Heterogeneous timescales are spatially represented

There is little doubt that cognitive processes involved in perceiving and comparing the duration of the sounds played by a musical instrument are very different from those involved, for instance, in recalling when our latest holidays took place. However, despite their clear difference, we ascribe both processes to the time domain, either in the form of a duration or of a mental time travel. Here we maintain that, despite the several meanings of “time” and its intrinsically heterogeneous nature, a common and somehow surprising characteristic of its representation is its spatial nature. Evidence will be provided suggesting that sensory time (the sounds of the instrument), time travel (last holidays) as well as conceptual time are all spatially represented. Whether this spatial format is the same for all “categories” of time is still unclear, as well as it is unclear whether category-specific spatial layouts exist. The possibility that several heterogeneous aspects of time might be all processed in spatial terms allows to better understand a number of apparently disconnected findings within the vast time domain. We will describe the prominent role played by writing habits in giving a direction to this representation. The possibility will be discussed that a more general tendency exists to represent in a spatial format all ordered sequences, as well as abstract concepts. The interest on the way humans process and perceive time can be readily explained by the ubiquity of time in human life and by its multifaceted and sometimes elusive characteristics. Moreover, time processing is relevant for several domains, including not only philosophy, linguistics, cognitive science, and neuroscience, but also anthropology, history, and biology. Humans are equipped with a device (or several devices, this does not seem to be currently known) allowing them to deal effectively with all aspects of time, including duration estimates, time travel, and time conceptualization. A key empirical finding supporting our purposes here is that, across a number of formats and paradigms, time processing interacts with spatial processing and with the mechanisms subtending the deployment of spatial attention. This allows to consider the way humans perceive different time durations, and represent heterogeneous time concepts, as spatial in nature. A thorough review on these aspects can be found in Bonato et al. (2012a), where a number of analogies with the spatial representation within the numerical domain -which is also characterized by ordinalityare discussed. Themore consistent line of evidence supporting the possibility that time is spatially represented comes from those studies showing an interaction between a time-related characteristic and the response side. Regardless of the aspect of time that is under investigation (perceptual or representational), the typical experimental manipulation requires the use of two stimulus-response mappings, whereby the same time-related stimulus receives a left response in one mapping and a right response in the other. By using two temporal durations Vallesi et al. (2008, exp 1) have shown consistent associations between left side and shorter durations (1 s) and between right side and long durations (3 s). The interaction does not depend on the responding hand but on the side of space where response is performed (exp 3 and 4) and was still present when more durations (six, from 0.5 to 3.5 s) had to be compared (exp 5; see Conson et al., 2008 for similar findings with auditory stimuli). An interaction with lateralized response keys is also found when the to-be-judged temporal references are related to mental time travel/time concepts, (e.g., beforeafter). These studies typically adopt much longer temporal durations. It is the case, for instance, of the months of the year, whereby faster responses are found when one of the first months (e.g., February) has to be responded to with a left-sided response and one of the last months (e.g., January) with a right-sided response (Gevers et al., 2003). Evidence for the automaticity of this association stems from the fact that it was found also when the task did not involve explicit access to the time aspect (Gevers et al., 2003). Leftright associations can be elicited also when the timeframe in a temporal comparison task extends to several years, as shown by Weger and Pratt (2008) who asked for a comparison of actors who were famous several decades ago (e.g., Charlie Chaplin) vs. those who were famous at the time of testing (e.g., Brad Pitt). Ishihara et al. (2008) provided evidence that left-before vs. right-after associations are present also for brief, sensory stimuli (sounds with a 20ms duration). In their study, the time of appearance of a target sound after a sequence of seven equally spaced (500ms interval) sounds had to be compared with the temporal lag of half a second dividing the preceding sounds. Participants showed a preference for right-sided responses when the target sound was presented later in time with respect to the expected interval (delay of 215ms) and a preference