Self-body recognition through a mirror: Easing spatial-consistency requirements for rubber hand illusion

Typical experiments to induce the rubber hand illusion (RHI) require experimental participants to gaze at a fake hand while tactile stimuli are provided to both the fake and hidden actual hands in a synchronous manner. However, under such conditions, postural and apparent disagreement between a seen fake hand and hidden actual hand prevents illusory body ownership. Provided that humans recognize mirror images as copies of the real world in spite of their spatial uncertainties or incongruence, the sensory disagreement may be accepted in RHI settings if using a mirror to show a fake hand. The present study performed two experiments to reveal how self-body recognition of a fake hand feature via mirror affects the RHI. These experiments were conducted in an RHI environment involving voluntary hand movements to investigate not only body ownership but also agency. The first experiment (Experiment 1) examined whether illusory ownership of a fake hand seen in a mirror could be induced. Then, we examined whether the RHI using a mirror image allows disagreement in orientation between the rubber and actual hands (Experiment 2). Subjective evaluations using a RHI questionnaire demonstrated that evoked embodiment of the rubber hand was stronger in the presence of a mirror than in the absence of it (Experiment 1) and that participants experienced the RHI even if the actual and rubber hands were incongruent in terms of orientation (45 °; Experiment 2). No significant difference was found in the change of perceived finger location (proprioceptive drift) between these experiments. These findings suggest that the use of a mirror masks subtle spatial incongruency or degrades the contribution of visual cues for spatial recognition and facilitates multisensory integration for bodily illusions.


Introduction
between the actual and fake hands lead to more intense illusions of body ownership. 8 For instance, when the fake and actual hands are spatially incongruent (e.g., when 9 they have different postures), the body ownership illusion is weakened [4][5][6][7][8]. 10 According to Constantini and Haggard, a hand angle mismatch of 20 o between the 11 veridical and fake hands distracts from the body ownership illusion [6]. Nevertheless, 12 this angle value varies among studies [8,9]. 13 Usually in experiments on the RHI paradigm, participants are asked to gaze at the 14 fake hand because "RHI is understood as a visual adaptation of proprioceptive 15 position" [5]. In contrast, in the present study, they gazed at the fake hand through a 16 mirror. Under these settings, the RHI is experienced [10][11][12]. Despite the reversed 17 orientation and uncertainty of geometrical information including depth, humans 18 recognize a mirror image as a copy of the real environment. In other words, humans 19 accept the low reliability of the image in a mirror. This tolerance of less reliable 20 images may also hold in RHI settings, and it is possible that incongruency between the 21 fake and actual hands do not disturb the illusory experience of body ownership. One 22 study supports this possibility; in it, the body ownership illusion was evoked 23 irrespective of the egocentric or allocentric fake hand images in a mirror [11]. The 24 present study pursues these hypothetical propositions. 25 In Experiment 1, we test whether the body ownership illusion is elicited when 26 experimental participants gaze at the mirror image of a fake hand. This experiment is 27 not merely a replication of earlier ones [10,11] as we tested that our settings involving 28 self-generated hand motions and tactile stimuli cause the illusory experiences. 29 Experiment 2 investigates whether the incongruency of the fake and actual hands' 30 postures (angles) is accepted in the RHI setting involving a mirror. Although Kontaris 31 and Downing [11] found that first-person and third-person perspective images of a fake 32 hand in a mirror did not differ in terms of the subjective experience of the 33 body-ownership illusion, they did not deal with situations where the actual and fake 34 hands were angled. Exp. 1 and Exp. 2 were conducted in active RHI settings to 35 investigate the body ownership and agency. Although these two senses can be 36 distinguished during involuntary hand movements, they are strongly related during 37 voluntary hand movements that enhance the precision of proprioceptive cues [5]. In 38 our active RHI settings, the effect of mirror and congruency of hand posture on body 39 ownership and agency can be investigated. 40 Although studies on the body ownership of phantom limbs have commonly 41 employed a mirror (e.g., [13]), those on the RHI have thus far rarely experimented 42 with mirrors. For instance, in the framework of the RHI, Holmes et al. used a mirror 43 to investigate how position bias caused by a mirror affects the reaching task of an 44 unseen hand [14]. Additionally, Preston et al. examined illusory ownership over a 45 whole-body mannequin seen in a mirror [15]. RHI settings using a mirror may share 46 some traits with body ownership illusions when facing an avatar image in a virtual 47 environment [16]. In two previous studies [15,16], whole-body ownership illusion was 48 induced by using a mirror or images from a third-person view.

51
Fifteen paid university students participated in the experiments (9 males and 6 52 females; mean and standard deviation of age in years: 24 and 9.2) with written 53 informed consent. The participants were recruited using a local advertisement poster.  Exp. 2 such that participants did not realize the expected effects of using mirror 60 images. Hence, in Exp. 2, participants could be more familiar with the process 61 including how they should move their hands to effectively cause the illusory experience. 62

63
This study was conducted with the approval of the internal review board, School of 64 Engineering, Nagoya University (♯17-12).

Apparatus for inducing the RHI under self-generated movement 66
As shown in Fig 1(a, b) and Fig 2(a, b, c), the main components of the apparatus 67 were a fake rubber hand, hand gloves, a cuboid frame, and a cloth for blindfolding 68 that occluded each participant's right shoulder, arm, and hand. The fake rubber hand 69 was composed of a rubber glove containing wires and cotton. The same rubber glove 70 was worn by participants. Two acrylic rods were fixed to the fake hand as in Fig 2(b).

71
Participants held these rods with the thumbs and palms of right hand and moved the hands in the same manner as the fake and right hands. When the acrylic rods were 74 removed from the fake hand, the fake hand remained still on the desktop, and an 75 asynchronous condition was configured in which the actual and fake hands were 76 independent. A similar setup was used in [17].  The mirror was located facing the participant at a height in the range of 15-25 cm.

79
The distance between the participant and the mirror was 50-60 cm. Within these 80 height and distance combinations, the spatial configuration was adjusted such that  In order for participants to familiarize themselves with the experimental setup, before illusion [18][19][20][21][22]. Each experimental condition was tested in a randomized order in a 105 single session, and two sessions were performed for individual participants.

106
Participants were instructed to gaze at the fake hand directly in without-mirror 107 conditions and through a mirror in with-mirror conditions.  No. Condition 4 Synchronous without a mirror 5 Synchronous with a mirror 6 Asynchronous without a mirror 7 Asynchronous with a mirror  125 We investigated how spatial incongruency between the actual and fake hands 126 influenced illusory body ownership. Table 2 shows the four conditions tested in 127 Experiment 2. They included the combination of two factors: synchronization 128 (synchronous or asynchronous) and the mirror (presence or absence). In all conditions, 129 actual hand was tilted at 45 o against the fake hand (Fig 3(b)). Under conditions 4 and 130 5, the motions of actual and fake hands were synchronized whereas they were not The magnitude of an illusion is often assessed by the questionnaire proposed by 138 Botvinick and Cohen [1]. In addition, we added questionnaire items regarding 139 agency [23,24] and modified some statements to fit our RHI setting. After each trial, 140 participants responded to the six questionnaire items shown in Table 3. These items 141 were provided in plain English. Each item was rated by using a seven-point scale (-3: I 142 disagree with the statement, to +3: I agree with the statement). Q1 (body ownership) 143 and Q2 (agency) concerned the illusory experience of the RHI whereas the others were 144 prepared as controls for the body ownership statement. positions of the middle finger in a parallel direction (Fig 3(c)). This measurement 154 criterion was same when the actual right hand was tilted 45 o (Fig 3(d)). PD was     and Preston [12], which investigated the mirror RHI. Other related studies have 221 reported body ownership through a mirror at the same extent [10] or weaker [11] as 222 compared to the condition without a mirror. However, these studies investigated 223 passive RHI, in which the experimental conditions differed from our study and that of 224 study of Jenkinson and Preston [12] in terms of involvements of hand movements.

225
In Experiment 2, ownership over the fake hand was observed even when the fake for 'mirror-asynchronous' condition. Humans are likely to misunderstand their body 241 motion seen through a mirror [26], and transformation into a third-person view 242 decreases the precision of visual and motor perception [27]. Many studies have 243 discussed the differences between first-person and third-person views, and suggested 244 their effects on the self-body recognition [10,12,28].

245
The enhancement of the illusion by a mirror could be explained by two possible 246 reasons if self-body recognition is obscured by a mirror. First, the mirror decreases the 247 reliability and precision of visual cues and masks the inconsistency of perceived 248 signals. Provided that the perceived hand is determined by a multisensory integration 249 processes of sensory cues [29][30][31], noisy or imprecise images in a mirror may alter the 250 weightings of unisensory signals for multisensory integration [12]. For instance, in a 251 visual-haptic integration task, a decrease in the reliability of visual cues led to the 252 relative increase in the contribution of haptic cues [32][33][34]. In particular, body 253 ownership is thought to be induced in the process of inferring the environment from 254 perceived signals [35]. Therefore, illusory body ownership would be influenced by the 255 change of weightings of multisensory cues. In our experimental settings, visual, tactile, 256 and proprioceptive cues were involved, and the use of a mirror degraded reliability 257 regarding the visually provided hand posture. The weightings of sensory cues would 258 be different between conditions with and without a mirror. Such an effect of less 259 reliable images in a mirror was pointed out in [11], in which even laterality was 260 masked by using mirror images of a fake hand. In other words, in classical RHI 261 settings where the fake hand is gazed at directly, because of the greater contribution of 262 visual cues, visible incongruency may cause the visual sense to fail to capture other 263 cues and prevent the occurrence of the illusion.

264
The second reason is that visual cues are weakened by a mirror and this accelerates 265 multisensory integration. The body-ownership illusion is considered to be caused by 266 visuo-tactile integration, and a greater degree of integration would produce a more 267 intense illusion. Sensory integration under RHI settings has been demonstrated in 268 terms of spatial and temporal accordance [3][4][5]. In addition, multisensory integration 269 is enhanced when the stimuli to individual sensory channels are weak [36,37]. These 270 three aspects are known as the spatial rule, temporal rule, and inverse 271 effectiveness [38], which was studied neurophysiologically from the perspective of the 272 activity of superior colliculus cells [36] and inter-trial phase coherency of scalp EEG 273 (electroencephalogram) during RHI tasks [39,40]. In our experiment, the use of a 274 mirror might have weakened the visual stimuli and could be related to the principle of 275 inverse effectiveness, and multisensory integration might have been enhanced in 276 comparison with using the settings without a mirror.

278
This study investigated the role of gazing at a fake hand in a mirror under RHI 279 conditions. The subjective experience was assessed based on a questionnaire. The 280 illusory experiences of body ownership and agency were evoked when the fake hand in 281 a mirror was gazed at (Experiment 1). The magnitudes of subjective experiences were 282 higher (body ownership) or equally high (agency) compared with the condition where 283 the fake hand was directly seen without a mirror. Furthermore, using the mirror, the 284 illusory experience was reported even when the actual and fake hands were placed at 285 incongruent postures (Experiment 2). Although the underlying mechanism of these 286 observations is unknown, gazing at a fake hand in a mirror offers a promising 287 approach for robustly inducing intense RHI experiences.