Content uploaded by Olga Dobrushina
Author content
All content in this area was uploaded by Olga Dobrushina on Aug 17, 2021
Content may be subject to copyright.
1
Sensory integration in interoception: Interplay between top-down and bottom-up
processing
Dobrushina Olga R1* ORCID: https://orcid.org/0000-0002-9493-4212
Arina Galina A2 ORCID: https://orcid.org/0000-0003-1321-9354
Dobrynina Larisa A1 ORCID: https://orcid.org/0000-0001-9929-2725
Novikova Evgenia S1 ORCID: https://orcid.org/0000-0001-5236-7259
Gubanova Mariia V.1 ORCID: https://orcid.org/0000-0002-9893-712X
Belopasova Anastasia V1 ORCID: https://orcid.org/0000-0003-3124-2443
Vorobeva Viktoriia P2 ORCID: https://orcid.org/0000-0001-9111-983X
Suslina Anastasia D1 ORCID: https://orcid.org/0000-0001-7425-1572
Pechenkova Ekaterina V4 ORCID: https://orcid.org/0000-0003-3409-3703
Perepelkina Olga S3 ORCID: https://orcid.org/0000-0001-9357-8407
Kremneva Elena I1 ORCID: https://orcid.org/0000-0001-9396-6063
Krotenkova Marina V1 ORCID: https://orcid.org/0000-0003-3820-4554
1 Research Center of Neurology, Moscow, Russia, Volokolamskoe 80, 125367 Moscow, Russia
2 M.V. Lomonosov Moscow State University, Faculty of Psychology, Mochovaya 11, Bld. 9,
125009 Moscow, Russia
3 Intel Corporation, Internet of Things Group, 2200 Mission College Blvd., Santa Clara, CA
95052, USA
4 HSE University, Armyanskiy Ln. 4, Bld. 2, 101000 Moscow, Russia
* Correspondence:
Olga Dobrushina
125367 Volokolamskoe Shosse 80, Moscow, Russia
+7 (916) 130-31-44
dobrushina@mipz.ru
2
Abstract
Although the neural systems supporting interoception have been outlined in general, the exact processes
underlying the integration of visceral signals still await research. Based on the predictive coding concept,
we aimed to reveal the neural networks responsible for the bottom-up (stimulus-dependent) and top-down
(model-dependent) processing of interoceptive information. In a study of 30 female participants, we
utilized two classical body perception experiments — the rubber hand illusion and a heartbeat detection
task (cardioception), with the latter being implemented in fMRI settings. We interpreted a stronger rubber
hand illusion, as measured by higher proprioceptive drift, as a tendency to rely on actual sensory
experience, i.e. bottom-up processing, while lower proprioceptive drift served as an indicator of the
prevalence of top-down, model-based influences. To reveal the bottom-up and top-down processes in
cardioception, we performed a seed-based connectivity analysis of the heartbeat detection task, using as
seeds the areas with known roles in sensory integration and entering proprioceptive drift as a covariate.
The results revealed a left thalamus-dependent network positively associated with proprioceptive drift
(bottom-up processing) and a left amygdala-dependent network negatively associated with drift (top-
down processing). Bottom-up processing was related to thalamic connectivity with the left frontal
operculum and anterior insula, anterior cingulate cortex, hypothalamus, right planum polare and right
inferior frontal gyrus. Top-down processing was related to amygdalar connectivity with the rostral
prefrontal cortex and an area involving the left frontal opercular and anterior insular cortex, with the latter
area being an intersection of the two networks. Thus, we revealed the neural mechanisms underlying the
integration of interoceptive information through the interaction between the current sensory experience
and internal models.
Keywords: interoception; sensory integration; functional MRI; brain connectivity; rubber hand illusion
3
1. Introduction
Perception of one’s body relies on exteroception — the perception of the body from the outside — and
interoception, which, in broad definition, is understood as sensing and integrating all aspects of the body’s
physiological state (visceral afference) and motivational needs (Tsakiris & Critchley, 2016). Integrated
representation of the bodily state is increasingly recognized to underlie not only homeostatic control and
allostatic adaptation (Kleckner et al., 2017), but also emotional processing, social cognition and the sense
of self (Adolfi et al., 2017; Barrett, 2017; Critchley & Harrison, 2013; Tsakiris, 2017). Aberrant
interoceptive processing is associated with various neurological and psychiatric diseases, including
depression (Eggart, Lange, Binser, Queri, & Müller-Oerlinghausen, 2019; Wiebking et al., 2015;
Wiebking & Northoff, 2015), eating disorders (Young et al., 2017), functional neurological symptoms
(Ricciardi et al., 2016), somatoform disorders (Schaefer, Egloff, & Witthöft, 2012), stroke (Raimo et al.,
2020) and neurodegenerative syndromes (Marshall et al., 2017). Thus, description of the mechanisms
underlying sensory integration in interoception is of major importance.
The neural systems supporting interoceptive awareness have been extensively investigated after the
pioneer study of Critchley et al. in 2004 (Critchley, Wiens, Rotshtein, Öhman, & Dolan, 2004), with the
majority of data coming from the heartbeat detection task (cardioception) (Kleckner, Wormwood,
Simmons, Barrett, & Quigley, 2015). Although the general description of interoception-related areas has
been already provided (Schulz, 2016), the exact neural mechanisms underlying the integration of
interoceptive signals remain poorly studied. In line with the influential predictive coding concept, it is
supposed that top-down predictions regarding the body state are compared with the current bottom-up
sensory information (Apps & Tsakiris, 2014; Owens, Friston, Low, Mathias, & Critchley, 2018).
However, the neural basis of these top-down and bottom-up processes, to our knowledge, has not been
investigated directly.
At the same time, the integration of sensory signals in body perception is widely studied on the basis of
the rubber hand illusion (Ehrsson, 2019), which is a model of exteroceptive rather than interoceptive
processing. During the rubber hand illusion experiment, a realistic rubber hand is placed on the table in
front of the participant, while their real hand is hidden under the table. The experimenter induces the
illusion with multiple synchronous brushings of the participant’s real and artificial hands, which usually
result in the sense of ownership over the rubber hand. The intensity of the illusion is measured with
questionaries and with the perceived displacement of the participant’s own hand towards the rubber hand
(proprioceptive drift).
The development of the illusion depends on multisensory processing of bodily signals (Ehrsson, Holmes,
& Passingham, 2005). In accordance with the predictive coding model of self-perception (Apps &
Tsakiris, 2014; Owens et al., 2018), the strength of the illusion depends on the interplay between bottom-
up processing of incoming signals (visual and tactile) and top-down predictive coding based on
expectations (Reader & Crucianelli, 2019). That is, when the participant’s current sensory experience
indicates that the rubber hand is their own (bottom-up processing), they still know that if you place the
hand under the table and do not remove it, it is going to be there, and also that human beings do not own
rubber hands (top-down predictive coding). Recent findings indicate that the amygdala may serve as a
mediator between these bottom-up and top-down processes in the rubber hand illusion (Reader &
Crucianelli, 2019; Spengler, Scheele, Kaiser, Heinrichs, & Hurlemann, 2019).
Since predictive coding is thought to underlie both exteroception and interoception (Apps & Tsakiris,
2014; Owens et al., 2018), we proposed that the rubber hand illusion features may reflect general
tendencies in the bottom-up and top-down processing of bodily signals. In the current study, we utilized
both the rubber hand illusion experiment and the heartbeat detection task, with the latter being performed
in functional magnetic resonance imaging (fMRI) settings. We quantified individual differences in
bottom-up and top-down sensory processing with the rubber hand illusion experiment and then evaluated
how these differences are associated with brain connectivity patterns during cardioception.
2. Methods
2.1 Participants
Volunteers were enrolled from the control group of a governmentally funded project on vascular brain
injury conducted at the Research Center of Neurology (Moscow). We enrolled 30 female right-handed
4
non-physician healthcare workers aged 40–65 years, with no history of cardiovascular events, such as
stroke or myocardial infarction, and no severe white matter hyperintensities according to structural MRI
(modified Fazekas score of 2 or less; Fazekas, Chawluk, Alavi, Hurtig, & Zimmerman, 1987). The
sample size was determined by analogy with the previous studies utilizing fMRI-based heartbeat
detection task (Schulz, 2016). All participants underwent MRI scanning, but three of them did not
complete the rubber hand illusion experiment due to personal reasons (unwillingness to make a second
visit), leading to 27 full cases.
The study protocol, including inclusion/exclusion criteria, procedures and analysis plan, was pre-
approved by the Ethics Committee and the Institutional Review Board of the Research Center of
Neurology, and all participants gave informed consent for participation. We report how we determined
our sample size, all data exclusions, all inclusion/exclusion criteria, whether inclusion/exclusion criteria
were established prior to data analysis, all manipulations, and all measures in the study. The dataset for
this study can be found in OpenNeuro repository, reference number ds003763
(https://openneuro.org/datasets/ds003763).
2.2 Assessment of the sensory integration with rubber hand illusion
We used the rubber hand illusion in a previously tested modification of dynamic illusion (Perepelkina,
Vorobeva, Melnikova, Arina, & Nikolaeva, 2018). During the experiment, participants sat still at a table
with their right hand hidden in a wooden box (Fig. 1). An artificial rubber hand was placed medial to the
real right hand at a distance of 15 cm. The left hand was resting on the knees. The participants were
wearing a black cloth covering the artificial hand up to the forearm. During the onset of the illusion,
synchronous stimulation to the real and artificial hands was applied with two identical brushes, with a
mean frequency of 0.5 Hz; the touches were slightly irregular in duration and interval.
To evaluate the development of the illusion, we used a behavioral measure of perceived proprioceptive
displacement of the real hand under the illusion (proprioceptive drift), and a subjective measure (the
ownership questionnaire). To calculate proprioceptive drift, we evaluated the perceived position of the
right hand before the start of experiment (a value characterizing the error in hand localization, named
“mislocalization”), after 15, 30, 60, 120 and 240 seconds of stimulation (onset phase) and also at 15, 30,
60, 120 and 240 seconds after the end of stimulation (fading phase). The participants were asked to close
their eyes, put their left index finger on one of the three fixed points on the top of the box, and to make a
pointing movement, stopping the left index finger above the tip of the right middle finger. This procedure
was repeated three times in total with the sequence of points randomized, with subsequent averaging.
Proprioceptive drift was calculated as the difference between the indicated starting position and the
indicated current position of the right arm, with positive values for displacement towards the artificial
hand. The ownership questionnaire was administered at the same timepoints, only at the onset phase. It
consisted of 6 questions, including 3 test statements related to the experience of illusion (test subscale)
and 3 control statements, rated with a 7-point Likert scale (control subscale) (Perepelkina, Romanov,
Arina, Volel, & Nikolaeva, 2019).
Fig. 1. Experimental setting in the rubber hand illusion
5
To assess the validity of the rubber hand illusion experiment, we, first tested the dependence of
proprioceptive drift from time and phase with a mixed linear model. The following model was used:
Proprioceptive drift ~ time*condition + mislocalization + (1|Subject)
That is, the time from the phase onset and type of condition (onset or fading) served as predictors, initial
mislocalization was included as a covariate due to its known influence on illusion development (Longo,
Schüür, Kammers, Tsakiris, & Haggard, 2008), and individual subjects effects were considered as a
source of random variance.
A second evaluation of illusion validity was performed on the basis of the ownership questionnaire. In a
proportional odds logistic regression model, we used as a dependent variable the results of the ownership
questionnaire (two subscales: test and control), while predictors included the subscale type (test or
control) and time of stimulation (15, 30, 60, 120 or 240 s):
Ownership questionnaire ~ time*subscale
The modelling was performed with the use of R Project 3.6.0 (https://www.R-project.org/,
RRID:SCR_001905) packages “lme4”, “lmerTest” and “MASS”.
To reveal the brain networks related to multisensory integration during cardioception, we entered
proprioceptive drift at the end of the onset phase (240 s) into the fMRI connectivity analysis (see section
2.4.3 below) as a covariate. We did not include the ownership questionnaire results into the fMRI data
analysis due to the difficulties of testing the effect of an ordinal value in a relatively small sample and
also due to its low informative value in the current sample (see Section 3.1 below).
2.3 Functional MRI: the heartbeat detection task
In order to evaluate interoceptive processing, we used a variant of the heartbeat detection task, translated
into Russian, which is described in detail in our previous publications (O.R. Dobrushina et al., 2020; Olga
R. Dobrushina et al., 2020). The task was implemented in the fMRI setting. In brief, the heartbeat
detection task consisted of sixteen blocks alternating eight cardioceptive (Heart) and eight exteroceptive
(Sound) conditions. Each task included an attending phase (20 sec), a response phase (6 sec), and a rest
phase with a fixation cross (6 sec). The Heart/Sound phase was indicated on the screen by an ECG/note
symbol, and at the start of the response phase a finger on a button pictogram appeared nearby. During the
Heart condition, participants were asked to listen to their own heartbeat and, during the response phase, to
press a button each time they felt a heartbeat. During the Sound task, “beep” tones were presented at the
6
individual rate of heartbeats with a similar random variance. Before the task, the level of sound was
individually adjusted to the minimum that the participant was able to discriminate from the noise of the
scanner. The participants were instructed to attend to the tones and, during the response phase, to press a
button each time they heard a tone. Instructions were first presented as a video clip outside of the scanner,
and then presented inside the scanner. The presentation of the task and collection of responses was
performed with the use of the Cogent Matlab Toolbox (http://www.vislab.ucl.ac.uk/cogent_2000.php,
RRID:SCR_015672). Pulses were recorded with a pulse oximeter by the Siemens Physiological
Monitoring Unit, and TAPAS PhysIO Toolbox
(https://www.tnu.ethz.ch/en/software/tapas/documentations/physio-toolbox.html) was used for artifact
correction, peak detection, and the extraction of interbeat intervals.
Interoceptive accuracy, a widely used index (Brener & Ring, 2016), was calculated according to the
formula:
In order to evaluate the relations between multisensory integration and interoceptive accuracy (IAc), we
entered the latter into the models describing the rubber hand illusion development (see Section 2.2):
Proprioceptive drift ~ time*condition + mislocalization + IAc + (1|Subject)
Ownership questionnaire ~ time*question type + IAc
2.4 Functional MRI: data acquisition, pre-processing, activation analysis
MRI was performed with a Siemens MAGNETOM Verio 3T scanner (Erlangen, Germany) located at the
Research Center of Neurology. A three-dimensional structural image consisted of a sagittal T1-weighted
3D-MPRAGE sequence (TR 1900 msec, TE 2.47 msec, voxel size 1x1x1 mm, FOV 250 mm). The
magnetic field map further used for correction of images was obtained with a double-echo gradient field
map sequence. Functional images were acquired using Т2*-gradient echo imaging sequences (TR 2000
ms, TE 21 ms, voxel size 3×3×3 mm, FOV 192 mm). Four extra functional volumes were acquired at the
start of the session and discarded by the scanner software to prevent the usage of artifactual data obtained
before the magnetic equilibrium could be reached.
Analysis of the fMRI data was performed with the use of Statistical Parametric Mapping (SPM) 12
(www.fil.ion.ucl.ac.uk/spm, RRID:SCR_007037) and Conn 18b (http://www.nitrc.org/projects/conn,
RRID:SCR_009550) packages. The initial pre-processing in SPM included standard steps: slice-timing
correction, calculation of the voxel displacement map, realignment and unwrapping of the functional
images, co-registration of the structural and functional images, spatial normalization into standard
Montreal Neurological Institute (MNI) space, and spatial smoothing using a Gaussian kernel of 8 mm full
width at half maximum. Physiological noise resulting from central hemodynamics was estimated with the
use of the TAPAS PhysIO Toolbox on the basis of the RETROICOR algorithm (Glover, Li, & Ress,
2000; Kasper et al., 2017).
During the activation analysis, performed with SPM, the attending phases of the two conditions (Heart
and Sound) entered the model. Realignment parameters and physiological noise regressors were entered
as first-level covariates. Activation maps for the cardioceptive and exteroceptive conditions for each
participant were calculated with the use of the “Heart vs. Sound” and “Sound vs. Heart” contrasts. Due to
the high variation in performance in the heartbeat detection task, interoceptive accuracy (IAc) was
included into the group cardioceptive activation analysis as a covariate. (Brener & Ring, 2016)
2.5 Functional MRI: connectivity analysis
The connectivity analysis was performed on the basis of the attending phases of the cardioceptive and
exteroceptive conditions. After basic preprocessing in SPM (see Section 2.4.1), several additional steps
were performed in Conn. The realignment parameters, ART-based motion-scrubbing outliers and
physiological noise regressors were entered as first-level covariates. Signal from the white matter and the
cerebrospinal fluid volume estimated with the anatomical component-based analysis (aCompCor) was
7
regressed out. A temporal highpass frequency filter was applied to the data, restricting the analysis to
frequencies >0.008 Hz.
In order to reveal the circuits of sensory integration, we performed a seed-based connectivity analysis
(seed-to-voxel), using as regions of interest (ROI) the brain areas involved in the sensory processing of
bodily signals (Grivaz, Blanke, & Serino, 2017; Reader & Crucianelli, 2019; Spengler et al., 2019; Tyll,
Budinger, & Noesselt, 2011) — see Table 1. The ROI for the right and left intraparietal sulcus (rIPS,
lIPS), left anterior insula (laIns) and right ventral premotor cortex (rPMv) were constructed as 10 mm
spheres around the coordinates described in a meta-analysis by Grivaz et al. (Grivaz et al., 2017), while
the exact masks for the right and left amygdala (Reader & Crucianelli, 2019; Spengler et al., 2019) and
thalamus (Tyll et al., 2011) were taken from the HarvardOxford Atlas (Desikan et al., 2006), which is a
build-in Conn toolbox atlas. We also used as seed ROI the activation clusters from the cardioceptive and
exteroceptive conditions.
In the second-level analysis, the proprioceptive drift at the end of the rubber hand illusion onset phase —
a measure characterizing the balance between the bottom-up and top-down processes in sensory
integration — was entered as a covariate. The connectivity analysis was performed for the Heart vs. Rest,
Sound vs. Rest and Heart vs. Sound condition contrasts. Multiple comparison control was implemented
with family-wise error rate (FWE) at the cluster-level (additionally corrected for the number of ROI: p-
FWE = .05/10 = .005), given a cluster-defining voxel-wise statistical threshold of p<.001 uncorrected.
8
Table 1. Regions of Interest Used for the Connectivity Analysis of Sensory Integration
Regions of Interest (ROI)
MNI coordinates (x, y, z) of the
center of mass
Source
Intraparietal sulcus left
-38 -48 56
A 10 mm spheres around the
coordinates described in a meta-
analysis by Grivaz at al. (2017)
Intraparietal sulcus right
30 -56 52
Anterior insula left
-34 18 8
Ventral premotor cortex right
48 8 32
Amygdala Left
-23 -5 -18
Exact mask from the
HarvardOxford Atlas.
The major role of the amygdala
in sensory body processing is
outlined in a study by Spengler
et al. (2019).
Amygdala Right
23 -4 -18
Thalamus Left
-10 -19 6
Exact mask from the
HarvardOxford Atlas.
The role of the thalamus in detail
outlined by sensory body
processing is outlined by Tyll et
al. (2011)
Thalamus Right
11 18 7
Anterior insula right
36 18 -1
Activation cluster from the
cardioceptive task — see Fig. 2
below
Superior temporal gyrus right
62 -32 7
Activation cluster from the
exteroceptive task (sound
detection) — see Fig. 3 below
3. Results
3.1 Rubber hand illusion
According to the results of mixed linear modeling, the proposed dependence of the proprioceptive drift
from time (p = .02) and phase (onset or fading, p < .00001) was observed, with the effect of
mislocalization being significant (p = .0004). The interaction of time and condition also appeared to be
significant (p = .009). Evaluation of the ownership question results revealed a significant effect of
question type (test/control; participants were more likely to agree with the test statements; p < .0001),
with no significant dependence from time (p = 0.08) or any interaction of time and question type (p = .2).
Thus, we observed the classical sequence of onset and fading of the rubber hand illusion as measured by
proprioceptive drift, while the dynamics of the ownership questionnaire in the current sample were
nonsignificant. The IAc appeared to be unrelated to the dynamics of proprioceptive drift (p = .5) and the
ownership questionnaire (p = .08).
3.2 Functional MRI: activation during the heartbeat and sound detection tasks
During the heartbeat detection task, we observed activation in the right anterior insula: a cluster of 86
voxels with a peak at (36; 23; -10), cluster-level pFWE-corr = .041 (see Fig. 2). During the sound
detection task, we observed activation in the right superior temporal gyrus: a cluster of 262 voxels with a
peak at (69; -31; 8), pFWE-corr = .001 (see Fig. 3).
9
Fig. 2. Functional MRI activation related to the cardioceptive task
A cluster of 86 voxels is observed in the right dorsal anterior insula (peak at 36; 23; -10; cluster-level
pFWE-corr = .04).
Fig. 3. Functional MRI activation related to the exteroceptive task
A cluster of 262 voxels is observed in the right superior temporal gyrus (peak at 69; -31; 8; cluster-level
pFWE-corr = .001),
3.3 Functional MRI: connectivity related to the sensory integration processes
3.3.1 Connectivity related to sensory integration during cardioception
A seed-to-voxel analysis was performed to reveal the associations between proprioceptive drift and
connectivity of the areas involved in sensory processing (see Table 1 above for the list of analyzed ROIs).
In accordance with a previously developed conceptualization of sensory integration in the rubber hand
illusion (Reader & Crucianelli, 2019; Spengler et al., 2019), we interpreted a positive association with
drift as indicative of bottom-up processing and a negative association as indicative of top-down
processing. Therefore, we applied directional statistical tests and reviewed the positive and negative
associations separately below.
During the cardioceptive condition, we found significant effects for the left thalamus, right anterior insula
and left amygdala (see table 2). The left thalamus showed connectivity with clusters in the right inferior
frontal gyrus (IFG), anterior cingulate cortex (ACC), left frontal operculum extending to anterior insula,
hypothalamus and right planum polare, which was positively associated with drift, indicative of bottom-
up processing (see Fig. 4). Another finding — connectivity between the right anterior insula and the
hypothalamus — probably represented the same network related to bottom-up processing: the thalamus
10
connectivity map involved a minor area in the right anterior insula which did not enter the results due to
the low significance of effect (see Fig. 4). In contrast, the left amygdala showed connectivity with clusters
in the bilateral frontal poles, left frontal operculum extending to anterior insula and left amygdala (the
connectivity within the amygdala), which was negatively associated with drift and thus indicative of top-
down processing (see Fig. 5).
Table 2. Connectivity related to sensory integration during the heartbeat detection task
Seed area
Direction of
association
with drift
Cluster localization
MNI
coordinates (x,
y, z) of the
peak
Cluster
size, voxels
(2×2×2
mm3)
Cluster
size p-
FWE
Thalamus
Left
Positive
(bottom-up
processing)
Inferior Frontal Gyrus
(IFG) Right
+46 +32 -10
578
.000001
Anterior Cingulate Cortex
(ACC) Bilateral
-6 +52 +22
307
.0004
Frontal Operculum,
Anterior Insula Left
-36 +22 -2
260
.0001
Hypothalamus
+4 -4 -8
248
.002
Planum Polare Right
+48 +8 -14
211
.005
Anterior
Insula Right
Positive
(bottom-up
processing)
Hypothalamus
+12 -2 +4
284
.0005
Amygdala
Left
Negative
(top-down
processing)
Frontal Pole Right
+34 +56 +20
509
.000001
Frontal Pole Left
-42 +50 +14
426
.00001
Frontal Operculum,
Anterior Insula Left
-36 +28 +2
260
.0006
Amygdala Left
-12 -4 -14
198
.003
11
Fig. 4. Connectivity of the left thalamus related to sensory integration during the cardioceptive task
The connectivity map is thresholded at p-FWE <.005 at the cluster-level with cluster-defining voxel-wise
statistical threshold of p<.001 uncorrected. Red color corresponds to a positive association with drift,
which is indicative of bottom-up processing.
12
Fig. 5. Connectivity of the left amygdala related to sensory integration during the cardioceptive task
The connectivity map is thresholded at p-FWE <.005 at the cluster-level with cluster-defining voxel-wise
statistical threshold of p<.001 uncorrected. Blue color corresponds to a negative association with drift,
which is indicative of top-down processing.
3.3.2 Connectivity related to sensory integration during exteroception
The seed-to-voxel analysis for the associations between proprioceptive drift and connectivity of the areas
involved in sensory processing during the sound detection task revealed significant effects for the left
thalamus and left anterior insula (see Table 3; the list of analyzed ROIs is described in Table 1).
Connectivity of the left thalamus with several regions (the right superior temporal gyrus extending to
Heschl's gyrus, angular gyrus, supramarginal gyrus and also the hypothalamus) was positively associated
with drift — bottom-up processing (see Fig. 6). At the same time, the left anterior insula showed
connectivity with a cluster in the left amygdala, which was negatively associated with drift — top-down
processing (see Fig. 7).
Table 3. Connectivity related to sensory integration during the sound detection task
Seed area
Direction of
association
with drift
Cluster localization
MNI
coordinates (x,
y, z) of the
peak
Cluster
size, voxels
(2×2×2
mm3)
Cluster
size p-
FWE
Thalamus
Left
Positive
(bottom-up
processing)
Superior Temporal Gyrus,
Heschl's Gyrus, Angular
Gyrus, Supramarginal Gyrus
Right
+66 -40 +20
565
.000001
Hypothalamus
-6 +4 -10
201
.004
13
Anterior
Insula left
Negative
(top-down
processing)
Left Amygdala
-32 +2 -26
235
.001
Fig. 6. Connectivity of the left thalamus related to sensory integration during the sound detection task
The connectivity map is thresholded at p-FWE <.005 at the cluster-level, cluster-defining voxel-wise
statistical threshold of p<.001 uncorrected. Yellow color corresponds to a positive association with drift,
which is indicative of bottom-up processing.
Fig. 7. Connectivity of the left anterior insula related to sensory integration during the sound detection
task
The connectivity map is thresholded at p-FWE <.005 at the cluster-level, cluster-defining voxel-wise
statistical threshold of p<.001 uncorrected. Cyan color corresponds to a negative association with drift,
which is indicative of top-down processing. The left amygdala is seen on the slice.
3.3.3 Comparison of the connectivity related to sensory integration during the cardioceptive and
exteroceptive tasks
Comparing the connectivity related to sensory integration in the heartbeat and sound detection tasks
(Heart vs. Sound or vice versa, effect of proprioceptive drift), we found no significant differences. To
evaluate the commonalities between the two tasks, we first manually analyzed the overlapping images of
the sensory integration related connectivity, and then performed a conjunction analysis.
An overlap map of the connectivity of the left thalamus related to sensory integration in the cardioceptive
and exteroceptive conditions is presented in Fig. 8 (a liberal statistical threshold is used: p-FWE <.05 at
the cluster-level, cluster-defining voxel-wise statistical threshold of p<.005 uncorrected). Conjunction
14
analysis (see Table 3) showed significant effects for a large cluster in the right superior temporal gyrus
extending to the planum polare and temporal operculum and another cluster in the hypothalamus. In
addition, on the overlap image there is a common area in the right IFG (see Fig. 8).
Fig. 8. Connectivity of the left thalamus related to sensory integration: overlap between the cardioceptive
and exteroceptive conditions
The connectivity map is thresholded at p-FWE <.05 at the cluster-level, cluster-defining voxel-wise
statistical threshold of p<.005 uncorrected. Red color corresponds to the Heart condition, and yellow to
the Sound condition. Both red and yellow colors correspond to a positive association with drift, which is
indicative of bottom-up processing.
Table 3. Connectivity related to sensory integration: conjunction between the cardioceptive and
exteroceptive conditions
Seed area
Direction of
association
with drift
Cluster
localization
MNI
coordinates (x,
y, z) of the
peak
Cluster
size, voxels
(2×2×2
mm3)
Cluster
size p-
FWE
Thalamus
Left
Positive
(bottom-up
processing)
Superior
Temporal
Gyrus,
Temporal
Operculum,
Planum Polare
Right
66 -44 4
540
<.001
Hypothalamus
0 4 -10
253
.006
Amygdala
Left
Negative
(top-down
processing)
Amygdala Left
-12 -4 -6
225
.008
Planum Polare,
Posterior Insula
Right
42 -4 -14
201
.01
15
An overlap map of the connectivity of the left amygdala related to sensory integration in the cardioceptive
and exteroceptive conditions is presented in Fig. 9 (a liberal statistical threshold is used: p-FWE <.05 at
the cluster-level, cluster-defining voxel-wise statistical threshold of p<.005 uncorrected). Conjunction
analysis (see Table 3) showed a significant effect for two areas: a cluster around the left amygdala and a
cluster in the right planum polare extending to the posterior insula (p-FWE <.05 at the cluster-level,
cluster-defining voxel-wise statistical threshold of p<.001 uncorrected). At the same time, on the overlap
image we also see common areas in the bilateral frontal poles and in the areas involving the left frontal
operculum and anterior insula (see Fig. 9).
Fig. 9. Connectivity of the left amygdala related to sensory integration: overlap between the cardioceptive
and exteroceptive conditions
The connectivity map is thresholded at p-FWE <.05 at the cluster-level, cluster-defining voxel-wise
statistical threshold of p<.005 uncorrected. Blue color corresponds to the Heart condition, and cyan to the
Sound condition. Both blue and cyan colors correspond to a negative association with drift, which is
indicative of top-down processing.
Analyzing the commonalities in the connectivity of the right anterior insula related to sensory integration
in the interoceptive and exteroceptive conditions, we found a tendency for overlap in the bilateral
thalamus. However, this finding was not proven by the conjunction analysis.
4. Discussion
In the current study we utilized two classical models of bodily perception: the rubber hand illusion (exter-
nal body) and the heartbeat detection task (internal body). We found no association between interoceptive
accuracy and rubber hand illusion dynamics. The existing data are controversial: Tsakiris et al. found an
inverse correlation between accuracy in a heartbeat counting task and proprioceptive drift in a classical
rubber hand illusion (Tsakiris, Tajadura-Jiménez, & Costantini, 2011), Suzuki et al. found a direct corre-
lation between accuracy in a heartbeat discrimination task and proprioceptive drift in an original virtual
16
reality hand illusion with cardiac feedback (Suzuki, Garfinkel, Critchley, & Seth, 2013) and Horváth et al.
found no association between accuracy in a heartbeat counting task and angular proprioceptive drift
(Horváth et al., 2020). Such a variety of findings may suggest some mediational factors which remain to
be revealed.
Proprioceptive drift appeared to be the most informative value characterizing the development of the
rubber hand illusion in our sample, and we suggested that it may characterize important individual
differences in body-related sensory processing. On the basis of previous studies of body perception and in
line with the predictive coding concept, we proposed that higher proprioceptive drift indicates a tendency
to rely on bottom-up signal processing, while lower proprioceptive drift may be related to the dominance
of critical top-down influences (Apps & Tsakiris, 2014; Owens et al., 2018; Reader & Crucianelli, 2019).
Indeed, higher proprioceptive drift appeared to be associated with enhanced connectivity of the thalamus
with the left frontal operculum and anterior insula, ACC, hypothalamus, right planum polare and right
IFG. The thalamus, first, is a relay point for the visceral afferents on their way to the anterior insula
(Craig, 2009; Saper, 2002), the area responsible for interoceptive awareness (Critchley et al., 2004), and,
thus, thalamic-insular connectivity represents a bottom-up pathway. Second, the thalamus participates in
the integration of sensory information, presumably coordinating distant cortical areas by means of
thalamo-cortical feed-forward connections (Tyll et al., 2011). The integratory role of the thalamus
explains its connectivity with other revealed areas. The anterior cingulate cortex and hypothalamus are
functionally coupled with the anterior insula within the salience network (Seeley et al., 2007) and are also
involved in the processing of visceral signals (Ceunen, Vlaeyen, & Van Diest, 2016). The planum
temporale participates in sound source location (Battal, Rezk, Mattioni, Vadlamudi, & Collignon, 2019),
which is highly relevant to the cardioceptive task — the heartbeat can be both felt and listened to (Hall,
Lopes, & Yu, 2019). The revealed area in the right IFG (see Fig. 5), according to the parcellation by
Hartwigsen et al., involves the right anterior part of the IFG, which is responsible for social-cognitive and
emotional processing (Hartwigsen, Neef, Camilleri, Margulies, & Eickhoff, 2019). More specifically, it is
involved in filtering out irrelevant information (interference suppression) (Vaidya et al., 2005; Zhao et al.,
2017) and self-face recognition (van Veluw & Chance, 2014). We propose that during the heartbeat
detection task, the right IFG might have a similar regulatory role, such as filtering out the sensory
information irrelevant to the heartbeat and recognition of information related the one’s own heart. Thus,
we hypothesize that the revealed network, coordinated by the thalamus, is responsible for bottom-up
integration of interoceptive information.
Another network, coordinated by the amygdala and presumably associated with top-down processing of
interoceptive information (negative association with proprioceptive drift), included large areas in the
bilateral frontal poles (rostral prefrontal cortex) and in the left frontal operculum extending to the anterior
insula. These areas are known to be functionally linked within the salience network (Seeley et al., 2007).
The revealed role of the amygdala was highly consistent with the results of Spengler et al., who observed
a dramatic increase in the speed of the rubber hand illusion formation in patients with amygdala damage
and also after the suppression of the amygdala with intranasal oxytocin (Spengler et al., 2019). Based on
the revealed dependency, the authors proposed that a protective mechanism against the distortion of body
schema is amygdala-dependent. The data of our study supports a more general conclusion: the amygdala
may mediate the top-down, model-based (vs. stimuli-based) processing of bodily signals. This role
converges with the known role of amygdala activation in hypervigilance (LeDoux, 2007), since epistemic
vigilance interferes with openness to incoming information (Sperber et al., 2010).
Within the revealed top-down processing network, the amygdala appeared to be functionally linked with
the lateral part of the rostral prefrontal cortex (see Fig. 5), which is responsible for stimulus-independent
attending. Stimulus-independent attending (cf. mental attention or James’s intellectual attention) is mental
processing based on self-generated thought rather than on the current sensory input, in contrast with
stimulus-oriented attending (Burgess, Gilbert, & Dumontheil, 2007). The revealed negative association of
the connectivity between the amygdala and the lateral rostral prefrontal cortex with proprioceptive drift is
consistent with the core assumption underlying our study — the interpretation of low illusion strength as a
dominance of top-down processing. It is likely that people who tend not to develop the illusion rely on the
stimulus-independent strategy of bodily information processing: the perception is dominated by top-down
expectations, based on internal models, rather than by the current sensory experience.
17
Interestingly, the same area in the left frontal operculum extending to the anterior insula was involved in
both bottom-up and top-down cardioceptive processing (see Figs. 4 and 5), suggesting its mediatory role.
The frontal opercular and anterior insulae are frequently coactivated, and this area is proposed to play a
key role in interoceptive awareness (Caseras et al., 2013; Critchley et al., 2004; Zaki, Davis, & Ochsner,
2012). Accumulating evidence supports a more general role for this area — selective attention towards the
relevant stimuli (Cai, Ryali, Chen, Li, & Menon, 2014; Higo, Mars, Boorman, Buch, & Rushworth, 2011;
Uddin, Nomi, Hébert-Seropian, Ghaziri, & Boucher, 2017; Varjačić et al., 2018). We propose that the
regulatory influence from the rostral prefrontal cortex to the operculo-insular cortex may serve to inhibit
attention towards the actual interoceptive input, suppressing the bottom-up signal processing in favor of
the top-down.
The revealed networks related to bottom-up and top-down processing showed similarity across the
heartbeat detection and sound detection tasks, which is probably due to the sensory integration processes
common to the two tasks. During the experiment, ever-present interoceptive flow coexisted with
exteroceptive information, such as scanner sounds and visual presentation, leading to a need for
multisensory processing — consistent with the continuous multimodal input presented in real life. The
described bottom-up and top-down processes represent higher-order sensory integration, which is likely
to be amodal. The heartbeat detection task involves attributing/not attributing the incoming sensory
information to a multisensory (amodal) image of the heart, which is built on the basis of the previous
sensory experience (visceral afference, sound) and socially governed models (visual image, spatial
localization). These processes are known to underlie perception in general, including interoception,
perception of the external body (such as the rubber hand illusion) and perception of environmental stimuli
(Briscoe, 2016). From the other point of view, the integration of interoceptive and exteroceptive
information is functionally coupled within the Bayesian systems supporting allostasis and self-awareness
(Quigley, Kanoski, Grill, Barrett, & Tsakiris, 2021). The brain is thought to constantly generate holistic,
multimodal predictions and to test them using all incoming information, both exteroceptive and
interoceptive (Barrett, 2017). Thus, it is not surprising that the neural systems supporting such processes
have modality-universal chains.
The study has limitations. First, we used a single experimental model of interoception — the heartbeat
detection task. The current field of human interoception research mostly relies on cardioception studies; at
the same time, there is controversy regarding the possibility of unitary view on interoception across
modalities, with ambivalent evidence (Garfinkel et al., 2016; Herbert, Muth, Pollatos, & Herbert, 2012;
Murphy, Catmur, & Bird, 2018; Pollatos, Herbert, Mai, & Kammer, 2016; Whitehead & Drescher, 1980).
On the neuroanatomical level, there is a high convergence between the pathways implementing different
interoceptive modalities, with the central role being played by the insular cortex (Craig, 2009; Saper,
2002). We propose that the study findings may have implications for interoception in general rather that
only for cardioception, since the bottom-up and top-down processes of predictive coding are higher-order,
apparently amodal. This hypothesis needs to be tested in subsequent research. As second limitation is that
in order to avoid excessive heterogeneity of the sample we included only females.
5. Conclusion
This study uncovers the neural systems supporting the bottom-up and top-down processing of bodily
signals. Bottom-up, stimulus-dependent processing was found to involve the thalamus, left frontal
operculum and anterior insula, ACC, hypothalamus, right planum polare and right IFG. Top-down,
model-dependent processing was supported by the rostral prefrontal cortex and the same area involving
the left frontal opercular and anterior insular cortex. These findings are in line with the well-grounded
predictive coding concept and with accumulated data about the functional roles of the revealed areas. The
highlighted interplay between bottom-up and top-down processing reflects the interaction between the
current sensory experience and socially governed internal models, such as body schema, during
perception. This interaction may allow perception to reach the amodal level, where incoming information
is linked to universal meanings and the subjective bodily reality is constructed.
6. Declarations of interest: none.
18
References
Adolfi, F., Couto, B., Richter, F., Decety, J., Lopez, J., Sigman, M., … Ibáñez, A. (2017). Convergence of
interoception, emotion, and social cognition: A twofold fMRI meta-analysis and lesion approach.
Cortex, 88, 124–142. https://doi.org/10.1016/j.cortex.2016.12.019
Apps, M. A. J., & Tsakiris, M. (2014). The free-energy self: A predictive coding account of self-
recognition. Neuroscience and Biobehavioral Reviews, Vol. 41, pp. 85–97.
https://doi.org/10.1016/j.neubiorev.2013.01.029
Barrett, L. F. (2017). The theory of constructed emotion: an active inference account of interoception and
categorization. Social Cognitive and Affective Neuroscience, 12(1), 1–23.
https://doi.org/10.1093/scan/nsw154
Battal, C., Rezk, M., Mattioni, S., Vadlamudi, J., & Collignon, O. (2019). Representation of Auditory
Motion Directions and Sound Source Locations in the Human Planum Temporale. The Journal of
Neuroscience, 39(12), 2208–2220. https://doi.org/10.1523/JNEUROSCI.2289-18.2018
Brener, J., & Ring, C. (2016). Towards a psychophysics of interoceptive processes: the measurement of
heartbeat detection. Philosophical Transactions of the Royal Society of London. Series B, Biological
Sciences, 371(1708). https://doi.org/10.1098/rstb.2016.0015
Briscoe, R. E. (2016). Multisensory Processing and Perceptual Consciousness: Part I. Philosophy
Compass, 11(2), 121–133. https://doi.org/10.1111/phc3.12227
Burgess, P. W., Gilbert, S. J., & Dumontheil, I. (2007). Function and localization within rostral prefrontal
cortex (area 10). Philosophical Transactions of the Royal Society B: Biological Sciences, 362(1481),
887–899. https://doi.org/10.1098/rstb.2007.2095
Cai, W., Ryali, S., Chen, T., Li, C. S. R., & Menon, V. (2014). Dissociable roles of right inferior frontal
cortex and anterior insula in inhibitory control: Evidence from intrinsic and task-related functional
parcellation, Connectivity, And response profile analyses across multiple datasets. Journal of
Neuroscience, 34(44), 14652–14667. https://doi.org/10.1523/JNEUROSCI.3048-14.2014
Caseras, X., Murphy, K., Mataix-Cols, D., López-Solà, M., Soriano-Mas, C., Ortriz, H., … Torrubia, R.
(2013). Anatomical and functional overlap within the insula and anterior cingulate cortex during
interoception and phobic symptom provocation. Human Brain Mapping, 34(5), 1220–1229.
https://doi.org/10.1002/hbm.21503
Ceunen, E., Vlaeyen, J. W. S., & Van Diest, I. (2016). On the origin of interoception. Frontiers in
Psychology, Vol. 7. https://doi.org/10.3389/fpsyg.2016.00743
Craig, A. D. (2009). How do you feel — now? The anterior insula and human awareness. Nature Reviews
Neuroscience, 10(1), 59–70. https://doi.org/10.1038/nrn2555
Critchley, H. D., & Harrison, N. A. (2013, February 20). Visceral Influences on Brain and Behavior.
Neuron, Vol. 77, pp. 624–638. https://doi.org/10.1016/j.neuron.2013.02.008
Critchley, H. D., Wiens, S., Rotshtein, P., Öhman, A., & Dolan, R. J. (2004). Neural systems supporting
interoceptive awareness. Nature Neuroscience, 7(2), 189–195. https://doi.org/10.1038/nn1176
Desikan, R. S., Ségonne, F., Fischl, B., Quinn, B. T., Dickerson, B. C., Blacker, D., … Killiany, R. J.
(2006). An automated labeling system for subdividing the human cerebral cortex on MRI scans into
gyral based regions of interest. NeuroImage, 31(3), 968–980.
https://doi.org/10.1016/j.neuroimage.2006.01.021
Dobrushina, O.R., Dobrynina, L. A., Arina, G. A., Kremneva, E. I., Suslina, A. D., Gubanova, M. V., …
Krotenkova, M. V. (2020). [The interrelation between interoception and emotional intelligence: a
functional neuroimaging study]. Zh Vyssh Nerv Deiat I P Pavlova, 70(2), 206–216.
https://doi.org/10.31857/s0044467720020069
Dobrushina, Olga R., Arina, G. A., Dobrynina, L. A., Suslina, A. D., Solodchik, P. O., Belopasova, A. V.,
… Krotenkova, M. V. (2020). The ability to understand emotions is associated with interoception‐
19
related insular activation and white matter integrity during aging. Psychophysiology, 57(3), e13537.
https://doi.org/10.1111/psyp.13537
Eggart, M., Lange, A., Binser, M. J., Queri, S., & Müller-Oerlinghausen, B. (2019). Major Depressive
Disorder Is Associated with Impaired Interoceptive Accuracy: A Systematic Review. Brain
Sciences, 9(6), 131. https://doi.org/10.3390/brainsci9060131
Ehrsson, H. H. (2019). Multisensory processes in body ownership. In Multisensory Perception: From
Laboratory to Clinic (pp. 179–200). https://doi.org/10.1016/B978-0-12-812492-5.00008-5
Ehrsson, H. H., Holmes, N. P., & Passingham, R. E. (2005). Touching a rubber hand: Feeling of body
ownership is associated with activity in multisensory brain areas. Journal of Neuroscience, 25(45),
10564–10573. https://doi.org/10.1523/JNEUROSCI.0800-05.2005
Fazekas, F., Chawluk, J., Alavi, A., Hurtig, H., & Zimmerman, R. (1987). MR signal abnormalities at 1.5
T in Alzheimer’s dementia and normal aging. American Journal of Roentgenology, 149(2), 351–
356. https://doi.org/10.2214/ajr.149.2.351
Garfinkel, S. N., Manassei, M. F., Hamilton-Fletcher, G., In den Bosch, Y., Critchley, H. D., & Engels,
M. (2016). Interoceptive dimensions across cardiac and respiratory axes. Philosophical
Transactions of the Royal Society B: Biological Sciences, 371(1708), 20160014.
https://doi.org/10.1098/rstb.2016.0014
Glover, G. H., Li, T. Q., & Ress, D. (2000). Image-based method for retrospective correction of
physiological motion effects in fMRI: RETROICOR. Magnetic Resonance in Medicine, 44(1), 162–
167.
Grivaz, P., Blanke, O., & Serino, A. (2017). Common and distinct brain regions processing multisensory
bodily signals for peripersonal space and body ownership. NeuroImage, 147, 602–618.
https://doi.org/10.1016/j.neuroimage.2016.12.052
Hall, J. K., Lopes, B., & Yu, H. (2019). Interoception Enhanced via the Ears? Journal of
Psychophysiology, 33(3), 165–170. https://doi.org/10.1027/0269-8803/a000219
Hartwigsen, G., Neef, N. E., Camilleri, J. A., Margulies, D. S., & Eickhoff, S. B. (2019). Functional
Segregation of the Right Inferior Frontal Gyrus: Evidence From Coactivation-Based Parcellation.
Cerebral Cortex, 29(4), 1532–1546. https://doi.org/10.1093/cercor/bhy049
Herbert, B. M., Muth, E. R., Pollatos, O., & Herbert, C. (2012). Interoception across Modalities: On the
Relationship between Cardiac Awareness and the Sensitivity for Gastric Functions. PLoS ONE,
7(5), e36646. https://doi.org/10.1037/xge0000366
Higo, T., Mars, R. B., Boorman, E. D., Buch, E. R., & Rushworth, M. F. S. (2011). Distributed and causal
influence of frontal operculum in task control. Proceedings of the National Academy of Sciences of
the United States of America, 108(10), 4230–4235. https://doi.org/10.1073/pnas.1013361108
Horváth, Á., Ferentzi, E., Bogdány, T., Szolcsányi, T., Witthöft, M., & Köteles, F. (2020). Proprioception
but not cardiac interoception is related to the rubber hand illusion. Cortex, 132, 361–373.
https://doi.org/10.1016/j.cortex.2020.08.026
Kasper, L., Bollmann, S., Diaconescu, A. O., Hutton, C., Heinzle, J., Iglesias, S., … Stephan, K. E.
(2017). The PhysIO Toolbox for Modeling Physiological Noise in fMRI Data. Journal of
Neuroscience Methods, 276, 56–72. https://doi.org/10.1016/J.JNEUMETH.2016.10.019
Kleckner, I. R., Wormwood, J. B., Simmons, W. K., Barrett, L. F., & Quigley, K. S. (2015).
Methodological recommendations for a heartbeat detection-based measure of interoceptive
sensitivity. Psychophysiology, 52(11), 1432–1440. https://doi.org/10.1111/psyp.12503
Kleckner, I. R., Zhang, J., Touroutoglou, A., Chanes, L., Xia, C., Simmons, W. K., … Barrett, L. F.
(2017). Evidence for a Large-Scale Brain System Supporting Allostasis and Interoception in
Humans. Nature Human Behaviour, 1. https://doi.org/10.1038/s41562-017-0069
LeDoux, J. (2007). The amygdala. Current Biology, 17(20), R868–R874.
20
https://doi.org/10.1016/j.cub.2007.08.005
Longo, M. R., Schüür, F., Kammers, M. P. M., Tsakiris, M., & Haggard, P. (2008). What is embodiment?
A psychometric approach. Cognition, 107(3), 978–998.
https://doi.org/10.1016/j.cognition.2007.12.004
Marshall, C. R., Hardy, C. J. D., Russell, L. L., Clark, C. N., Dick, K. M., Brotherhood, E. V., … Warren,
J. D. (2017). Impaired Interoceptive Accuracy in Semantic Variant Primary Progressive Aphasia.
Frontiers in Neurology, 8. https://doi.org/10.3389/fneur.2017.00610
Murphy, J., Catmur, C., & Bird, G. (2018). Alexithymia is associated with a multidomain,
multidimensional failure of interoception: Evidence from novel tests. Journal of Experimental
Psychology: General, 147(3), 398–408. https://doi.org/10.1037/xge0000366
Owens, A. P., Friston, K. J., Low, D. A., Mathias, C. J., & Critchley, H. D. (2018). Investigating the
relationship between cardiac interoception and autonomic cardiac control using a predictive coding
framework. Autonomic Neuroscience, 210, 65–71. https://doi.org/10.1016/j.autneu.2018.01.001
Perepelkina, O., Romanov, D., Arina, G., Volel, B., & Nikolaeva, V. (2019). Multisensory mechanisms of
body perception in somatoform disorders. Journal of Psychosomatic Research, 127, 109837.
https://doi.org/10.1016/j.jpsychores.2019.109837
Perepelkina, O., Vorobeva, V., Melnikova, O., Arina, G., & Nikolaeva, V. (2018). Artificial hand
illusions dynamics: Onset and fading of static rubber and virtual moving hand illusions.
Consciousness and Cognition, 65, 216–227. https://doi.org/10.1016/j.concog.2018.09.005
Pollatos, O., Herbert, B. M., Mai, S., & Kammer, T. (2016). Changes in interoceptive processes following
brain stimulation. Philosophical Transactions of the Royal Society B: Biological Sciences,
371(1708), 20160016. https://doi.org/10.1098/rstb.2016.0016
Quigley, K. S., Kanoski, S., Grill, W. M., Barrett, L. F., & Tsakiris, M. (2021). Functions of
Interoception: From Energy Regulation to Experience of the Self. Trends in Neurosciences, 44(1),
29–38. https://doi.org/10.1016/j.tins.2020.09.008
Raimo, S., Boccia, M., Di Vita, A., Iona, T., Cropano, M., Ammendolia, A., … Palermo, L. (2020).
Interoceptive awareness in focal brain-damaged patients. Neurological Sciences, 41(6), 1627–1631.
https://doi.org/10.1007/s10072-019-04172-z
Reader, A. T., & Crucianelli, L. (2019, September 25). A multisensory perspective on the role of the
amygdala in body ownership. Journal of Neuroscience, Vol. 39, pp. 7645–7647.
https://doi.org/10.1523/JNEUROSCI.0971-19.2019
Ricciardi, L., Demartini, B., Crucianelli, L., Krahé, C., Edwards, M. J., & Fotopoulou, A. (2016).
Interoceptive awareness in patients with functional neurological symptoms. Biological Psychology,
113, 68–74. https://doi.org/10.1016/j.biopsycho.2015.10.009
Saper, C. B. (2002). The Central Autonomic Nervous System: Conscious Visceral Perception and
Autonomic Pattern Generation. Annual Review of Neuroscience, 25(1), 433–469.
https://doi.org/10.1146/annurev.neuro.25.032502.111311
Schaefer, M., Egloff, B., & Witthöft, M. (2012). Is interoceptive awareness really altered in somatoform
disorders? Testing competing theories with two paradigms of heartbeat perception. Journal of
Abnormal Psychology, 121(3), 719–724. https://doi.org/10.1037/a0028509
Schulz, S. M. (2016). Neural correlates of heart-focused interoception: a functional magnetic resonance
imaging meta-analysis. Philosophical Transactions of the Royal Society B: Biological Sciences,
371(1708), 20160018. https://doi.org/10.1098/rstb.2016.0018
Seeley, W. W., Menon, V., Schatzberg, A. F., Keller, J., Glover, G. H., Kenna, H., … Greicius, M. D.
(2007). Dissociable intrinsic connectivity networks for salience processing and executive control.
The Journal of Neuroscience : The Official Journal of the Society for Neuroscience, 27(9), 2349–
2356. https://doi.org/10.1523/JNEUROSCI.5587-06.2007
21
Spengler, F. B., Scheele, D., Kaiser, S., Heinrichs, M., & Hurlemann, R. (2019). A protective mechanism
against illusory perceptions is amygdala-dependent. Journal of Neuroscience, 39(17), 3301–3308.
https://doi.org/10.1523/JNEUROSCI.2577-18.2019
Sperber, D., Clément, F., Heintz, C., Mascaro, O., Mercier, H., Origgi, G., & Wilson, D. (2010).
Epistemic Vigilance. Mind & Language, 25(4), 359–393. https://doi.org/10.1111/j.1468-
0017.2010.01394.x
Suzuki, K., Garfinkel, S. N., Critchley, H. D., & Seth, A. K. (2013). Multisensory integration across
exteroceptive and interoceptive domains modulates self-experience in the rubber-hand illusion.
Neuropsychologia, 51(13), 2909–2917. https://doi.org/10.1016/j.neuropsychologia.2013.08.014
Tsakiris, M. (2017). The multisensory basis of the self: From body to identity to others. Quarterly
Journal of Experimental Psychology, 70(4), 597–609.
https://doi.org/10.1080/17470218.2016.1181768
Tsakiris, M., & Critchley, H. (2016). Interoception beyond homeostasis: affect, cognition and mental
health. Philosophical Transactions of the Royal Society B: Biological Sciences, 371(1708),
20160002. https://doi.org/10.1098/rstb.2016.0002
Tsakiris, M., Tajadura-Jiménez, A., & Costantini, M. (2011). Just a heartbeat away from one’s
body:Interoceptive sensitivity predicts malleability of body-representations. Proceedings of the
Royal Society B: Biological Sciences, 278(1717), 2470–2476.
https://doi.org/10.1098/rspb.2010.2547
Tyll, S., Budinger, E., & Noesselt, T. (2011). Thalamic Influences on Multisensory Integration.
Communicative & Integrative Biology, 4(4). https://doi.org/10.4161/CIB.4.4.15222
Uddin, L. Q., Nomi, J. S., Hébert-Seropian, B., Ghaziri, J., & Boucher, O. (2017). Structure and Function
of the Human Insula. Journal of Clinical Neurophysiology : Official Publication of the American
Electroencephalographic Society, 34(4), 300–306.
https://doi.org/10.1097/WNP.0000000000000377
Vaidya, C. J., Bunge, S. A., Dudukovic, N. M., Zalecki, C. A., Elliott, G. R., & Gabrieli, J. D. E. (2005).
Altered Neural Substrates of Cognitive Control in Childhood ADHD: Evidence From Functional
Magnetic Resonance Imaging. American Journal of Psychiatry, 162(9), 1605–1613.
https://doi.org/10.1176/appi.ajp.162.9.1605
van Veluw, S. J., & Chance, S. A. (2014). Differentiating between self and others: an ALE meta-analysis
of fMRI studies of self-recognition and theory of mind. Brain Imaging and Behavior, 8(1), 24–38.
https://doi.org/10.1007/s11682-013-9266-8
Varjačić, A., Mantini, D., Levenstein, J., Slavkova, E. D., Demeyere, N., & Gillebert, C. R. (2018). The
role of left insula in executive set-switching: Lesion evidence from an acute stroke cohort. Cortex,
107, 92–101. https://doi.org/10.1016/j.cortex.2017.11.009
Whitehead, W. E., & Drescher, V. M. (1980). Perception of Gastric Contractions and Self-Control of
Gastric Motility. Psychophysiology, 17(6), 552–558. https://doi.org/10.1111/j.1469-
8986.1980.tb02296.x
Wiebking, C., de Greck, M., Duncan, N. W., Tempelmann, C., Bajbouj, M., & Northoff, G. (2015).
Interoception in insula subregions as a possible state marker for depression — an exploratory fMRI
study investigating healthy, depressed and remitted participants. Frontiers in Behavioral
Neuroscience, 9, 82. https://doi.org/10.3389/fnbeh.2015.00082
Wiebking, C., & Northoff, G. (2015). Neural activity during interoceptive awareness and its associations
with alexithymia-An fMRI study in major depressive disorder and non-psychiatric controls.
Frontiers in Psychology, 6, 589. https://doi.org/10.3389/fpsyg.2015.00589
Young, H. A., Williams, C., Pink, A. E., Freegard, G., Owens, A., & Benton, D. (2017). Getting to the
heart of the matter: Does aberrant interoceptive processing contribute towards emotional eating?
PloS One, 12(10), e0186312. https://doi.org/10.1371/journal.pone.0186312
22
Zaki, J., Davis, J. I., & Ochsner, K. N. (2012). Overlapping activity in anterior insula during interoception
and emotional experience. NeuroImage, 62(1), 493–499.
https://doi.org/10.1016/j.neuroimage.2012.05.012
Zhao, H., Qiao, L., Fan, D., Zhang, S., Turel, O., Li, Y., … He, Q. (2017). Modulation of Brain Activity
with Noninvasive Transcranial Direct Current Stimulation (tDCS): Clinical Applications and Safety
Concerns. Frontiers in Psychology, 8, 685. https://doi.org/10.3389/fpsyg.2017.00685
