Heartfelt imitation: High interoceptive awareness is linked to greater automatic imitation

Abstract

The Silences of the Archives, the Reknown of the Story.
The Martin Guerre affair has been told many times since Jean de Coras and Guillaume Lesueur published their stories in 1561. It is in many ways a perfect intrigue with uncanny resemblance, persuasive deception and a surprizing end when the two Martin stood face to face, memory to memory, before captivated judges and a guilty feeling Bertrande de Rols. The historian wanted to go beyond the known story in order to discover the world of the heroes. This research led to disappointments and surprizes as documents were discovered concerning the environment of Artigat’s inhabitants and bearing directly on the main characters thanks to notarial contracts. Along the way, study of the works of Coras and Lesueur took a new direction. Coming back to the affair a quarter century later did not result in finding new documents (some are perhaps still buried in Spanish archives), but by going back over her tracks, the historian could only be struck by the silences of the archives that refuse to reveal their secrets and, at the same time, by the possible openings they suggest, by the intuition that almost invisible threads link here and there characters and events.

Full text

1
Heartfelt imitation: high interoceptive 1
awareness is linked to greater automatic 2
imitation. 3
4
Vivien Ainley1, Marcel Brass2, & Manos Tsakiris1 5
6
7
8
1Lab of Action & Body, Department of Psychology, Royal Holloway, University of 9
London, UK 10
2Department of Experimental Psychology, Ghent University 11
12
Corresponding author: Vivien Ainley, Department of Psychology, Royal Holloway, 13
University of London, Egham, Surrey, UK. Tel. +44(0)1784276551, Fax. 14
+44(0)1784434347, E-mail: Vivien.Ainley.2008@live.rhul.ac.uk 15
16
17
Funding: This study was funded by the European Platform for Life Sciences, Mind 18
Sciences and Humanities, Volkswagen Foundation (II/85 064) and European 19
Research Council Starting Investigator Grant (ERC-2010-StG-262853) to MT20

2
Abstract 21
‘Interoceptive awareness’, defined as the individual’s awareness of internal body 22
signals, modulates self/other distinction under conditions of multisensory integration. 23
We examined here, for the first time, the potential impact of interoceptive awareness 24
on self/other distinction in the motor domain. In automatic imitation, inhibition of 25
imitation is an index of an individual’s success in distinguishing internally generated 26
motor representations from those triggered by observing another person’s action. This 27
is measured by the ‘congruency effect’, which is the difference between mean 28
reaction times when the observed action is ‘incongruent’ with the required action and 29
when it is ‘congruent’. The present study compared the congruency effect in a typical 30
finger lifting paradigm, with interoceptive awareness measured by heartbeat 31
perception. Contrary to expectation, interoceptive awareness was positively correlated 32
with the congruency effect and this effect depended on mean reaction times in the 33
incongruent condition, indicating that good heartbeat perceivers had more difficulty 34
inhibiting the tendency to imitate. Potentially, high interoceptive awareness involves 35
stronger interoceptive representations of the consequences of an action, implying 36
higher empathy, greater motor reactivity in response to observed action and hence a 37
greater tendency to imitate. Our results may also tentatively be explained within a 38
predictive coding account of interoception. 39
40
41
42
Highlights 43
• Interoceptive awareness modulates self/other distinctions in body-awareness 44
tasks. 45
• Automatic imitation also indexes the ability to distinguish ‘self’ from ‘other’. 46
• In a finger-lifting task, good heartbeat perceivers had larger ‘congruency 47
effects’. 48
• Interoceptive awareness correlated with difficulty in inhibiting imitation. 49
50

3
1. Introduction 51
The ability to distinguish between self and other is crucial to all aspects of self-52
processing and has relevance for action-awareness (Farrer et al., 2003), body-53
awareness (Tsakiris, 2013), empathy (Singer et al., 2004) and social cognition 54
(Lamm, Batson, & Decety, 2007). In the motor domain, self/other distinction has been 55
extensively studied using ‘automatic imitation’ paradigms (Brass, Bekkering, & Prinz, 56
2001; Catmur, Walsh, & Heyes, 2007), where the ability to resist imitating an action 57
performed by another person is taken to indicate a stronger sense of self (Spengler, 58
Brass, Kühn, & Schütz-Bosbach, 2010). Recent theories propose, however, that the 59
self is grounded in ‘interoception’, which refers to the signals arising from within the 60
body (Craig, 2010; Damasio, 2010; Seth, 2013). Awareness of such internal signals 61
has been shown to influence the ability to distinguish between self and other in 62
multisensory contexts (Suzuki, Garfinkel, Critchley, & Seth, 2013; Tsakiris, Tajadura-63
Jiménez, & Costantini, 2011). Given the inter-connectedness of perception and action 64
(Friston, 2010; Hommel, 2009) the purpose of this study was to investigate whether 65
awareness of interoceptive cues similarly impacts on self/other distinction in the 66
domain of action. 67
68
Humans have a tendency to involuntarily imitate actions that they observe. Thus, 69
when an individual is required to perform a given action, observing another person 70
perform an identical action typically facilitates performance, whereas observing a 71
different action generally interferes with it, even when the observed action is entirely 72
task-irrelevant (see Heyes, 2010, for a review). Although the term ‘automatic 73
imitation’ is commonly used, the phenomenon rarely involves true imitation, in that 74
people actually seldom perform the wrong action. They must, however, resist a 75
tendency to copy the action they observe. The ability to inhibit imitation is measured 76
by ‘the congruency effect’, which is the difference between the slower mean reaction 77
time (RT) typically found when the required and observed actions are ‘incongruent’ 78
(i.e. different) and the faster mean RT when the desired and observed actions are 79
‘congruent’ (Brass, Bekkering, Wohlschläger, & Prinz, 2000). 80
81
According to the Theory of Event Coding, automatic imitation occurs because actions 82
are coded in terms of their goals and thus their sensory consequences. The distinction 83
between perception and action is thus a false dichotomy (Hommel, Müsseler, 84
Aschersleben, & Prinz, 2001) and seeing an action necessarily primes the motor 85
representation of that action. The Associative Sequence Learning (ASL) theory 86
(Catmur, Walsh, & Heyes, 2009), suggests that visual and motor components of 87
actions are linked by long-term stimulus response (SR) bonds, such that the activation 88
of a visual mental representation necessarily predicts a motor representation (Heyes, 89
2010). More recently, the theory of predictive coding has linked perception and action 90
within a unified framework that may, in future, elucidate the neural mechanisms 91
behind automatic imitation (Adams, Shipp, & Friston, 2012; Friston, 2010). 92
93
Not only does automatic imitation rarely involve imitation but neither is it truly 94
‘automatic’, because it is not immune to interference by other processes. According to 95
the ASL model (Catmur et al., 2009) these processes can be divided into ‘input 96
modulation’, which alters the extent to which the relevant long-term SR bond is 97
activated, and ‘output modulation’, where social factors potentially inhibit the 98
involuntary imitation (Heyes, 2010). Input modulation is demonstrated by selective 99
attention to one’s own actions, which reduces imitation (Bortoletto, Mattingley, & 100

4
Cunnington, 2013; Chong, Cunnington, Williams, & Mattingley, 2009). Automatic 101
imitation also can be reduced by modest amounts of training (Cook, Press, Dickinson, 102
& Heyes, 2010; Gillmeister, Catmur, Brass, & Heyes, 2008; Heyes, Bird, Johnson, & 103
Haggard, 2005; Heyes & Bird, 2007), which reverses the muscle specificity of the 104
motor-evoked potentials (MEPs) produced by TMS (Catmur et al., 2007). 105
106
Output modulation depends on the top-down influence of participants’ traits and 107
social attitudes. Eye contact, or priming with pro-social cues, enhances the 108
congruency effect (Leighton, Bird, Orsini, & Heyes, 2010; Wang & Hamilton, 2012; 109
Wang, Newport, & Hamilton, 2011). Similarly, a desire to affiliate to the person 110
observed increases automatic imitation in both experimental settings and social 111
interaction (Lakin & Chartrand, 2003; Wang & Hamilton, 2012). People scoring high 112
in ‘self-monitoring’ (Snyder, 1974), or who have an interdependent self-construal, 113
have a greater tendency to mimic others, possibly as an unconscious affiliation 114
strategy (Cheng & Chartrand, 2003; Obhi, Hogeveen, & Pascual-Leone, 2011). 115
Interestingly, priming participants with examples of interdependent self-construal 116
increases the amplitude of MEPs elicited by TMS (Obhi et al., 2011), indicating that 117
these top-down influences increase cortical excitability in the motor areas that 118
produce imitation. 119
120
Automatic imitation is one of a number of phenomena which involve ‘self/other 121
overlap’, defined as “any phenomenon whereby an observer engages a state similar 122
to that of the target, via activation of the observer’s personal representations for 123
experiencing the observed state, whether through direct perception or simulation” 124
(Preston & Hofelich, 2012). These shared representations occur at a very early, 125
preconscious, processing stage. The ability to inhibit imitation requires that the 126
individual distinguishes between internally generated motor representations and those 127
that are triggered by observing other people’s actions (Brass, Ruby, & Spengler, 128
2009). Successfully inhibiting the tendency to imitate activates cortical areas thought 129
to be involved in discriminating between self and other (Brass, Derrfuss, & von 130
Cramon, 2005; Brass et al., 2009; Brass & Heyes, 2005). The most active of these 131
regions - the temporal parietal junction and anterior fronto-median cortex (BA10) - 132
are related to perspective taking, feelings of agency and theory of mind (Wang, 133
Ramsey, & Hamilton, 2011). Greater activation in BA10 correlates with smaller 134
congruency effects and thus with better self/other distinction (Spengler et al., 2009). 135
Furthermore, experimentally increasing self-focus reduces the congruency effect, by 136
reducing RTs on incongruent trials (Spengler, Brass, Kühn, & Schütz-Bosbach, 137
2010). Similarly, observing an action increases the amplitude of MEPs if that action is 138
attributed to another individual but reduces cortico-spinal excitability when the action 139
is illusorily attributed to the self (Schutz-Bosbach, Mancini, Aglioti, & Haggard, 140
2006). 141
142
Automatic imitation can therefore be characterised as a tool to measure how 143
effectively the self can be distinguished from others (Spengler, von Cramon, & Brass, 144
2009). The purpose of the current experiment was to investigate how the congruency 145
effect is linked to ‘interoceptive awareness’ - a fundamental dimension of self-146
awareness that has been the focus of recent research in body ownership (Tsakiris et 147
al., 2011), self-recognition (Tajadura-Jiménez & Tsakiris, 2013) and empathy 148
(Fukushima, Terasawa, & Umeda, 2011). 149
150

5
Recent neuroscientific models of the self emphasize the role of ‘interoception’ (Craig, 151
2010; Critchley & Harrison, 2013; Hayes & Northoff, 2012; Panksepp & Northoff, 152
2009) defined as “the afferent information arising from within the body, affecting the 153
cognition, emotion or behaviour of an organism, with or without awareness” 154
(Cameron, 2001). Insular cortex, which is activated by all feelings arising within the 155
body (Craig, 2010; Critchley & Harrison, 2013; Singer, Critchley, & Preuschoff, 156
2009; Wiebking et al., 2013; Zaki, Davis, & Ochsner, 2012), may underpin this 157
fundamental representation of self (Craig, 2009; Seth, 2013; but see also Philippi et 158
al., 2012). Recent predictive coding accounts of cortical function (Clark, 2013; 159
Friston, 2010) similarly propose interoceptive information as an essential component 160
of the self (Apps & Tsakiris, 2013; Seth, Suzuki, & Critchley, 2011). ‘Interoceptive 161
awareness’, which is the extent to which internal signals reach consciousness, has 162
been extensively studied in relation to emotion, stemming originally from William 163
James’ theory that emotion comprises unconscious bodily responses (Damasio & 164
Carvalho, 2013; James, 1890). 165
166
Recent studies have begun to investigate the contribution of interoceptive awareness 167
to self-processing. In the rubber hand illusion, people with low interoceptive 168
awareness are more likely to claim ownership over a prosthetic hand, (Tsakiris et al., 169
2011) and similarly experience a stronger illusory identification with a stranger’s face 170
when they observe that face being stroked synchronously with felt touch on their own 171
face (Tajadura-Jiménez & Tsakiris, 2013). Conversely, enhanced self-focus, through 172
mirror self-observation, a self-photograph or self-relevant words, can improve 173
interoceptive awareness in people for whom this is initially low (Ainley, Maister, 174
Brokfeld, Farmer, & Tsakiris, 2013; Ainley, Tajadura-Jiménez, Fotopoulou, & 175
Tsakiris, 2012; Maister, Tsiakkas, & Tsakiris, 2013). Individuals who see a virtual 176
image of their own hand (Suzuki, Garfinkel, Critchley, & Seth, 2013) or of their 177
whole body (Aspell et al., 2013) have a greater sense of self-identification with, and 178
self-location towards, the image under conditions of cardio-visual synchrony. 179
180
Despite these investigations into the contribution of interoceptive awareness to 181
self/other distinction in multisensory contexts, little is known about the potential role 182
of interoception in the action system, for example in automatic imitation. This lack of 183
empirical research is striking, given that human actions are thought to be driven by 184
the goal of homeostatic control, which is signaled interoceptively (Craig, 2010; 185
Damasio, 2010; Seth, 2013). Theoretical accounts of the neural basis of perception 186
and action stress their inter-connectedness (Friston, 2010; Schütz-Bosbach & Prinz, 187
2007). While it has been previously assumed that the sensory consequences of an 188
action are primarily exteroceptive, empathy for pain (Avenanti, Bueti, Galati, & 189
Aglioti, 2005; Singer et al., 2004) and overlapping cortical activation during the 190
experience, observation or imagination of disgust (Wicker et al., 2003) can only be 191
explained if actions involve a representation of their interoceptive sensory 192
consequences (Heyes & Bird, 2007). 193
194
Given that the ability to inhibit automatic imitation seems to index better self/other 195
distinction, at the level of visual and motor representation, and also that people with 196
high interoceptive awareness appear more reliably able to distinguish their own 197
bodies from those of others, at a multisensory level, we hypothesised that in an 198
automatic imitation paradigm individuals with high interoceptive awareness would 199
successfully inhibit the tendency to imitate, whereas those with low interoceptive 200

6
awareness would exhibit less self/other distinction and would therefore have a greater 201
tendency to automatic imitation. 202
203
‘Interoceptive awareness’ is generally assessed using a heartbeat perception task 204
(Schandry, 1981; Whitehead & Drescher, 1980). Such measures correlate with 205
awareness of gastric cues (Herbert, Muth, Pollatos, & Herbert, 2012; Whitehead & 206
Drescher, 1980). We used the Mental Tracking task (Schandry, 1981) which is well-207
validated (Knoll & Hodapp, 1992), with good test retest reliability (Mussgay, 208
Klinkenberg, & Rüddel, 1999; Werner, Kerschreiter, Kindermann, & Duschek, 2013) 209
and which discriminates well between individuals. The measure we have called 210
‘interoceptive awareness’ in this study assesses the accuracy of cardiac awareness, by 211
comparing the subjectively reported number of heartbeats experienced with the 212
number (objectively) recorded (Cuenen, Van Diest, & Vlaeyen, 2012; Garfinkel & 213
Critchley, 2013). Gender, body mass index (BMI), and resting heart rate were also 214
recorded, as possible confounds of the heartbeat perception task (Cameron, 2002). 215
Automatic imitation was assessed using an established inhibition imitation paradigm 216
developed by Brass and colleagues (Brass et al., 2005; Spengler et al., 2009). It was 217
anticipated that people who performed accurately in heartbeat perception would also 218
be more accurate during the automatic attention task (show a smaller congruency 219
effect). However, both these variables might be affected by participants’ general 220
willingness and ability to attend to the tests. Attention is a possible source of input 221
modulation in automatic imitation (Davis, 1983; Kaplan & Iacoboni, 2006; Preston & 222
Hofelich, 2012). It has also been reported (Matthias, Schandry, Duschek, & Pollatos, 223
2009) that interoceptive awareness is linked to scores on the d2 test (Brickenkamp & 224
Zilmer, 1998), which measures individual differences in motivation and attention. We 225
accordingly administered the d2 test as a check for this potential confound. 226
227
2. Method 228
2.1 Ethics statement 229
The study was approved by the Department of Psychology Ethics Committee, Royal 230
Holloway University of London. All participants gave written informed consent and 231
were free to withdraw from the experiment at will. 232
233
2.2 Participants 234
Participants were 45 students at Royal Holloway University of London who 235
participated for course credit. All declared themselves right handed and had normal or 236
corrected to normal vision. The data for 2 participants was excluded for excessive 237
numbers of errors (more than 10%, i.e. 3SD above the mean) in the action imitation 238
task, indicating a failure to concentrate and follow the instructions. Of the remaining 239
43 participants, mean age = 19.6 (SD = 4.9), 9 were male. 240
241
2.3 Stimuli 242
The stimuli consisted of sequences of 5 frames (Brass et al., 2005; Spengler et al., 243
2009). Each video stared with a frame showing the hand, which mirrored the right 244
hand of the subject, in the starting position, for 2s. The next two frames, each lasting 245
34ms, presented a number (either 1 or 2) and simultaneously showed the finger 246
movement (if any). The fourth frame showed the finger in the end position for 1.3s, 247
with the number (1 or 2) superimposed. Between trials, the screen turned black for 248
2.7s. Each video trial was thus 6s duration. The video hand was presented on a blue 249
rectangular background, measuring 22 x 12cm. 250

7
251
Figure 1. Example of the video stimuli for the index finger, from Brass et al., (2005) 252
253
254
There were six possible video sequences, consisting of each of the two fingers (index 255
or middle) in each of three conditions (baseline, congruent or incongruent). 256
Participants were required to lift either the index (1) or middle (2) finger in response 257
to a number appearing on the screen. The three possible conditions (for the index 258
finger) are shown in Figure 1. Thus in the baseline condition, simultaneous with the 259
appearance of the number, the video hand remained static. In the congruent condition 260
the video hand lifted the finger that corresponded to the number shown (i.e. the index 261
finger was lifted when the number 1 appeared). In the incongruent condition the video 262
hand lifted the ‘wrong’ finger (i.e. the middle finger was lifted when the number 1 263
appeared). 264
265
2.4 Procedure 266
2.4.1 Interoceptive awareness 267
After giving informed consent, participants’ gender, age, height and weight were 268
recorded. Heartbeat signals were acquired with a piezo-electric pulse transducer, fitted 269
to the participant’s left index finger and connected to a physiological data unit (26T 270
PowerLab, AD Instruments) sampling at 1 kHz which recorded the derived electrical 271
signal onto a second PC running LabChart6 software (AD Instruments). Instructions 272
for the Mental Tracking Method (Schandry, 1981) were presented over noise-273
attenuating headphones. The onset and offset of each heartbeat counting trial were 274
cued by the words “go” and “stop”, presented audiovisually. We used a standard 275
instruction (Ehlers, Breuer, Dohn, & Fiegenbaum, 1995) whereby participants were 276
asked to concentrate hard and try to silently count their own heartbeats, simply by 277
“listening” to their bodies, without taking their pulse. The three trials (25s, 35s & 45s) 278
were presented in random order. A criticism of the Mental Tracking Method is that 279
participants may estimate the elapsed interval and then use knowledge of their own 280
heart rate to guess the number of heartbeats. We therefore asked individuals to 281
estimate the length of three, randomly presented, intervals (19s, 37s, 49s) and to 282
provide an estimate of their resting heart rate (Dunn et al., 2010). 283
284

8
2.4.2 Action imitation 285
The stimuli were viewed on a standard PC, using Presentation software 286
(Neurobehavioral Systems, Albany, CA). Participants were seated about 60 cm in 287
front of the screen and were instructed to execute their movements as quickly and 288
accurately as possible. Participants placed the index and middle fingers of their right 289
hand on a serial response box which was linked to another PC which recorded the 290
times of all finger movements, using Spike2 software (Cambridge Electronic Design, 291
Cambridge UK). This recorded the onset of the visible stimulus on screen (i.e. the 292
number 1 or 2, which coincided with the onset of movement of the video hand) and 293
also recorded whenever the participant lifted an index or middle finger. Following 6 294
tests trials, 150 trials experimental trials were presented in three blocks of 5mins, with 295
obligatory rests of at least 2mins between blocks. The order of the presentation of the 296
trials was fully randomised and comprised 25 trials in each of the 6 conditions. 297
298
2.4.3 The d2 test of attention 299
Finally, the d2 test was administered (Brickenkamp & Zilmer, 1998). This is a widely 300
used measure of selective visual attention. The test items consist of the letters d and p 301
with up to four dashes, arranged either individually or in pairs, above and/or below 302
each letter. The subject is given 20s to scan across each of the 14 closely printed test 303
lines, during which they must identify and cross out every letter d which has exactly 304
two dashes, while ignoring all other distractor letters. The d2 test produces several 305
norm-referenced scores, of which the most commonly reported are the total number of 306
items processed (TN) regardless of whether these are correct or incorrect (this is a 307
measure of processing speed), the percentage of errors made (E%) and the total 308
number of items processed correctly (TN-E). This final score is designed to provide a 309
measure of the capacity to selectively orient to relevant aspects of the task, while 310
screening out irrelevant ones. 311
312
2.5 Data reduction 313
2.5.1 Interoceptive Awareness 314
LabChart6 was employed to identify and count the number of R-wave peaks on the 315
heart trace recorded for each participant in each trial, as well as to calculate the 316
average heart rates for each trial (Jennings et al., 1981). Every heart trace was visually 317
inspected for artefacts and the number of R-wave peaks was recounted manually, if 318
necessary. No participant was excluded due to artefacts. Interoceptive awareness was 319
calculated as (1/3Σ (1-(|recorded heartbeats – counted heartbeats|/recorded 320
heartbeats)) (Schandry, 1981). Higher scores indicate higher interoceptive awareness. 321
As a control on guessing, the participant’s ability to estimate the length of an elapsed 322
interval was also calculated as (1/3Σ (1-(|estimated elapsed time – actual elapsed 323
time|/actual elapsed time)) which we called the “time modulus” measure (Dunn et al., 324
2010). 325
326
2.5.2 Action imitation 327
Data was extracted using Matlab (mathworks.com) and analysed with Microsoft 328
Excel. The mean reaction time (RT) was calculated for each of the 6 conditions 329
(congruent, incongruent and baseline, for each of the two fingers). The ‘congruency 330
effect’ was found by subtracting the mean RT for congruent trials from the mean RT 331
for incongruent trials. 332
333
3. Results 334

9
3.1 Error analysis 335
RT errors were removed before analysis. There were 2 possible sources of RT errors. 336
Firstly, participants occasionally lifted the wrong finger. Secondly, in common with 337
most RT analyses, some response times were omitted as outliers (Miller & Diego, 338
1991). Thus RTs less than 80ms or greater than 800ms were excluded from the RT 339
analysis (Brass, Bekkering, & Prinz, 2001). The rate for all errors was 2.3% of trials. 340
Two participants were excluded for total errors > 10% i.e. 3SD above the mean. The 341
distribution of errors was thereafter approximately Normal, skewness = .64, kurtosis = 342
-.16. 343
344
Paired sample t tests (with Bonferroni correction for multiple comparisons and a 345
significance level of 0.017) showed that there were significantly more errors in the 346
incongruent condition than in the baseline, t(42) = 5.07, p < .001, but no significant 347
difference between the numbers of errors in the congruent condition and baseline, 348
t(42) = 0.82, p = .42, replicating the finding of Brass et al. (2005). 349
350
3.2 Reaction Time (RT) Analysis 351
Repeated measures ANOVA was performed, with both the finger (index or middle) 352
and the condition (congruent, incongruent and baseline) as within-subjects variables. 353
Mauchley’s test of Sphericity was significant; therefore Greenhouse Geisser 354
corrections were applied. There was a main effect of condition (RTs in the 355
incongruent conditions were slower), F(2, 84) = 186.4, p < .001. This indicates 356
significant automatic imitation i.e. slower mean RTs in the incongruent than 357
congruent condition, for both fingers (Brass et al., 2000; Brass, Derrfuss, & von 358
Cramon, 2005). There was a main effect of finger, F(1, 42) = 13.2, p = .001 (reaction 359
times were generally faster for the middle finger), as shown in Figure 2. The 360
interaction of finger and condition was also significant, F(2, 84) = 8.9, p < .001. 361
Paired samples t tests (with Bonferroni correction and a significance level of 0.008) 362
showed that, compared with RTs in the baseline, RTs in the incongruent condition 363
were significantly longer when participants were required to lift their index finger 364
rather than their middle finger, t(42) = 3.32, p = .002. However, there was no 365
significant difference between the two fingers for RTs in the congruent condition, 366
compared with the baseline, t(42) = .57, p = .57. Despite the significantly shorter RTs 367
for the middle finger, particularly in the incongruent condition, the relationships 368
between interoceptive awareness and the various reaction time measures in our study 369
were very similar for the two fingers. For the remaining analysis we therefore used 370
the mean of the data for the index and middle fingers, to give a single measure of 371
average RT in each condition. 372
373
Figure 2. Mean reaction times by condition and by finger (Errors bars = SEM) 374
375

10
376
To investigate the relationship between interoceptive awareness and the congruency 377
effect, we calculated the latter, in the standard way (as the mean RT in the 378
incongruent condition minus the mean RT in the congruent condition), for the average 379
of the two fingers, for each participant. Correlations between interoceptive awareness 380
and differences in RTs between conditions are shown in Table 1. Interoceptive 381
awareness was positively correlated with the congruency effect (Figure 3) and this 382
was wholly accounted for by RTs in the incongruent condition. Interoceptive 383
awareness was significantly correlated with the difference between mean RTs in the 384
incongruent condition and the baseline but not with the difference between mean RTs 385
in the congruent and baseline conditions. 386
387
Figure 3. Scatter diagram of the average congruency effect against interoceptive 388
awareness 389
390
391
392
Table 1. Correlations between interoceptive awareness (IA) and RT measures 393
394

11
IA & ‘the congruency effect’ (mean RT in incongruent condition
minus the congruent condition)
r = .41
p = .006**
IA & mean RT in the incongruent condition minus the baseline
r = .45
p = .002**
IA & mean RT in the congruent condition minus the baseline
r = -.04
p = .73
** significant at the 1% level 395
396
The wide range of mean RTs amongst our participants (318ms - 513ms, median 397
398ms) might have affected our results. We therefore calculated the percentage 398
difference in RTs between the incongruent and congruent conditions using the 399
formula [{(mean RT incongruent - mean RT congruent)/mean RT baseline} x 100]. 400
This statistic was also significantly positively correlated with interoceptive awareness, 401
r = .40, p = .008. 402
403
In this experiment we recorded a number of confounding variables known to impact 404
on interoceptive awareness, namely gender, Body Mass Index (BMI), resting heart 405
rate, and two measures designed to assess possible guessing on the Mental Tracking 406
task (i.e. the ‘time modulus’ measure of the participant’s ability to estimate elapsed 407
time, and the participant’s belief about his/her heart rate). An independent samples t 408
test (with equal variances not assumed) showed no effect of gender on interoceptive 409
awareness, t(41) = 1.32, p = .24. Likewise the correlation of interoceptive awareness 410
and BMI was not significant, r = -.20, p = .21. Although people with slower hearts are 411
often better heartbeat perceivers (Ainley et al., 2012; Cameron, 2001; Knapp-Kline & 412
Kline, 2005), in this sample the correlation of interoceptive awareness and average 413
heart rate did not reach significance r = -.22, p = .16. 414
415
The ‘time modulus’ measure (of participants’ ability to estimate the length of an 416
elapsed interval) was correlated with interoceptive awareness, r = .35, p = .02 but the 417
correlation of interoceptive awareness and participants’ estimates of their own heart 418
rates was not significant, r = -.08, p = .62. 419
420
Table 2. Hierarchical multiple regression with the average congruency effect as the 421
dependent variable 422
Step 1
Step 2
Step 3
Step 4
Step 5
Step 6
Independent
variables
Beta
(p)
Interoceptive
awareness (IA)
1.74
(p = .36)
1.96
(p = .16)
1.90
(p =.16)
.52
(p = .14)
0.40 **
(p = .009)
0.41**
(p = .006)
Average heart
rate (HR)
0.22
(p = .82)
0.26
(p = .78)
‘Time modulus’
-0.76
(p = .64)
-0.66
(p = .66)
-0.10
(p = .29)

12
* significant at the 5% level 423
** significant at the 1% level 424
425
Given previous correlations in the literature between interoceptive awareness and both 426
participants’ average heart rates and the ‘time modulus’ measure (Cameron, 2002; 427
Dunn et al., 2010), we performed a hierarchical multiple regression analysis with the 428
average congruency effect as the dependent variable and independent variables 429
comprising interoceptive awareness, average heart rate, ‘time modulus’, and their 430
interactions. Only interoceptive awareness had any significant effect on the 431
congruency effect (see Table 2). 432
433
Results of the d2 test of attention were analysed in terms of the total number of items 434
processed (TN), total number correct (TN-E) and percentage of errors (E%). 435
Compared with published norms, d2 scores for our participants (mean TN = 516, 436
mean (TN-E) = 493) were at the 70th percentile for students. Previous research 437
(Matthias et al., 2009) found significant correlation between interoceptive awareness 438
and TN but in this experiment none of the d2 measures were correlated with 439
interoceptive awareness, for TN r = .03, p = .87, for (TN-E) r = .04, p = .82 and for 440
(E%) r = -.02, p = .92. To replicate the analysis of Matthias et al. (2009), we split the 441
data using their cut off at interoceptive awareness = .85 but found no significant 442
difference in any d2 measures between ‘good’ (interoceptive awareness > .85, n = 5) 443
and ‘poor’ (interoceptive awareness < .85, n = 38) heartbeat perceivers (e.g. for TN, 444
F(1, 41) = .46, p = .50). There were likewise no significant correlations between any 445
of the d2 measures and the average congruency effect, for TN r = .18, p = .24, for 446
(TN-E) r = .11, p = .47, and for (E%) r = .15, p = .32. 447
448
4. Discussion 449
We investigated the relationship between interoceptive awareness and automatic 450
imitation, measuring interoceptive awareness (IA) with a well-validated heartbeat 451
perception task (Schandry, 1981) and automatic imitation by a widely used finger-452
lifting paradigm (Brass et al., 2005). The expected ‘congruency effect’ was obtained, 453
i.e. mean reaction times (RTs) were slower when the observed and required actions 454
were incongruent and were faster when they were congruent (compared with the 455
baseline of no observed movement). Interoceptive awareness was significantly 456
positively correlated with the congruency effect. This was fully accounted for by the 457
difference between RTs in the incongruent condition and the baseline. There were no 458
Interaction of
IA & ‘time
modulus’
0.19
(p = .86)
Interaction of
IA & HR
-1.47
(p = .31)
-1.55
(p = .25)
-1.49
(p = .26)
-.15
(p = .69)
Interaction of
‘time modulus’
& HR
0.86
(p = .62)
0.85
(p = .62)
1.22
(p = .23)
.15
(p = .40)
.11
(p = .45)
Adjusted R2
(p)
.08
(p = .18)
.10
(p = .11)
.13
(p = .06)
.12*
(p = .05)
.14**
(p = .02)
.15**
(p = .006)

13
significant effects of interoceptive awareness on RTs difference between the 459
congruent and the baseline. Thus the relationship we observed depended on RTs the 460
incongruent condition and thus on interference between the observed and required 461
action (Blakemore & Frith, 2005), indicating that people with high interoceptive 462
awareness had greater difficulty inhibiting the tendency to automatically imitate. Had 463
there been a motor facilitation effect, it would have taken the form of shorter RTs on 464
congruent trials. RTs in the incongruent condition were significantly slower for the 465
index finger than for the middle finger, probably because lifting an index finger is a 466
more familiar experience than the isolated lifting of a middle finger, with a 467
consequently stronger, learned associative bond. 468
469
The result we obtained was contrary to our original hypothesis. Experiments in 470
multisensory integration have suggested that people with high interoceptive 471
awareness are better at making self/other body ownership distinctions (Tajadura-472
Jiménez & Tsakiris, 2013; Tsakiris et al., 2011). We hypothesized that this effect 473
might translate into the motor domain. The ability to inhibit imitation is assumed to 474
index self/other distinction (Spengler et al., 2009) and we therefore predicted that 475
people with high interoceptive awareness would more successfully inhibit the 476
tendency to imitate. Our results show that, on the contrary, they were more inclined to 477
imitate, implying greater self/other overlap. 478
479
Despite the findings from body-ownership paradigms, which suggest that high 480
interoceptive awareness is linked to better ability to make self/other distinctions, this 481
is likely to be context dependent. Thus while low interoceptive awareness might 482
predict greater ability to distinguish between self and other in cases of multisensory 483
body-related integration (Tajadura-Jiménez & Tsakiris, 2013; Tsakiris et al., 2011), in 484
other contexts high interoceptive awareness seems to suggest greater self/other 485
overlap. A fundamental difference between self/other distinction in the automatic 486
imitation task and self/other distinction in the rubber hand illusion is that confusion in 487
the automatic imitation task is at a representational level and at a point in time where 488
participants have no sensory information about their own movements. The link 489
between interoceptive awareness and automatic imitation may therefore be indirect 490
and depend on the sensitivity of people with high interoceptive awareness to social 491
influences. Thus the concentration of our effect in incongruent cues indicates that it 492
depended on the action observation aspect of the task and therefore on output 493
modulation, rather than the preparation of the individual’s own action (input 494
modulation). The lack of correlation between the congruency effect and the d2 test 495
also supports this conclusion. The d2 test scores are measures of “the capacity to 496
selectively orient to relevant aspects of the task while screening out irrelevant ones” 497
(Zimmerman & Frimm, 2002). The d2 was included to counter the criticism that if we 498
had found the hypothesised correlation between high accuracy in both the heartbeat 499
detection and the automatic imitation tasks, this might have reflected the participants’ 500
level of motivation and attention. We did not replicate previous reports of a 501
correlation between high interoceptive awareness and selective and divided attention 502
(Matthias et al., 2009), indicating that general differences in individuals’ motivation 503
and attention to the tasks were unlikely to have confounded our results. 504
505
In terms of the Associative Sequence Learning model of automatic imitation (Catmur 506
et al., 2009) output modulation is occasioned by social factors which influence 507
individuals to suppress or enhance the tendency to imitate. High interoceptive 508

14
awareness has been linked to anxiety (Domschke et al., 2010) and particularly to 509
social anxiety (Terasawa, Shibata, Moriguchi, & Umeda, 2013). We did not assess 510
trait anxiety in this study but potentially, if our high interoceptive awareness 511
participants were more socially anxious, they might have had a greater desire to 512
affiliate, which could have enhanced their tendency to imitate. 513
514
A potential source of output modulation is affective empathy, which is assumed to 515
involve shared representations between one’s own emotional state and that of another 516
individual (Decety & Jackson, 2004; Iacoboni, 2009; Preston & Hofelich, 2012; Zaki, 517
Weber, Bolger, & Ochsner, 2009). People with high interoceptive awareness are 518
thought to exhibit greater empathy (Ernst et al., 2013; Terasawa, Shibata, Moriguchi, 519
& Umeda, 2013), perhaps because they have a stronger interoceptive representation of 520
the consequences of an observed action, for example, they are more sensitive to 521
masked fear conditioning (Katkin, Wiens, & Ohman, 2001). Scores on the empathetic 522
concern scale of the Interpersonal Reactivity Index (Davis, 1983) correlate with the 523
amplitude of heartbeat evoked potentials (Fukushima, Terasawa, & Umeda, 2011), 524
which are larger in people with high interoceptive awareness (Pollatos & Schandry, 525
2004). Empathy has, in turn, been linked to action observation. Kaplan and Iacoboni 526
(2006) found that when participants observed another individual reaching for a cup, 527
inferior frontal mirror activity was greater in those people who had higher scores on 528
the Empathetic Concern subscale. Such motor activity in response to action 529
observation is also linked to a greater tendency to imitate (Catmur et al., 2007; Obhi 530
et al., 2011; Schutz-Bosbach et al., 2006). Empathy is inversely correlated with 531
narcissism and it has recently been shown that individuals who are high in trait 532
narcissism - thus displaying a lack of empathy and concern for others - have a greater 533
ability to inhibit automatic imitation (Obhi, Hogeveen, Giacomin, & Jordan, 2013). 534
Thus high interoceptive awareness may involve stronger interoceptive representation 535
of the consequences of an action, implying higher empathy, greater mirror neuron 536
activity in response to observed action and hence a greater tendency to imitate. 537
538
Our results may alternatively depend on some hitherto unexplored aspect of 539
interoceptive awareness and its relationship to the action system. Given that accounts 540
of cortical function, including both the Theory of Event Coding (Hommel, 2009) and 541
predictive coding (Clark, 2013; Friston, 2010) stress that perception and action are 542
reciprocally connected, further research is needed to confirm whether interoceptive 543
awareness impacts not only on action in interoceptive systems but on motor activity 544
as well. The basis of inter-individual differences in interoceptive awareness is not 545
well understood (Verdejo-Garcia, Clark, & Dunn, 2012). Such differences have 546
generally been assumed to depend simply on the strength of interoceptive signals 547
arising within the body, which are conveyed principally by the vagus nerve (Craig, 548
2003; Cameron, 2002; Critchley et al., 2007). However, interoceptive awareness may 549
perhaps be interpreted in a predictive coding context (Friston, 2010; Seth et al., 2011). 550
551
Hypothetically, high interoceptive awareness might relate to the high ‘precision’ of 552
interoceptive signals, which could, in turn, account for the high levels of autonomic 553
activity that have been observed in people with good interoceptive awareness 554
(Herbert, Pollatos, Flor, Enck, & Schandry, 2010; Pollatos, Füstös, & Critchley, 555
2012). Although very speculative, it seems possible that interoceptive signals are 556
more reliable and attended (i.e. more precise) in people with high interoceptive 557
awareness, which would account for these individuals’ reduced liability to body 558

15
ownership illusions. Given that interoceptive awareness affects perception of the 559
body, it is also likely to modulate action representations. It has recently been indicated 560
that in order to avoid mirroring another person’s actions it is essential to reduce the 561
precision of proprioceptive precision errors (Friston, Mattout, & Kilner, 2011). If 562
people with high interoceptive awareness have initially precise proprioceptive 563
precision errors, then their tendency to imitate others may be accounted for. 564
Potentially, recently observed individual differences in levels of neurotransmitters in 565
the insula (e.g. Wiebking et al., 2013) may provide the means to unravel the links 566
between interoceptive signals and proprioceptive, motor and autonomic reflexes. 567
568
5. Conclusion 569
Interoceptive awareness, measured by the accuracy with which people perceive their 570
own heartbeats, is known to modulate self/other distinction in multisensory contexts. 571
Here we demonstrate for the first time that interoceptive awareness also impacts on 572
shared representations in the motor domain, such that people with high interoceptive 573
awareness have greater difficulty in inhibiting the tendency to imitate, in a standard 574
automatic imitation paradigm. 575
576
6. Acknowledgements: We thank Professor Narender Ramnani for the loan of 577
equipment. 578
579
7. References 580
Adams, R. A., Shipp, S., & Friston, K. (2012). Predictions not commands: active 581
inference in the motor system. Brain Structure & Function, 218(3), 611–643. 582
doi:10.1007/s00429-012-0475-5 583
Ainley, V., Maister, L., Brokfeld, J., Farmer, H., & Tsakiris, M. (2013). More of 584
myself: Manipulating interoceptive awareness by heightened attention to bodily 585
and narrative aspects of the self. Consciousness and Cognition, 22(4), 1231–586
1238. doi:10.1016/j.concog.2013.08.004 587
Ainley, V., Tajadura-Jiménez, A., Fotopoulou, A., & Tsakiris, M. (2012). Looking 588
into myself: Changes in interoceptive sensitivity during mirror self-observation. 589
Psychophysiology, 49(11), 1672–1676. 590
Apps, M. A. J., & Tsakiris, M. (2013). The free energy self: A predictive coding 591
account of self-recognition. Neuroscience and Biobehavioral Reviews, 1–13. 592
doi:dx.doi.org/10.1016/j.neubiorev.2013.01.029 593
Aspell, J. E., Heydrich, L., Marillier, G., Lavanchy, T., Herbelin, B., & Blanke, O. 594
(2013). Turning Body and Self Inside Out: Visualized Heartbeats Alter Bodily 595
Self-Consciousness and Tactile Perception. Psychological Science, 24(12), 596
2445–2453. doi:10.1177/0956797613498395 597
Avenanti, A., Bueti, D., Galati, G., & Aglioti, S. M. (2005). Transcranial magnetic 598
stimulation highlights the sensorimotor side of empathy for pain. Nature 599
Neuroscience, 8(7), 955–60. doi:10.1038/nn1481 600

16
Blakemore, S. J., & Frith, C. D. (2005). The role of motor contagion in the prediction 601
of action. Neuropsychologia, 43(2), 260–7. 602
doi:10.1016/j.neuropsychologia.2004.11.012 603
Bortoletto, M., Mattingley, J. B., & Cunnington, R. (2013). Effects of context on 604
visuomotor interference depends on the perspective of observed actions. PloS 605
One, 8(1), e53248. doi:10.1371/journal.pone.0053248 606
Brass, M., Bekkering, H., & Prinz, W. (2001). Movement observation affects 607
movement execution in a simple response task. Acta Psychologia, 106, 3–22. 608
Brass, M., Bekkering, H., Wohlschläger, A., & Prinz, W. (2000). Compatibility 609
between observed and executed finger movements: Comparing symbolic, spatial, 610
and imitative cues. Brain and Cognition, 44(2), 124–43. 611
doi:10.1006/brcg.2000.1225 612
Brass, M., Derrfuss, J., & von Cramon, D. Y. (2005). The inhibition of imitative and 613
overlearned responses: a functional double dissociation. Neuropsychologia, 614
43(1), 89–98. doi:10.1016/j.neuropsychologia.2004.06.018 615
Brass, M., & Heyes, C. (2005). Imitation: is cognitive neuroscience solving the 616
correspondence problem? Trends in Cognitive Sciences, 9(10), 489–95. 617
doi:10.1016/j.tics.2005.08.007 618
Brass, M., Ruby, P., & Spengler, S. (2009). Inhibition of imitative behaviour and 619
social cognition. Philosophical Transactions of the Royal Society of London. 620
Series B, Biological Sciences, 364, 2359–2367. doi:10.1098/rstb.2009.0066 621
Brickenkamp, R., & Zilmer, E. (1998). The d2 test of attention. Oxford: Hogrefe Ltd. 622
Cameron, O. G. (2001). Interoception: the inside story - a model for psychosomatic 623
processes. Psychosomatic Medicine, 63(5), 697–710. 624
Cameron, O. G. (2002). Visceral Sensory Neuroscience. Oxford: Oxford University 625
Press. 626
Catmur, C., Walsh, V., & Heyes, C. (2007). Sensorimotor learning configures the 627
human mirror system. Current Biology, 17(17), 1527–31. 628
doi:10.1016/j.cub.2007.08.006 629
Catmur, C., Walsh, V., & Heyes, C. (2009). Associative sequence learning: the role of 630
experience in the development of imitation and the mirror system. Philosophical 631
Transactions of the Royal Society B: Biological Sciences, 364(1528), 2369–632
2380. doi:10.1098/rstb.2009.0048 633
Cheng, C. M., & Chartrand, T. L. (2003). Self-monitoring without awareness: Using 634
mimicry as a nonconscious affiliation strategy. Journal of Personality and Social 635
Psychology, 85(6), 1170–1179. 636

17
Chong, T. T.-J., Cunnington, R., Williams, M., & Mattingley, J. B. (2009). The role 637
of selective attention in matching observed and executed actions. 638
Neuropsychologia, 47(3), 786–95. doi:10.1016/j.neuropsychologia.2008.12.008 639
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of 640
cognitive science. The Behavioral and Brain Sciences, 36(3), 181–204. 641
doi:10.1017/S0140525X12000477 642
Cook, R., Press, C., Dickinson, A., & Heyes, C. (2010). Acquisition of automatic 643
imitation is sensitive to sensorimotor contingency. Journal of Experimental 644
Psychology. Human Perception and Performance, 36(4), 840–52. 645
doi:10.1037/a0019256 646
Craig, A. D. (2003). Interoception: the sense of the physiological condition of the 647
body. Current Opinion in Neurobiology, 13(4), 500–505. doi:10.1016/S0959-648
4388(03)00090-4 649
Craig, A. D. (2009). How do you feel--now? The anterior insula and human 650
awareness. Nature Reviews. Neuroscience, 10(1), 59–70. doi:10.1038/nrn2555 651
Craig, A. D. (2010). The sentient self. Brain Structure and Function, 214, 563–577. 652
doi:10.1007/s00429-010-0248-y 653
Critchley, H. D., & Harrison, N. (2013). Visceral Influences on Brain and Behavior. 654
Neuron, 77, 624–638. doi:doi.org/10.1016/j.neuron.2013.02.008 655
Critchley, H. D., Lewis, P. A., Orth, M., Josephs, O., Deichmann, R., Trimble, M. R., 656
& Dolan, R. J. (2007). Vagus nerve stimulation for treatment-resistant 657
depression: behavioral and neural effects on encoding negative material. 658
Psychosomatic Medicine, 69(1), 17–22. doi:10.1097/PSY.0b013e31802e106d 659
Critchley, H. D., Wiens, S., Rotshtein, P., Ohman, A., & Dolan, R. J. (2004). Neural 660
systems supporting interoceptive awareness. Nature Neuroscience, 7(2), 189–661
195. doi:10.1038/nn1176 662
Cuenen, E., Van Diest, I., & Vlaeyen, J. (2012). Accuracy and awareness of 663
perception: Related, yet distinct. Biological Psychology, 92(2), 426–427. 664
Damasio, A. R. (2010). Self comes to mind: Constructing the conscious brain. 665
London: William Heinemann. 666
Damasio, A. R., & Carvalho, G. B. (2013). The nature of feelings: evolutionary and 667
neurobiological origins. Nature Reviews Neuroscience, 14(2), 143–152. 668
doi:10.1038/nrn3403 669
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a 670
multidimensional approach. Journal of Personality and Social Psychology, 671
44(1), 113–126. 672

18
Domschke, K., Stevens, S., Pfleiderer, B., & Gerlach, A. L. (2010). Interoceptive 673
sensitivity in anxiety and anxiety disorders: an overview and integration of 674
neurobiological findings. Clinical Psychology Review, 30(1), 1–11. 675
doi:10.1016/j.cpr.2009.08.008 676
Dunn, B. D., Galton, H. C., Morgan, R., Evans, D., Oliver, C., Meyer, M., … 677
Dalgleish, T. (2010). Listening to your heart. How interoception shapes emotion 678
experience and intuitive decision making. Psychological Science, 21(12), 1835–679
44. doi:10.1177/0956797610389191 680
Ehlers, A., Breuer, P., Dohn, D., & Fiegenbaum, W. (1995). Heartbeat perception and 681
panic disorder: possible explanations for discrepant findings. Behaviour 682
Research and Therapy, 33(I), 69–76. 683
Ernst, J., Northoff, G., Böker, H., Seifritz, E., & Grimm, S. (2013). Interoceptive 684
awareness enhances neural activity during empathy. Human Brain Mapping, 685
34(7), 1615–24. doi:10.1002/hbm.22014 686
Fairclough, S. H., & Goodwin, L. (2007). The effect of psychological stress and 687
relaxation on interoceptive accuracy: Implications for symptom perception. 688
Journal of Psychosomatic Research, 62(3), 289–95. 689
doi:10.1016/j.jpsychores.2006.10.017 690
Farrer, C., Franck, N., Georgieff, N., Frith, C. D., Decety, J., & Jeannerod, M. (2003). 691
Modulating the experience of agency: a positron emission tomography study. 692
NeuroImage, 18(2), 324–333. doi:10.1016/S1053-8119(02)00041-1 693
Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews. 694
Neuroscience, 11(2), 127–38. doi:10.1038/nrn2787 695
Friston, K., Mattout, J., & Kilner, J. (2011). Action understanding and active 696
inference. Biological Cybernetics, 104, 137–60. doi:10.1007/s00422-011-0424-z 697
Fukushima, H., Terasawa, Y., & Umeda, S. (2011). Association between 698
interoception and empathy: evidence from heartbeat-evoked brain potential. 699
International Journal of Psychophysiology, 79(2), 259–65. 700
doi:10.1016/j.ijpsycho.2010.10.015 701
Garfinkel, S. N., & Critchley, H. D. (2013). Interoception, emotion and brain: new 702
insights link internal physiology to social behaviour. Commentary on: “Anterior 703
insular cortex mediates bodily sensibility and social anxiety” by Terasawa et al. 704
(2012). Social Cognitive and Affective Neuroscience, 8(3), 231–4. 705
doi:10.1093/scan/nss140 706
Gillmeister, H., Catmur, C., Brass, M., & Heyes, C. (2008). Experience-based 707
priming of body parts: A study of action imitation. Brain Research, 1217, 157–708
170. 709
Halasz, V., & Cunnington, R. (2012). Unconscious effects of action on perception. 710
Brain Science, 2, 130–146. 711

19
Hayes, D. J., & Northoff, G. (2012). Common brain activations for painful and 712
nonpainful aversive stimuli. BMC Neuroscience, 13(1), 60. doi:10.1186/1471-713
2202-13-60 714
Herbert, B. M., Muth, E. R., Pollatos, O., & Herbert, C. (2012). Interoception across 715
modalities: On the relationship between cardiac awareness and the sensitivity for 716
gastric functions. PloS One, 7(5), e36646. doi:10.1371/journal.pone.0036646 717
Herbert, B. M., Pollatos, O., Flor, H., Enck, P., & Schandry, R. (2010). Cardiac 718
awareness and autonomic cardiac reactivity during emotional picture viewing 719
and mental stress. Psychophysiology, 47(2), 342–54. doi:10.1111/j.1469-720
8986.2009.00931.x 721
Heyes, C. (2010). Automatic imitation. Psychological Bulletin, 44, 1–88. 722
Heyes, C., & Bird, G. (2007). Mirroring, association, and the correspondence 723
problem. In K. M. Haggard P, Rossetti Y (Ed.), Sensorimotor Foundations of 724
Higher Cognition, Attention and Performance (pp. 461–480). Oxford University 725
Press. 726
Heyes, C., Bird, G., Johnson, H., & Haggard, P. (2005). Experience modulates 727
automatic imitation. Cognitive Brain Research, 22(2), 233–40. 728
doi:10.1016/j.cogbrainres.2004.09.009 729
Hommel, B. (2009). Action control according to TEC (theory of event coding). 730
Psychological Research, 73(4), 512–526. doi:10.1007/s00426-009-0234-2 731
Hommel, B., Müsseler, J., Aschersleben, G., & Prinz, W. (2001). The Theory of 732
Event Coding (TEC): a framework for perception and action planning. The 733
Behavioral and Brain Sciences, 24(5), 849–78. 734
James, W. (1890). The Principles of Psychology. New York: Henry Holt & Co. 735
Jennings, R., Berg, K., Hutcheson, J., Obrist, P., Porges, S., & Turpin, G. (1981). 736
Publication guidelines for heart rate studies in man. Psychophysiology, 18(3), 737
227–231. 738
Kaplan, J. T., & Iacoboni, M. (2006). Getting a grip on other minds: mirror neurons, 739
intention understanding, and cognitive empathy. Social Neuroscience, 1(3-4), 740
175–83. doi:10.1080/17470910600985605 741
Katkin, E. S., Wiens, S., & Ohman, A. (2001). Nonconscious fear conditioning, 742
visceral perception, and the development of gut feelings. Psychological Science, 743
12(5), 366–70. 744
Knapp-Kline, K., & Kline, J. P. (2005). Heart rate, heart rate variability, and heartbeat 745
detection with the method of constant stimuli: slow and steady wins the race. 746
Biological Psychology, 69(3), 387–396. doi:10.1016/j.biopsycho.2004.09.002 747

20
Knoll, J. K., & Hodapp, V. (1992). A comparison between two methods for assessing 748
heartbeat perception. Psychophysiology, 29(2), 218–222. 749
Lamm, C., Batson, C. D., & Decety, J. (2007). The neural substrate of human 750
empathy: effects of perspective-taking and cognitive appraisal. Journal of 751
Cognitive Neuroscience, 19(1), 42–58. doi:10.1162/jocn.2007.19.1.42 752
Leighton, J., Bird, G., Orsini, C., & Heyes, C. (2010). Social attitudes modulate 753
automatic imitation. Journal of Experimental Social Psychology, 46(6), 905–910. 754
doi:10.1016/j.jesp.2010.07.001 755
Longo, M. R., & Bertenthal, B. I. (2009). Attention modulates the specificity of 756
automatic imitation to human actors. Experimental Brain Research, 192(4), 739–757
44. doi:10.1007/s00221-008-1649-5 758
Maister, L., Tsiakkas, E., & Tsakiris, M. (2013). I feel your fear: Shared touch 759
between faces facilitates recognition of fearful facial expressions. Emotion, 760
13(1), 7–13. doi:http://psycnet.apa.org/doi/10.1037/a0030884 761
Matthias, E., Schandry, R., Duschek, S., & Pollatos, O. (2009). On the relationship 762
between interoceptive awareness and the attentional processing of visual stimuli. 763
International Journal of Psychophysiology, 72(2), 154–9. 764
doi:10.1016/j.ijpsycho.2008.12.001 765
Miller, J., & Diego, S. (1991). Reaction time analysis with outlier exclusion": Bias 766
varies with sample size. The Quarterly Journal of Experimental Psychology 767
Section A, 43(4), 37–41. 768
Montoya, P., Schandry, R., & Müller, A. (1993). Heartbeat evoked potentials (HEP): 769
topography and influence of cardiac awareness and focus of attention. 770
Electroencephalography and Clinical Neurophysiology, 88(3), 163–72. 771
Mussgay, L., Klinkenberg, N., & Rüddel, H. (1999). Heartbeat perception in patients 772
with depressive, somatoform, and personality disorders. Journal of 773
Psychophysiology, 13, 27–36. 774
Obhi, S. S., Hogeveen, J., Giacomin, M., & Jordan, C. H. (2013). Automatic Imitation 775
Is Reduced in Narcissists. Journal of Experimental Psychology. Human 776
Perception and Performance. doi:10.1037/a0034056 777
Obhi, S. S., Hogeveen, J., & Pascual-Leone, A. (2011). Resonating with others: the 778
effects of self-construal type on motor cortical output. The Journal of 779
Neuroscience, 31(41), 14531–14535. doi:10.1523/JNEUROSCI.3186-11.2011 780
Panksepp, J., & Northoff, G. (2009). The trans-species core SELF: the emergence of 781
active cultural and neuro-ecological agents through self-related processing within 782
subcortical-cortical midline networks. Consciousness and Cognition, 18(1), 193–783
215. doi:10.1016/j.concog.2008.03.002 784

21
Philippi, C. L., Feinstein, J. S., Khalsa, S. S., Damasio, A. R., Tranel, D., Landini, G., 785
… Rudrauf, D. (2012). Preserved self-awareness following extensive bilateral 786
brain damage to the insula, anterior cingulate, and medial prefrontal cortices. 787
PloS One, 7(8), e38413. doi:10.1371/journal.pone.0038413 788
Pollatos, O., Füstös, J., & Critchley, H. D. (2012). On the generalised embodiment of 789
pain: how interoceptive sensitivity modulates cutaneous pain perception. Pain, 790
153(8), 1680–6. doi:10.1016/j.pain.2012.04.030 791
Pollatos, O., & Schandry, R. (2004). Accuracy of heartbeat perception is reflected in 792
the amplitude of the heartbeat-evoked brain potential. Psychophysiology, 41(3), 793
476–482. doi:10.1111/1469-8986.2004.00170.x 794
Preston, S. D., & Hofelich, A. J. (2012). The Many Faces of Empathy: Parsing 795
Empathic Phenomena through a Proximate, Dynamic-Systems View of 796
Representing the Other in the Self. Emotion Review, 4(1), 24–33. 797
doi:10.1177/1754073911421378 798
Schandry, R. (1981). Heartbeat perception and emotional experience. 799
Psychophysiology, 18(4), 483–488. 800
Schutz-Bosbach, S., Mancini, B., Aglioti, S. M., & Haggard, P. (2006). Self and Other 801
in the Human Motor System. Current Biology, 16, 1830–1834. 802
doi:10.1016/j.cub.2006.07.048 803
Schütz-Bosbach, S., & Prinz, W. (2007). Perceptual resonance: action-induced 804
modulation of perception. Trends in Cognitive Sciences, 11(8), 349–55. 805
doi:10.1016/j.tics.2007.06.005 806
Seth, A. (2013). Interoceptive inference, emotion, and the embodied self. Trends in 807
Cognitive Sciences, 1–9. doi:10.1016/j.tics.2013.09.007 808
Seth, A., Suzuki, K., & Critchley, H. D. (2011). An interoceptive predictive coding 809
model of conscious presence. Frontiers in Psychology, 2, 395. 810
doi:10.3389/fpsyg.2011.00395 811
Singer, T., Critchley, H. D., & Preuschoff, K. (2009). A common role of insula in 812
feelings, empathy and uncertainty. Trends in Cognitive Sciences, 13(8), 334–40. 813
doi:10.1016/j.tics.2009.05.001 814
Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R. J., & Frith, C. D. (2004). 815
Empathy for pain involves the affective but not sensory components of pain. 816
Science, 303(5661), 1157–62. doi:10.1126/science.1093535 817
Snyder, M. (1974). Self-monitoring of expressive behavior. Journal of Personality 818
and Social Psychology, 30(4), 526–537. doi:10.1037/h0037039 819
Spengler, S., Brass, M., Kühn, S., & Schütz-Bosbach, S. (2010). Minimizing motor 820
mimicry by myself: Self-focus enhances online action-control mechanisms 821

22
during motor contagion. Consciousness and Cognition, 19, 98–106. 822
doi:10.1016/j.concog.2009.12.014 823
Spengler, S., von Cramon, D., & Brass, M. (2009). Control of shared representations 824
relies on key processes involved in mental state attribution. Human Brain 825
Mapping, 30(11), 3704–18. doi:10.1002/hbm.20800 826
Stevens, S., Gerlach, A. L., Cludius, B., Silkens, A., Craske, M. G., & Hermann, C. 827
(2011). Heartbeat perception in social anxiety before and during speech 828
anticipation. Behaviour Research and Therapy, 49(2), 138–43. 829
doi:10.1016/j.brat.2010.11.009 830
Suzuki, K., Garfinkel, S. N., Critchley, H. D., & Seth, A. K. (2013). Multisensory 831
integration across exteroceptive and interoceptive domains modulates self-832
experience in the rubber-hand illusion. Neuropsychologia, 1–9. 833
doi:10.1016/j.neuropsychologia.2013.08.014 834
Tajadura-Jiménez, A., & Tsakiris, M. (2013). Balancing the “Inner” and the “Outer” 835
Self: Interoceptive Sensitivity Modulates Self-Other Boundaries. Journal of 836
Experimental Psychology. General, 44(0). doi:10.1037/a0033171 837
Terasawa, Y., Shibata, M., Moriguchi, Y., & Umeda, S. (2013). Anterior insular 838
cortex mediates bodily sensibility and social anxiety. Social Cognitive and 839
Affective Neuroscience, 8, 259–266. doi:10.1093/scan/nss108 840
Tsakiris, M. (2013). Self-Specificity of the External Body. Neuropsychoanalysis, 841
15(1), 66–69. 842
Tsakiris, M., Tajadura-Jiménez, A., & Costantini, M. (2011). Just a heartbeat away 843
from one’s body: interoceptive sensitivity predicts malleability of body-844
representations. Proceedings of the Royal Society B: Biological Sciences, 278, 845
2470–6. doi:10.1098/rspb.2010.2547 846
Verdejo-Garcia, A., Clark, L., & Dunn, B. D. (2012). The role of interoception in 847
addiction: A critical review. Neuroscience and Biobehavioral Reviews, 36(8), 848
1857–1869. doi:10.1016/j.neubiorev.2012.05.007 849
Wang, Y., & Hamilton, A. (2012). Social top-down response modulation (STORM): a 850
model of the control of mimicry in social interaction. Frontiers in Human 851
Neuroscience, 6, 153. doi:10.3389/fnhum.2012.00153 852
Wang, Y., Newport, R., & Hamilton, A. (2011). Eye contact enhances mimicry of 853
intransitive hand movements. Biology Letters, 7(1), 7–10. 854
doi:10.1098/rsbl.2010.0279 855
Wang, Y., Ramsey, R., & Hamilton, A. F. (2011). The control of mimicry by eye 856
contact is mediated by medial prefrontal cortex. The Journal of Neuroscience, 857
31(33), 12001–12010. doi:10.1523/JNEUROSCI.0845-11.2011 858

23
Werner, N. S., Kerschreiter, R., Kindermann, N. K., & Duschek, S. (2013). 859
Interoceptive Awareness as a Moderator of Affective Responses to Social 860
Exclusion. Journal of Psychophysiology, 27(1), 39–50. doi:10.1027/0269-861
8803/a000086 862
Whitehead, W. E., & Drescher, V. M. (1980). Perception of gastric contractions and 863
self-control of gastric motility. Psychophysiology, 17(6), 552–558. 864
doi:10.1111/j.1469-8986.1980.tb02296.x 865
Wicker, B., Keysers, C., Plailly, J., Royet, J. P., Gallese, V., & Rizzolatti, G. (2003). 866
Both of us disgusted in My insula: the common neural basis of seeing and feeling 867
disgust. Neuron, 40(3), 655–64. 868
Wiebking, C., Duncan, N. W., Tiret, B., Hayes, D. J., Marjańska, M., Doyon, J., … 869
Northoff, G. (2013). GABA in the insula - a predictor of the neural response to 870
interoceptive awareness. NeuroImage. doi:10.1016/j.neuroimage.2013.04.042 871
Zaki, J., Davis, J. I., & Ochsner, K. N. (2012). Overlapping activity in anterior insula 872
during interoception and emotional experience. NeuroImage, 62(1), 493–9. 873
doi:10.1016/j.neuroimage.2012.05.012 874
Zimmerman, P., & Frimm, B. (2002). Test of attentional performance (TAP). Version 875
2.3. www.psytest.net. 876
877