Content uploaded by Giovanni B Caputo
Author content
All content in this area was uploaded by Giovanni B Caputo on Nov 14, 2020
Content may be subject to copyright.
Short Communication
Strange-face illusions during inter-subjective gazing
Giovanni B. Caputo
DIPSU, University of Urbino, via Saffi 15, 61029 Urbino, Italy
article info
Article history:
Received 16 February 2012
Available online 13 September 2012
Keywords:
Dissociative-identity
Embodiment
Empathy
Joint-action
Carl G. Jung
Mirror
Numinosity
Out-of-body experience (OBE)
Self
Synchronicity
abstract
In normal observers, gazing at one’s own face in the mirror for a few minutes, at a low
illumination level, triggers the perception of strange faces, a new visual illusion that has
been named ‘strange-face in the mirror’. Individuals see huge distortions of their own faces,
but they often see monstrous beings, archetypal faces, faces of relatives and deceased, and
animals. In the experiment described here, strange-face illusions were perceived when two
individuals, in a dimly lit room, gazed at each other in the face. Inter-subjective gazing
compared to mirror-gazing produced a higher number of different strange-faces. Inter-sub-
jective strange-face illusions were always dissociative of the subject’s self and supported
moderate feeling of their reality, indicating a temporary lost of self-agency. Unconscious
synchronization of event-related responses to illusions was found between members in
some pairs. Synchrony of illusions may indicate that unconscious response–coordination
is caused by the illusion–conjunction of crossed dissociative strange-faces, which are per-
ceived as projections into each other’s visual face of reciprocal embodied representations
within the pair. Inter-subjective strange-face illusions may be explained by the subject’s
embodied representations (somaesthetic, kinaesthetic and motor facial pattern) and the
other’s visual face binding. Unconscious facial mimicry may promote inter-subjective
illusion–conjunction, then unconscious joint-action and response–coordination.
Ó2012 Elsevier Inc. All rights reserved.
1. Introduction
Gazing at one’s own reflected face in a mirror, at a low illumination level, reportedly leads to experiences termed ‘strange-
face in the mirror’ illusions (Caputo, 2010a). In a study set-up under controlled laboratory conditions, all of the 50 healthy
young adults (mean age 23 years) began to perceive strange-face illusions after about 1 min. When a 10 min session of mir-
ror gazing ended, participants experienced striking effects such as huge deformations of one’s own face (reported by 66% of
individuals); a relative’s face with the changed features (18%), of whom 8% were still alive and 10% were deceased; an un-
known person (28%); an archetypal face, such as a numinous child, a young androgyne, a very old woman, an ancestor or a
shaman (28%); an animal face such as a cat or a pig (18%); and monstrous beings such as a witch or a skull (48%).
Previous clinical studies used mirrors in bereavement therapy. Research with the psychomanteum showed that individ-
uals experienced hallucinations when a mirror was displayed in a darkened room, with black opaque curtain cloth on walls,
and the mirror reflected the space above while the seated individuals did not see their reflected face or body (Terhune &
Smith, 2006). Experiences with the psychomanteum are hallucinations of dialogue, sounds, light, bodily sensations and
smell, in addition to visual imagery in the mirror. On the contrary, strange-faces during mirror gazing are likely to be clas-
sified as illusions because they have a distinctive visual character and are specific of the dysmorphic perceptions of one’s
own face. Another difference is that strange-faces are produced by a visual stimulus, whereas hallucinations can happen
when physical stimulus is absent. For instance, hallucinations are regarded as false-positive responses in psychophysical
tasks (Bentall, 1990).
1053-8100/$ - see front matter Ó2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.concog.2012.08.007
E-mail address: giovanni.caputo@uniurb.it
Consciousness and Cognition 22 (2013) 324–329
Contents lists available at SciVerse ScienceDirect
Consciousness and Cognition
journal homepage: www.elsevier.com/locate/concog
Strange-face illusions may be similar to some out-of-body experiences that have been studied experimentally in healthy
individuals, who perceive an illusory own body in extracorporeal space (Lenggenhager, Tadi, Metzinger, & Blanke, 2007;
Sforza & Blanke, 2012). However, experimental out-of-body experiences involve illusory self-identification with the double,
whereas strange-face illusions are always dissociative of the self in healthy individuals (Caputo, 2010b). Instead, in schizo-
phrenic patients, identification with strange-faces during mirror-gazing can be frequent, probably due to psychotic ego
dysfunction and misattribution of agency (Caputo et al., 2012).
The purpose of the present study is to investigate whether strange-face illusions were perceived when two individuals, in
a dimly lit room, gazed at each other in the face. In preliminary trials, I found that each participant of the experimental pair
described spontaneously illusions of deformed faces and strange-faces of the other participant, even if they had not been told
in the initial instructions that this may happen.
Strange-face illusions that the subject perceives in the mirror are seen into the other’s metamorphosed face during inter-
subjective gazing. This phenomenon would show that individuals share strange-face illusions within a pair, hence it would
shed light on the process of ‘projection’ of unconscious mental contents between individuals, as found in dynamic psychol-
ogy (Jung, 1977). Moreover, inter-subjectivity could emerge through synchronization of strange-face illusions between
individuals. This finding would provide a basis for scientific investigations of meditation techniques known in alchemy
and magic as two-vase opus (Ur group, 2001) that has parallel traditions in Buddhism and Taoism.
2. Methods
Ten adults participated in the experiment. They were naïve observers, unaware of the aim of the research; no psychiatric
disorders; mean age 27.3 years, s.d. 8.6; two men and eight women. Two pairs were formed of a man and a woman; others
pairs were formed by two women.
The experiment was conducted in a 4 m 4 m white room, quiet and obscured from external light. Two chairs were posi-
tioned around the center of the room, facing each other. The two participants were seated so as to maintain a distance of
1.0 m between faces. The space left empty between the chairs was 0.4 m wide and it was covered with a flat opaque panel.
A halogen lamp (Osram 12V, 10W) produced a fixed beam illumination. The spotlight was mounted under the flat panel, at
the very center, halfway between the two chairs. The spotlight beam pointed toward the floor in such a way to produce only
indirect, diffuse and relatively uniform lighting within the room. Illumination of faces was approx. 0.2 cd m
2
, measured by a
digital photometer with a wide-angle sensor (Pantec by Carlo Gavazzi, LM-20). This level of illumination allowed detailed
perception of fine face traits but attenuated color perception.
Instructions for the experimental task were the following: ‘‘You should maintain a neutral facial expression. Your task is to
look at the other participant; you should keep gazing into the eyes of the other participant. You may or may not perceive changes in
the face of the other participant. Press the button once you perceive a change and keep it pressed for as long as you perceive the
change. If you don’t perceive any change, then don’t press the button’’. The session began with a few minutes of light adaptation;
then a 10-min inter-subjective gazing.
Dependent variables of the experiment were the following: the event-related-responses, three Likert-type scales and the
phenomenological description. Event-related-responses were recorded through two response switches and digitally stored.
Each member of a pair had a response switch in one hand, covered by the other hand, with hands on legs, in order to com-
pletely hide finger movement. The frequency of event-related responses was defined by the number of times observers
pressed the response button, averaged per minute. The mean duration was the mean time they held down the response
button.
Time-series analysis was run on durations of illusions in each participant. Autocorrelation functions were calculated
between the values of the series of durations of illusions with the values lagged from 1 to 7: the illusion prior to the current
illusion is a lag of 1, two illusions prior to the current illusion is a lag of 2, etc. In the autocorrelation function, the difference-
transformation parameter was set to 1; this value removes linear trend. Statistical test of periodicity, within the series of
durations, used the null hypothesis that variations of durations were produced by white noise intrinsic to the process.
Statistical significance was calculated through standard errors.
Inter-subjective synchronization between event-related responses of the members of each pair was measured. Synchro-
nization of event-related responses was defined as follows: if, during a response by one member, the other member also
starts to respond, the responses by the pair were classified as synchronous. In other words, at a certain time both members
were pressing their respective buttons together, unaware of the other’s action. Moreover, the time-lag between the onsets of
two synchronous responses was measured. The percentage of synchronous responses of each participant was calculated over
the total number of his responses.
Self-evaluation was made on three Likert-type scales. Sentences of Likert-type scales were the following: ‘How often did
you notice anything strange?’, ‘How often did it seem real?’, ‘How often did you see another person?’. Responses given were
scored on a five-point Likert-type scale, ranging from ‘never’ (= 0), ‘rarely’ (= 1), to ‘very often’ (= 4).
Participants were asked to give a description of their phenomenological experience (‘What did you see?’). The descriptions
were classified according to categories found in previous studies (Caputo, 2010a). For each participant, the total number of
different strange-faces was counted.
G.B. Caputo/ Consciousness and Cognition 22 (2013) 324–329 325
Control trials were run in two pairs. A fixation dot was placed half-height on the chests of members. The task of each indi-
vidual was to gaze at the dot placed on the chest of the other individual and to respond to changes perceived in the body
around the dot. The same aforementioned experimental instructions were used. Results showed that illusory changes were
not perceived. Therefore, inter-subjective strange-face illusions arose specifically when gazing at each other face.
3. Results
Results showed that all participants perceived strange-face illusions whilst inter-subjective gazing. Hereunder, all data
are expressed as means ± SEM. Event-related-responses showed mean onset of first illusion 30.7 ± 7.9 s, mean frequency
2.4 ± .4 illusions per min, mean duration 4.5 ± .6 s.
Regression analysis of illusion durations in function of time did not show univocal results: for three participants the linear
trend was positive (mean .5 ± .05 s min
1
;p< .03); for another participant the linear trend was negative (.18 s min
1
;
p< .02); for six participants there was not significant linear trend.
Time-series analysis was performed on durations of illusions in eight participants (two participants were excluded due to
insufficient number of responses to carry time-series analysis). Statistically significant (p< .03) periodicity of lag 1 was found
on seven participants. It was caused by a negative correlation between the duration of an illusion and the duration of the
illusion that had previously occurred. Fig. 1 plots the autocorrelation functions of a representative pair of participants (their
frequencies of illusions were 3.3 and 3.1 illusions per min, respectively).
Inter-subjective synchronization between event-related responses of each pair showed that percentages of synchronous
responses were in 0–84% range among participants; mean synchrony across all participants was 39% ± 11. The binomial test
1
indicated that synchronization was statistically significant (p< .04) in five participants (50% of participants). In the other two
pairs, synchrony was probably absent because one member had a very low number of responses. The distribution of time-lags
of synchronous responses across all pairs was symmetric (skewness .2 ± .3); the mean of absolute values of time-lags was
2.6 ± .3 s. In follow-up interviews, it was found that pairs who responded in synchrony were completely unaware of the event.
For example, the members of the aforementioned pair had highest synchrony (84% and 76%, respectively) of all participants, but
had guessed to have responded together only once during the entire session.
Likert-type scales showed that the mean score was 2.5 ± .4 for the sentence ‘How often did you notice anything strange?’,
2.1 ± .3 for the sentence ‘How often did it seem real?’, and 3.0 ± .3 for the sentence ‘How often did you see another person?’. Cor-
relations between Likert-type scales showed that responses to the first and the third sentences were significantly correlated
(Pearson r= .83; p< .01).
Classification of phenomenological descriptions showed that illusions were deformed traits (reported by 9 out of 10 par-
ticipants), parents (2 out of 10), unknown persons (6 out of 10), archetypes (5 out of 10), animals (1 of 10) and monstrous
beings (8 out of 10). The mean total number of different strange-faces was 4.9 ± .7 per 10 min session. There were no par-
ticipants who identified themselves with strange-faces.
In order to make comparisons with previous works (Caputo, 2010a, 2010b), between-subject ANOVAs (2-level factor: in-
ter-subjective-gazing vs. mirror-gazing) were performed. Onset, frequency and duration of event-related responses were not
statistically different in inter-subjective gazing and mirror gazing.
2
The total number of different strange-faces was higher in
inter-subjective gazing than in mirror-gazing
3
(4.9 ± .7 vs. 2.6 ± .1; F(1, 58) = 31.8; p< .001;
g
2
= .354).
4. Discussion
Overall, the experimental findings show that inter-subjective gazing is a valuable technique to produce illusions of
strange-faces. Number of strange-faces increased in inter-subjective gazing with respect to mirror-gazing. Instead, psycho-
physical duration and frequency of illusions did not change.
The negative correlation found in time-series analysis between the durations of an illusion and the previous occurring
illusion indicates that strange-face illusions are under a controlling regulatory mechanism. This control may be due to a
supervisory system that subjectively produces, internally, the experience of ego agency and, externally, the sense of re-estab-
lishing reality upon illusions. Indeed, most healthy observers describe a temporary lost of control on strange-face illusions
(Caputo, 2010b), whereas, in schizophrenic patients, serious deficits of ego agency and lost sense of reality upon strange-face
illusions are found (Caputo et al., 2012).
In phenomenology, mirror self-experience points to the profoundly unsettling encounter with one’s specular double
(Merleau-Ponty, 1964a, 1964b). Mirror gazing has been considered a sort of inter-subjective interaction between an
1
In the binomial distribution, the probability of success (to respond in synchrony by chance ) for a pair of subjects was conservatively set to the algebraic sum
of response probabilities of the two subjects. Response probability of a subject was calculated by algebraic ratio between its cumulative duration of event-
related responses (equal to the algebraic product between number of responses and mean response duration) and the session duration. Among participants,
response probabilities ranged from .03 to .35; mean response probability was .17 ± .04 across participants.
2
In mirror-gazing, event-related-responses had onset 46.8 ± 6.6 s, frequency 2.0 ± .2 responses per min, duration 6.9 ± .8 s (calculated from data of Caputo
(2010b)).
3
In mirror-gazing, phenomenological descriptions showed that the total number of different strange-faces per 10 min was 2.6 ± .1 (calculated from dataof
Caputo (2010a)).
326 G.B. Caputo/ Consciousness and Cognition 22 (2013) 324–329
embodied individual and own reflected face (Rochat & Zahavi, 2011). On the basis of the results of the present experiment,
gazing at another person may be even more unsettling than mirror-gazing.
Self-evaluations through Likert-type scales indicate a feeling of reality of strange-faces, which may suggest a moderate
loss of self-agency during illusions (van den Bos & Jeannerod, 2002). Strangeness of the other, perceived in place of the
known other member of the pair is intense. Correlation between strangeness and otherness indicates a dissociative effect
of illusions. Moreover, participants never identified themselves with strange-faces. Therefore, strange-faces were always
dissociative of subject’s self. This conclusion agrees with previous results of dissociative experiences in mirror-gazing (Cap-
uto, 2010b).
An explanation of the increase in the number of strange-faces during inter-subjective gazing with respect to mirror-gaz-
ing may be based on inter-subjective empathy (Singer & Lamm, 2009). If empathy is involved, then one should expect a high-
er frequency of illusions in inter-subjective gazing than in mirror-gazing – instead gazing at one’s own face should increase
monitoring and control of self-produced emotions. Indeed, empathy seems to facilitate the fact that symbolic contents
emerge into the subject’s consciousness, whereas these contents are dissociative of the subject’s self and seen as projections
into the other’s face – a conclusion that sounds counterintuitive for the role of empathy.
Synchronization of inter-subjective responses is proved in a number of pairs (50% of participants). Synchronous responses
occur in spite of large time-lags in their onsets and happen without awareness. This result adds to previous studies, showing
that the spontaneous synchrony of responses in a pair emerges as soon as visual information is exchanged for joint-action
(Knoblich, Butterfill, & Sebanz, 2011). Response coordination can result from unconscious mimicry of facial expressions in a
pair (Sonnby-Borgstrom, 2002; van Baaren, Janssen, Chartrand, & Dijksterhuis, 2009). However, synchrony of inter-subjec-
Fig. 1. Graphs show autocorrelation functions of the series of durations of strange-face illusions by a pair. Bars indicate correlations for different lags of
illusions in the series (for example, the illusion prior to the current illusion is a lag of 1). Lines indicate confidence limits (2.0 standard error limits). At lag 1,
the correlation is negative (p< .03) in both graphs.
G.B. Caputo/ Consciousness and Cognition 22 (2013) 324–329 327
tive strange-face illusions goes beyond response–coordination in joint-action or mimicry in facial expression, since it shows
a conjunction of dissociative illusory contents within a pair.
An explanation of synchrony may be based on the hypothesis (Caputo, 2010b) that strange-face illusions during mirror-
gazing are produced by binding the observer’s embodied representations (somaesthetic, kinaesthetic and motor facial pat-
tern) and the visual image in the mirror. As demonstrated, in the case of an observer wearing a mask that rendered the ob-
server’s facial features invisible during mirror-gazing (Caputo, 2011), it is not the image itself (the specular worn mask) that
changes, but the observer’s embodied information that triggers strange-face illusions.
4
In this manner, the sentient and numi-
nous beings are perceived in the worn mask that becomes alive (e.g. swinging eyes, moving lips, breathing). This hypothesis of
binding can be further developed by postulating that the observer’s embodied representation can produce minimal changes of
motor patterns of the observer’s face (probably prototypical expressions), which happen when strange-face illusions are per-
ceived. Henceforth, these changes in facial expression during strange-face illusions can become relevant to inter-subjective sub-
ject-other gazing. The subject’s facial expression, which may change minimally during the subject’s illusion, can cause
unconscious mimicry by the other in the pair. Changes in the other’s embodied representations due to unconscious mimicry
can enhance the probability that the other, after a delay, binds the subject’s visual face into a strange-face illusion. Unconscious
mutual mimicries and reciprocal embodiments between the subject and the other can lead to unconscious joint-action about a
shared strange-face illusion, hence to response–coordination and synchrony. Archetypal strange-faces might reflect uncon-
scious prototypes of somaesthetic, kinaesthetic and motor facial patterns at the conscious level.
A consequence of this explanation is that synchrony of responses (response–coordination) actually reflects synchrony of
illusion contents (illusion–conjunction). The term ‘illusion–conjunction’ is used to indicate that the subject and the other pro-
duce crossed projections of unconscious contents into mutual visual faces within the pair. Projections give awareness of
strange-faces but are dissociative of subject’s self, because unconscious subject contents are always embodied into the
other’s face. In this sense, mirror-gazing is a special case of inter-subjective gazing that occurs when embodiment is made
upon the subject’s own specular face, which is seen as a dissociative other.
According to this explanation of synchrony, the finding of an increase in the number of strange-faces during inter-sub-
jective gazing, with respect to mirror-gazing, can be explained by twofold subject’s facial embodiments that increase illu-
sion–conjunctions – both from the subject’s bodily self who is gazing at the other and from the subject’s mimicry of the
other’s bodily face who gazes at the subject.
Strange-face illusions are strikingly similar to projections of self’s gods and devils into external numinous beings found by
archetypal psychology in dreams, deep meditation, alchemy and artistic productions (Jung, 1977). Jung’s idea of synchronic-
ity (Jung, 1970; Yiassemides, 2011) evidences a common feature across spiritual experiences: unconscious archetypal con-
tents can emerge to consciousness through an acausal co-occurrence in time of events that have similar meanings.
Synchronicity coordinates the psychic and the physic at the ‘psychoid’ level, an idea that agrees with our hypothesis that
archetypal strange-faces can reflect prototype motor facial patterns. Synchronicity produces the symbolic conjunction of
opposites, also named syzygy, within the self or between individuals (Jung, 1977). Synchrony of strange-face illusions indi-
cates, according to Jung’s idea of synchronicity, the crossed conjunction of conscious and unconscious contents within a dyad
– in fact, a quadruple syzygy.
References
Bentall, R. P. (1990). The illusion of reality: A review and integration of psychological research on hallucinations. Psychological Bulletin, 107, 82–95.
Caputo, G. B. (2010a). Strange-face-in-the-mirror illusion. Perception, 39, 1007–1008.
Caputo, G. B. (2010b). Apparitional experiences of new faces and dissociation of self-identity during mirror-gazing. Perceptual and Motor Skills, 110,
1125–1138.
Caputo, G. B. (2011). Mask in the mirror: The living mask illusion. Perception, 40, 1261–1264.
Caputo, G. B., Ferrucci, R., Bortolomasi, M., Giacopuzzi, M., Priori, A., & Zago, S. (2012). Visual perception during mirror gazing at one’s own face in
schizophrenia. Schizophrenia Research, 140, 46–50.
Jung, C. G (1970). Synchronicity: An acausal connecting principle. In C. G. Jung (Ed.). Collected works (Vol. 8). Princeton: Princeton University Press.
Jung, C. G. (1977). Mysterium coniunctionis. Princeton: Princeton University Press.
Knoblich, G., Butterfill, S., & Sebanz, N. (2011). Psychological research on joint action: Theory and data. In B. Ross (Ed.). The psychology of learning and
motivation (Vol. 54, pp. 59–101). Burlington: Academic Press.
Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317, 1096–1099.
Merleau-Ponty, M. (1964a). L’oeil et l’esprit. Paris: Gallimard.
Merleau-Ponty, M. (1964b). The child’s relations with others. In M. Merleau-Ponty (Ed.), The primacy of perception (pp. 96–155). Evanston: Northwestern
University Press.
Rochat, P., & Zahavi, D. (2011). The uncanny mirror: A re-framing of mirror self-experience. Consciousness and Cognition, 20, 204–213.
Sforza, A., & Blanke, O. (2012). Autoscopic phenomena: Clinical and experimental perspectives. In Hallucinations (part 2, pp. 187–201). Berlin: Springer.
Singer, T., & Lamm, C. (2009). The social neuroscience of empathy. Annals of the New York Academy of Sciences, 81–96.
Sonnby-Borgstrom, M. (2002). Automatic mimicry reactions as related to differences in emotional empathy. Scandinavian Journal Psychology, 43, 433–443.
Terhune, D. B., & Smith, M. D. (2006). The induction of anomalous experiences in a mirror-gazing facility: Suggestion, cognitive perceptual personality traits
and phenomenological state effects. Journal of Nervous and Mental Disease, 194, 415–421.
Ur group (2001). Introduction to magic. New York: Inner Traditions.
4
In mirror-gazing when wearing a mask, phenomenological descriptions showed that the total number of different strange-faces can be higher (5.6 ± .7
strange-faces per 10 min; Caputo, 2011) than gazing to one’s own face or gazing to another person’s; however, this result essentially depends on the artistic
power engraved into the mask.
328 G.B. Caputo/ Consciousness and Cognition 22 (2013) 324–329
van Baaren, R., Janssen, L., Chartrand, T., & Dijksterhuis, A. (2009). Where is the love? The social consequences and moderators of mimicry in humans.
Philosophical Transactions of the Royal Society B, 2381–2389.
van den Bos, E., & Jeannerod, M. (2002). Sense of body and sense of action both contribute to self-recognition. Cognition, 85, 177–187.
Yiassemides, A. (2011). Chronos in synchronicity: Manifestations of the psychoid reality. Journal of Analytical Psychology, 56, 451–470.
G.B. Caputo/ Consciousness and Cognition 22 (2013) 324–329 329
