Audio-visual sensory deprivation degrades visuo-tactile peri-personal space

Highlights

• Peripersonal space, interoceptive accuracy (IAcc) was measured after sensory deprivation.

• Phenomenology associated with sensory deprivation (SD) was assessed.

• Peri-trunk space was ill-defined in participants with better IAcc after SD.

• The gradient of peri-trunk space related to IAcc and aberrant self-phenomenology.

Abstract

Self-perception is scaffolded upon the integration of multisensory cues on the body, the space surrounding the body (i.e., the peri-personal space; PPS), and from within the body. We asked whether reducing information available from external space would change: PPS, interoceptive accuracy, and self-experience. Twenty participants were exposed to 15 min of audio-visual deprivation and performed: (i) a visuo-tactile interaction task measuring their PPS; (ii) a heartbeat perception task measuring interoceptive accuracy; and (iii) a series of questionnaires related to self-perception and mental illness. These tasks were carried out in two conditions: while exposed to a standard sensory environment and under a condition of audio-visual deprivation. Results suggest that while PPS becomes ill defined after audio-visual deprivation, interoceptive accuracy is unaltered at a group-level, with some participants improving and some worsening in interoceptive accuracy. Interestingly, correlational individual differences analyses revealed that changes in PPS after audio-visual deprivation were related to interoceptive accuracy and self-reports of “unusual experiences” on an individual subject basis. Taken together, the findings argue for a relationship between the malleability of PPS, interoceptive accuracy, and an inclination toward aberrant ideation often associated with mental illness.

Introduction

Prominent models of self-consciousness stress the role of integration between multisensory exteroceptive (i.e., processing of external sensory stimuli; Blanke, 2012, Blanke et al., 2015, Bermudez et al., 1995) and interoceptive (i.e., processing of sensory stimuli from within the body; Damasio, 2010, Seth et al., 2011, Craig, 2002, Craig, 2009, Critchley and Seth, 2012) signals as essential in the formation of a pre-reflexive form of bodily self-consciousness (BSC). BSC includes the feeling of owning a body, of being at a specific location in space, and of experiencing the world from a particular first-person perspective (Blanke & Metzinger, 2009). Intriguingly, signals relevant for BSC are not limited to the body itself – with crucial contributions from the tactile, proprioceptive and thermo-nocioceptive systems (see Haggard, Iannetti, & Longo, 2013 and Legrain, 2017, for theoretical postulates casting nocioception as inherently multisensory, fundamental in body representation, and most importantly for the current purpose, with a double function both in exteroception and interoception) - but they equally extend within one’s peri-personal space (PPS; Blanke et al., 2015, Noel, Grivaz, et al., 2015, Noel, Pfeiffer, et al., 2015, Salomon et al., 2017, Bernasconi et al., 2018); that is, the space immediately adjacent to and surrounding the body (Di Pellegrino et al., 1997, Rizzolatti et al., 1981, Rizzolatti et al., 1997, Serino et al., 2015). Indeed, all physical interactions between the individual’s body and the external environment take place within the PPS. Accordingly, by modulating multisensory cues not only from the body, but also within the PPS, it is possible to alter the different components of BSC (Noel, Grivaz, et al., 2015, Noel, Pfeiffer, et al., 2015, Salomon et al., 2017). For example, by manipulating visuo-tactile spatio-temporal congruencies (tactile on the body and visual in the PPS) it is possible to induce ownership for an artificial hand (as in the rubber hand illusion; RHI, Botvinick & Cohen, 1998), face (as in the enfacement illusion; Tsakiris, 2008) or even the whole body (as in the body-swap illusion, Petkova & Ehrsson, 2008). Further, the administration of controlled multisensory cues at the body and within the PPS may even shift the perceived location of the self in space (as in the full body illusion; FBI, Lenggenhager et al., 2007, Noel, Grivaz, et al., 2015, Noel, Pfeiffer, et al., 2015, Salomon et al., 2017) and the direction of the first-person perspective (Petkova and Ehrsson, 2008, Ionta et al., 2011).

On the other hand, focusing on internal as opposed to external bodily cues, experimentally induced altered states of BSC due to the administration of conflicting exteroceptive sensory signals (Botvinick and Cohen, 1998, Lenggenhager et al., 2007, Ehrsson, 2007) have been shown to affect aspects of interoception such as autonomic responses (Ehrsson, Wiech, Weiskopf, Dolan, & Passingham, 2007) and neural representations of cardiac afferents (Park et al., 2016). In the case of the RHI (Botvinick & Cohen, 1998), for instance, illusory ownership for a rubber hand is associated with a reduction of the skin temperature of the participant’s real hand (Moseley et al., 2008, but see de Haan et al., 2017) and an increase in its histamine reactivity (Barnsley et al., 2012). Similar temperature effects have been reported for the FBI (Salomon, Lim, Pfeiffer, Gassert, & Blanke, 2013). In addition, there exists a negative relation between an individual’s capacity for interoceptive accuracy (in this case operationalized as the ability to detect their own heartbeats without measuring their pulse) and their proneness to the RHI (Tsakiris, Tajadura-Jimenez, & Costantini, 2011) and the enfacement illusion (Tajadura-Jiménez et al., 2012, Tajadura-Jiménez and Tsakiris, 2013). These results suggest that individuals with low interoceptive accuracy might rely more heavily on external sensory cues in forming a representation of their bodily self, and hence be more prone to illusions mediated by multisensory external cues. More directly, recent data demonstrated a strong relationship between exteroceptive and interoceptive sensory signals in giving rise to BSC (Suzuki et al., 2013, Aspell et al., 2013, Adler et al., 2014). Indeed, artificially introduced matches (illusion condition; vs. mismatches or control condition) between cardiac and visual signals provoke the RHI (Suzuki et al., 2013) and the FBI (Aspell et al., 2013; see also Adler at al. (2014) for a FBI using respiratory signals), much in the same way that spatially concordant visual-tactile stimuli promote the illusion. Thus, it appears that there are strong relationships between the manner in which individuals process exteroceptive and interoceptive sensory signals, and in addition, this relationship appears to modulate BSC.

Although prior studies have attempted to manipulate interoceptive accuracy in order to measure concomitant body representation changes (Ainley et al., 2012, Ainley et al., 2013, Stevens et al., 2011, Khalsa et al., 2008, Fairclough and Goodwin, 2007, Maister and Tsakiris, 2013) few have studied the relationship between the processing of interoceptive and exteroceptive signals within and beyond the PPS. Indeed, within this framework, Legrain and colleagues have highlighted the nocioceptive system as one straddling interoceptive and exteroceptive domains (Haggard et al., 2013) to demonstrate that visual stimuli within but not beyond the PPS modulates nocioceptive processing, in particular in the temporal dimension (De Paepe et al., 2014, De Paepe et al., 2017, Filbrich et al., 2017; further see Bultitude, Walker, & Spence, 2017 for recent corroborative evidence from a different group). Nonetheless, a direct causal manipulation offsetting the weighting between exteroceptive and interoceptive signaling, and subsequently measuring the impact of this remapping on PPS and BSC is lacking. Indeed, while the relationship between interoceptive accuracy and BSC has been described in prior work (Tsakiris et al., 2011, Tajadura-Jiménez et al., 2012, Tajadura-Jiménez and Tsakiris, 2013, Ainley et al., 2012), only a single study (Ferri, Ardizzi, Ambrosecchia, & Gallese, 2013) has attempted to investigate whether interoceptive accuracy was associated with an autonomic response (more precisely, respiratory sinus arrhythmia) indexing PPS. Further, while the above mentioned study (Ferri et al., 2013) showed a positive association between interoceptive accuracy and an autonomic response to interpersonal stimulation (i.e., another person’s hand approaching the participant’s body) at the boundary of the PPS anchored on the hand (i.e., the peri-hand space; see Serino et al., 2015), no study has investigated the relation between interoception and fundamental characteristics of an individual’s PPS, such as it’s size (i.e., the spatial extent over which exteroceptive signals modulate somatosensory processing on the body) or it’s gradient (i.e., the sharpness in the division between peri- and extra-personal space), as well as their respective link (interoception and PPS) to the experience of the self in space.

Indeed, if the neural encoding of the PPS (and thus the boundary between the peri- and extra-personal space) is conceived as encoding the interface between the individual’s body (or body-related space) and the environment, it is possible that PPS and interoceptive accuracy interact in building one’s BSC (see Noel, De Niear, Lazzara, & Wallace, 2017, for a similar argument). If true, one would expect a relationship between the spatial extent (i.e., size) and/or shape (i.e., gradient or the way in which ‘far’ and ‘near’ space are distinguished) of one’s PPS and interoceptive accuracy. Further, it may be that certain PPS representations, such as the peri-trunk space (i.e., the PPS anchored on the trunk) – due to its association with self-location (Blanke, 2012, Blanke et al., 2015) – and not others, such as the peri-face space (i.e., the PPS anchored on the face), are related to interoception accuracy and BSC.

As a result of these questions, the first aim of this study is to highlight potential relationships between PPS, interoceptive accuracy, and BSC by attempting to manipulate the relative strength of exteroceptive and interoceptive signals. To this aim, we submitted healthy subjects to a short session of audio-visual deprivation in an anechoic chamber, in an attempt to reduce exteroceptive processing and hence potentially enhance interoception processing. That is, we conceived that the most direct approach in testing the relation between interoception and exteroception in shaping PPS and BSC was to eliminate or reduce the degree of exteroceptive signals available, potentially enhancing interoceptive processing (either due to the fact that the degree of exteroceptive information is reduced or as a consequence of a homeostatic process). After audio-visual deprivation in the anechoic chamber or after the same amount of time in a normal environment (as a control condition) – but importantly while still in either the anechoic or standard room – we assessed (1) participant’s interoceptive accuracy by means of a heartbeat-counting task (Schandry Task; Schandry, 1981), (2) the size and gradient of their peri-face and peri-trunk representation by means of a visuo-tactile space-dependent interaction task (for analogous task in the auditory domain see Noel, Grivaz, et al., 2015, Noel, Pfeiffer, et al., 2015, Galli et al., 2015, Serino et al., 2017), and (3) their phenomenological experience of the self-in-space during audio-visual deprivation. We hypothesized that as a consequence of the degraded exteroceptive signals; (1) audio-visual deprivation would result in (1) degraded division between the near and far space (i.e., shallower PPS gradients) for both the peri-face and peri-trunk, and (2) enhanced interoceptive accuracy – as other sensory evidence is diminished. Further, we predict that (3) audio-visual deprivation will result in anomalies of the phenomenology of the “self-in-space”, which would be most readily related to the PPS representation around the trunk (Noel, Grivaz, et al., 2015, Noel, Pfeiffer, et al., 2015, Serino et al., 2015, Salomon et al., 2017) than the face.

Furthermore, the manner in which the nervous system represents the body and integrates information from distinct sensory modalities has also been postulated to impact individual differences in personality traits (Damasio, 2010, Seth et al., 2011, James, 1890, Damasio, 2000, Seth, 2013, Berlucchi and Aglioti, 2010) and higher-order levels of cognition (Canzoneri et al., 2016, Stevenson et al., 2014, Stevenson et al., 2017, Postmes et al., 2014, Noel et al., 2016, Noel, Cascio, et al., 2017, Noel, Blanke, et al., 2017, Noel, Lytle, et al., 2017). In fact, prior studies have demonstrated a specific interplay between personality traits, the valence of stimuli utilized (e.g., spider vs. butterfly; de Haan et al., 2016), and PPS representation (Sambo and Iannetti, 2013, Fossataro et al., 2016, Iachini et al., 2015). Similarly, studies have linked interoceptive accuracy with emotional processing (Pollatos et al., 2007, Critchley et al., 2011, Werner et al., 2013, Durlik and Tsakiris, 2015) and have attempted to establish a functional relationship between interoceptive measures and psychopathology (Pollatos et al., 2007, Dunn et al., 2007, Mussgay et al., 1999, Domschke et al., 2010, Ehlers and Breuer, 1992, Noel et al., 2017). Therefore, in the present study we also tested how changes in exteroceptive and interoceptive processes induced by audio-visual deprivation affects personality traits by administering personality questionnaires (Mason et al., 1995, Claridge et al., 1996, Morrison et al., 2002, Launay and Slade, 1981) after audio-visual deprivation or the control condition. The questionnaires administered indexed psychosis-proneness, principally Schizotypy (Claridge et al., 1996, Mason et al., 1995) and hallucinations (Morrison et al., 2002). Similarly to above, we hypothesized that audio-visual deprivation would accentuate Schizotypic and hallucinatory reports, and that these would be closely tied to PPS representation and the experience of the “self-in-space”.