Testing the exteroceptive function of nociception: The role of visual experience in shaping the spatial representations of nociceptive inputs

Abstract

Adequately localizing pain is crucial to protect the body against physical damage and react to the stimulus in external space having caused such damage. Accordingly, it is hypothesized that nociceptive inputs are remapped from a somatotopic reference frame, representing the skin surface, towards a spatiotopic frame, representing the body parts in external space. This ability is thought to be developed and shaped by early visual experience. To test this hypothesis, normally sighted and early blind participants performed temporal order judgment tasks during which they judged which of two nociceptive stimuli applied on each hand's dorsum was perceived as first delivered. Crucially, tasks were performed with the hands either in an uncrossed posture or crossed over body midline. While early blinds were not affected by the posture, performances of the normally sighted participants decreased in the crossed condition relative to the uncrossed condition. This indicates that nociceptive stimuli were automatically remapped into a spatiotopic representation that interfered with somatotopy in normally sighted individuals, whereas early blinds seemed to mostly rely on a somatotopic representation to localize nociceptive inputs. Accordingly, the plasticity of the nociceptive system would not purely depend on bodily experiences but also on crossmodal interactions between nociception and vision during early sensory experience.

Introduction

Pain is as an unpleasant sensory and emotional experience associated with actual or potential tissue damage (IASP, 1994). It usually results from the activation of nociceptors, sensory receptors characterized by high activation thresholds, i.e., by the capacity to respond – at least under normal conditions – to stimuli of high intensity and potentially noxious (Belmonte & Viana, 2008). Pain has therefore an interoceptive function of warning the brain about the occurrence of sensory events having the potential to damage the body (Craig, 2003).

Among other functions of pain, its localization on the skin or in the viscera is of primary importance because it helps to identify which part of the body is being damaged. Conversely to the classical view (e.g., Kandel, Schwartz, Jessell, Siegelbaum, & Hudspeth, 2013), nociceptive inputs can provide detailed and accurate spatial information, suggesting finely-tuned mapping systems for pain (Moore & Schady, 1995). Surprisingly, studies having investigated the mapping organization of nociceptive inputs in the brain (Andersson et al., 1997, Baumgartner et al., 2010, Bingel et al., 2004, Henderson et al., 2007, Mancini et al., 2012) only focused on the somatotopic organization characterized by anatomical representations of the body surface based on the ordered projection of the receptor fields to spatially segregated subgroups of neurons (Penfield & Rasmussen, 1950). However, it has been repeatedly shown that innocuous tactile inputs can be recoded according to spatiotopic representations, i.e., representations that use external space as reference frame, taking the relative position of the limb on which a given stimulus is applied into account (Azanon and Soto-Faraco, 2008, Graziano et al., 1997, Heed and Azañon, 2014, Iwamura et al., 1993, Shore et al., 2002, Smania and Aglioti, 1995, Yamamoto and Kitazawa, 2001). While somatotopic maps allow coding the position of contacts on the skin surface, spatiotopic maps provide an appropriate readout for the brain allowing to identify the object in external space that is in contact with the body, and therefore planning an adequate spatially guided action towards that object (Brozzoli, Ehrsson, & Farne, 2014). Such complex ability to represent somatic information appears even more crucial for nociceptive and painful stimuli since it allows to detect and appropriately react against noxious stimuli that threathen the physical integrity of the body (Legrain & Torta, 2015, pp. 2–20). Demonstrating the brain's ability to map nociceptive inputs according to spatiotopic representations would provide evidence for the exteroceptive function of nociception (Haggard, Iannetti, & Longo, 2013), whose role would be to optimize the monitoring of space around the body and react to potential danger (Legrain, 2017).

Spatiotopic mapping of touch has been repeatedly demonstrated using temporal order judgment (TOJ) tasks during which participants judge the order of occurrence of two successive tactile stimuli, one applied to each hand, and separated by different temporal delays (Heed & Azañon, 2014). It is noteworthy that TOJ tasks are performed with the hands either in a normal uncrossed posture or crossed over the midsagittal plane of the body. Participants’ judgements are typically less accurate when their hands are crossed, and such an effect is accounted by the fact that the somatotopic representation (“Which hand is stimulated?”) mismatches the spatiotopic representation (“Where is the stimulated hand?”) (Shore et al., 2002, Yamamoto and Kitazawa, 2001). This indicates that, when judging the position of a tactile stimulation on the body, its position is automatically recoded according to spatiotopic frames of references (Azanon and Soto-Faraco, 2008, Heed and Azañon, 2014). Importantly, it has been suggested that spatiotopic representations of touch are not innate but instead develop during infancy (Azanon et al., 2017, Pagel et al., 2009). Accordingly, hand posture does not affect the performance of people with early and complete visual deprivation, suggesting that the ability to remap touch according to external frame of reference is, at least partially, shaped by early visual experiences (Crollen, Albouy et al., 2017, Röder et al., 2004).

The first aim of the present experiments was to test the hypothesis according to which nociceptive inputs are automatically coded according to spatiotopic reference frames. Normally sighted participants performed temporal order judgment tasks on thermal stimuli specifically and selectively activating skin nociceptors. Stimuli were applied on each hand dorsum and tasks were performed with the hands either uncrossed or crossed. We expected a decrease of performance in the crossed posture as compared to the uncrossed posture. The second aim was to test the hypothesis of the role of early visual experience in the development of the spatiotopic representation of nociception. We therefore compared the ability of early blind participants and matched blindfolded sighted controls in discriminating the temporal order of nociceptive stimuli. Considering that touch and nociception share the same spatial representation (Legrain & Torta, 2015, pp. 2–20), we would expect early blind participants to be unaffected by hand posture, indicating that their judgements mostly rely on somatotopic representations of nociception. However, due to the higher relevance of nociception in terms of survival and the underlying fundamental role of its spatial representation, we could also expect that early blind participants would be affected by the conflict between somato- and spatiotopic representations during crossed hand posture. In this line, the spatial mapping of nociceptive stimuli would be less dependent on external factors, such as early visual experience, than that of tactile inputs.