a given test event occurredbefore or aftera reference event (time dimension) or whether the test event was locatedeast or westof the reference event (space dimension). Participants’ task was made relative to a spatiotemporal reference frame randomly intermixed to maintain both temporal and spatial layout in participants’ mind. Our measures of interest were participants’ reaction times (RT) and error rates (ER) in order to assess the distance effects in time (Temporal Distance Effect; Arzy, 2009; Santiago, 2010) and its analog in space (Spatial Distance Effect; Denis, 2008; McNammarra, 1986; Noordzij, 2005). Absolute distance effects (aDE) consist in an increase of RT and ER when comparing two events in a spatially or temporally distant reference frame relative to a present or local reference frame. Relative distance effects (rDE) consist in the increase of RT and ER with relative temporal or spatial distance between the event to be compared and the reference, irrespective of its actual temporal or spatial value. aDE can be interpreted as a cost of self-projection in time or space i.e., a shift of the mental reference of self; conversely, rDE may result from the comparison process of temporal or spatial positions of events on the mental time map. One central question was whether the same mental mapping is used for mental travel in time and space. Our results are interpreted in the context of a preliminary model consisting of three distinct cognitive stages, namely: a memory stage in which temporal and spatial information are recalled from memory as allocentric information; a second projection stage in which information is projected in egocentric coordinates as a point on a mental map serving a third comparison stage in which the temporal or spatial position of an event is compared to the egocentric reference. Memory stage: no temporal or spatial distance bias was observed in the memorization pre-test, suggesting that DEs do not simply result from properties of memory encoding. The memorization score pattern was in fact very different for time and space respective to the identity of the events, suggesting at least a separate recall process. Projection stage: a significant aDE was observed in space and time i.e., a significant increase of RTs for both temporal (Past and Future reference) and spatial (Western and Eastern reference relative to Paris) shifts, for both space and time judgments. However, a ERs increase was only seen for temporal shifts in time judgments, and spatial shifts in space judgments. The fact that aDE was domain-unspecific for RTs and domain-specific pour ERs suggests that the shifts impacts the process up to the comparison stage, after the dimension selection. Representation and comparison stage: rDE were found in time and space judgments but no inter-domain rDE was found. The lack of time-space interference in this task strongly suggests the existence of parallel representations for time and space. Their selection may occur at the comparison stage in order to perform the task. Crucially, rDE were best fitted by a logarithmic function of distance, confirming previous results in temporal domain and extending it to spatial domain. A significant interaction between aDE and rDE consisting in a steeper slope of rDE in temporally shifted reference conditions for temporal judgments and in spatially shifted reference conditions for spatial judgments, show the same properties of encoding the distance on both representation of time and space. Taken together, our results suggest that participants build parallel representations for time and space and switched between them while maintaining the mental projection of both temporal and spatial reference. The properties of the observed rDE suggest that the representation of time and space are comparable but whether the effects found are due to the encoding or to the comparison process are yet to be elucidated.
