Penultimate Draft (Oct 21, 2013) Please cite the published version, IEICE Technical Report 113, no. 283, HCS2013-59: 53-57.
The Philosophy of the Feeling of Presence Masashi KASAKI†
Abstract The current state of the art development of human-shaped robots is partly motivated by an aspiration to produce efficient telepresence devices. The question of what it is like for a person to be present, however, has not received the attention it should deserve. Recent philosophical research on this question is fueled by psychiatric and neuropsychological studies of delusional misidentification syndromes. This paper reviews such philosophical research and articulates its significance for the development of telepresence devices. Keyword Telepresence, Presence, Delusional Misidentification Syndromes, Capgras Delusion, Philosophy Lee’s
1. Introduction
typology
is
conceptual,
and
may
not
be
The term ‘presence’ is of increasing importance for
substantial; it does not imply that the feelings of physical,
designing and developing media products and computer
social, and self presence are different mental states. The
interfaces. The term no doubt has a broad application to
conceptual distinctions have practical merits for designing
different
and programming telepresence devices. More substantially,
fields
concerning
media
in
general
and
interaction between human and machine in particular (it is
it might be possible to separate different kinds of feelings
a perfectly ordinary term anyway). At the same time,
of presence and measure them when a user operates a
however, ‘presence’ has different connotations in different
telepresence device in which she interacts with artificial
fields or even within a field. Serious attempts have been
environments. It is more difficult to separate different
made to clarify and define the connotations of the term,
kinds of feelings of presence when a user operates a
and to specify the precise measures for presence or the
telepresence
feeling of presence.
para-authentic environments (i.e., objects or persons at a
device
in
which
she
interacts
with
scholarship
and
remote place). For, if a user communicates and interacts
interchangeably
with
with actual humans via a telepresence device, their
‘telepresense,’ the term first coined by M. Minsky [1] to
reactions may affect how the user finds herself in the
refer to an operator ’s feeling of being at a remote place.
para-authentic environments, and vice versa.
In
the
technology,
context
of
telepresence
‘presence’ is
used
After an extensive review and scrutiny of the existing
First, the interlocutors may not find the avatar of the
literature on telepresence, K. M. Lee [2] proposes to
user humanlike
distinguish three subtypes of presence (for other reviews,
movements and voice. This may lead the interlocutors to
or
real,
say,
due
to
its
unnatural
see [3], [4], [5], [6]):
react to the avatar differently than they do to normal humans. As a result, the user’s feeling of social presence
Physical Presence: A psychological state in which
may decrease. Second, once this happens, the user may not
virtual (para-authentic or artificial) physical objects are
find herself immersed in the quasi-authentic environment,
experienced as actual physical objects in either sensory
resulting in a decrease in the feeling of self presence. It
or nonsensory ways.
seems plain that our feeling of social presence is tightly
Social Presence: A psychological state in which vir tual
connected to our feeling of self presence. Third, if the user
(para-authentic
are
only has decreased feeling of social presence (and thereby
experienced as actual social actors in either sensory or
decreased feeling of self presence), this may affect her
nonsensory ways.
avatar ’s reactions to the interlocutors. As a result, the
Self Presence: A psychological state in which virtual
degree
(para-authentic or artificial) self/selves are experienced
humanlike or real may decrease. In sum, a user ’s
as the actual self in either sensory or nonsensory ways.
feeling(s)
or
artificial)
social
actors
to
which of
the
presence
interlocutors are
find
the
inter-connected
avatar to
an
interlocutor ’s feeling(s) of presence toward the avatar of
the user.
unrecognized faces, whereas they exhibit a decrease in
This consideration suggests that it is important to take
SCR to unfamiliar faces. In contrast, patients with
account of both the user ’s and the interlocut or ’s feelings
Capgras delusion do not show any difference in SCR to
of presence in conceptualizing and designing telepresence
familiar
devices. The current research on telepresence mainly
hyporesponsiveness, [15]). Partly on the basis of these
focuses on the user ’s feeling(s) of presence, whereas
findings, the two-route model of facial recognition posits
philosophical research on presence often targets one’s
two different neural pathways responsible for overt and
feeling(s) of other ’s presence. Philosophical research on
covert
this score is motivated and fueled by psychiatric and
visual-limbic pathway (identifying faces) and the dorsal
neuropsychological studies of delusional misidentification
visual-limbic
syndromes. This paper aims to provide a review of
significance of faces) ([14], [16], [17], [18], [19], [20],
philosophical and psychological perspectives into the
but also see [21]). Prosopagnosia is regarded to stem from
feeling(s) of presence, and to articulate their significance
a defect in the former, and Capgras delusion from a defect
for the development of telepresence devices .
in the latter.
and
face
unfamiliar
recognition, pathway
faces
respectively:
(detecting
the
(emotional
the
ventral
valence
or
A fully adequate explanation of Capgras delusion needs to explain (1) what neurological mechanism underlies the
2. Delusional Misidentification Syndromes With a notable exception of A. Noë [7], recent
delusion, (2) what content and character the anomalous
philosophical research on presence has been advanced by
experience of the patient with the delusion has, and (3)
theorizing
delusional
why the patient forms and maintains the delusional belief.
misidentification syndromes ([8], [9], [10], [11], [12],
The two-route model is an important contribution to (1),
[13]).
and also involves particular answers to (2) and (3).
on
the
Delusional
symptoms
of
misidentification
syndromes
are
psychiatric disorders the symptoms of which involve
Different views on (2) and (3) have been proposed. On
monothematic, circumscribed delusions. Prominent ones
one-stage models of Capgras delusion, the anomalous
are Capgras, Frégoli, and Cotard delusions. The three
experience produced by the neurological disruption is
delusions
sufficient for eliciting the delusional belief ([22], [23],
are
different
in
the
thematic
content
of
[24]). One-stage models typically have it that the content
delusional belief as follows:
of the anomalous perceptual experience is thematically Capgras Delusion: At least one close person (a family
rich, e.g., ‘the [perceived] woman is not the [remembered]
member or friend) is replaced by an imposter (e.g., an
woman,’ ‘the [perceived] woman is not the person I think
alien or robot).
of as my wife,’ or even strongly ‘the [perceived] woman is
Frégoli Delusion: Different strangers are the same
not
person in disguise.
responsible for the patient’s pathology is in her experience,
Cotard Delusion: The patient herself is dead or does not
not in her cognitive process of forming or revising beliefs;
exist.
rather, the delusional belief is an outcome of the rational
my
wife but
an
imposter.’ The primal
defect
response to the anomalous experience. Although the three delusions are often discussed
On two-stage models of Capgras delusion, not only the
together, the main focus of the paper is on C apgras
anomalous experience but also some sort of cognitive
delusion, simply for lack of space. Patients with Capgras
impairment are necessary (and sufficient) for eliciting the
delusion have relatively unimpaired abilities to identify
delusional belief ([25], [26], [27]). Although many
and discern familiar faces. According to Ellis and Young
proponents of one or another form of the two-stage model
[14],
of
agree that the content of the anomalous experience is
prosopagnosia—a disorder of face recognition typically
thematically rich, some hold that it is relatively thin, e.g.,
associated
ventral
‘the [perceived] woman looks different.’ The delusional
occipitotemporal cortex. Prosopagnosic patients have a
belief that the [perceived] woman is an imposter is formed
selectively impaired ability to recognize faces, while
as an explanation of the anomalous experience. But the
having relatively unimpaired abilities to recognize other
explanation is an outcome of the irrational process of
objects. They nevertheless exhibit an increase in skin
forming and maintaining beliefs due to a cognitive defect
conductance response (SCR) to familiar but consciously
or bias. ([26], [28], [29], [30], [31]). R. McKay ([30], p.
Capgras with
delusion
is
bilateral
a
mirror
lesions
image in
345) argues that patients with Capgras delusion have a
most prominently, (i) how the contents of perceptual
bias toward explanatory adequacy and thereby “update
experience are rationally related to the contents of belief,
belief as if ignoring the relevant prior probabilities of
and (ii) what presence or the feeling of presence consist s
candidate hypotheses.”
in. These may be just different forms of the same problem.
Let o be the content of the anomalous experience, and h w and h s be two alternative hypotheses:
Philosophical work on (i) and (ii) both refer s to and is referred to in the literature on delusional misidentification syndromes in other fields.
h w : the [perceived] woman is in fact my wife. h s : the [perceived] woman is an stranger.
Two views are distinguished with regard to (i): the top-down account and the bottom-up account ([33]). The top-down account has it that the cont ent of the anomalous
P(h w ) and P(h s ) are subjective probabilities the patient
experience is at least in part dependent on or determined
assigns to h w and h s , prior to perceiving the woman, i.e.,
by a defect in the belief system. The bottom-up account
obtaining data o. After obtaining o, the patient needs to
has it that the anomalous experience distorts the belief
update her beliefs on o. Insofar as she obeys Bayes’
system. The two accounts are highly relevant to the debate
theorem, the posterior probabilities of h w and h s are
between empiricism and rationalism ([34]) (or to use
specified as follows:
different terminology, the debate between inferentialism and liberalism ([8])). Empiricism is the view that th e
P(ℎ𝑤 ⁄𝑜) =
P(ℎ𝑤 ) × P(𝑜⁄ℎ𝑤 ) 𝑃(𝑜)
contents of experience are rich, and thereby they alone are
P(ℎ𝑠 ⁄𝑜) =
P(ℎ𝑠 ) × P(𝑜⁄ℎ𝑠 ) 𝑃(𝑜)
the other hand, is the view that the content s of experience
sufficient to rationalize or justify beliefs. Rationalism, on are
relatively
thin,
and
thereby
they
need
some
background belief to rationalize or justify other beliefs. Even though the likelihood of o under h s , P(𝑜⁄ℎ𝑠 ), is
Rationalism is more closely associated with the two -factor
much higher than that under h w , P(𝑜⁄ℎ𝑤 ) (h s explains the
model than empiricism is; if the delusional belief is
occurrence of o much better than h w does), a normal
irrational, it must be accounted for by a cognitive defect in
person assigns to a much lower prior probability to h s than
the patient’s belief system, and the defect may be the lack
to h w . So the posterior probability of h s is not higher than
of one or another background belief.
that of h w for a normal person. The patient with Capgras
The patient with Capgras delusion is able to visually
delusion, due to the bias toward explanatory adequacy,
identify and discriminate familiar faces. It has been
ignores the priors of h w and h s , and hence the posterior
assumed that the visual content of the patient’s face
probability of h s is higher than that of h w for her.
perception
An advantage of the two-stage model over the one-stage
is
relatively
unchanged
by
the
neural
disruption. What is changed is the affective aspect of the
model is that the former can explain why patients with
perceptual
damage to ventromedial frontal regions of the brain do not
philosophers identify this affective aspect with the fee ling
form delusional beliefs, while they show symptoms
of (human) presence in one or another sense ([8], [9], [10],
similar to Capgras delusion, such as the reported feeling
[11], [12], [13]; also see [35], [36]). It should be
of unfamiliarity and emotional hyporesponsiveness ([32]).
emphasized that they define the feeling of presence in
They suffer from a defect in the first, experimental stage,
similar but importantly different fashions. But for lack of
but keep the second, cognitive stage of belief formation
space, only M. Ratcliffe’s account ([11], [12], [13]) is
and revision intact.
discussed here. Ratcliffe,
3. Philosophy and Delusional Misidentification
experience
by
relying
of
familiar
on
the
faces.
Many
phenomenological
considerations by Husserl and Merleau-Ponty, espouses a kind of enactivism:
Syndromes Delusional misidentification syndromes are important for many fields; not to mention psychiatry, they are
Perceptual
studied
recognition
dynamic set of correspondences between what is
mechanism. In addition, philosophers hold that they may
actually perceived and how potential activities will alter
shed light on several perennial philosophical problems:
that perception ([12], p. 208)
in
neuroscience
to
reveal
face
experience
involves
an
elaborate
and
another mirror image of Capgras delusion: the interlocutor On Ratcliffe’s view, the patient with Capgras delusion undergoes a change in the structure of experience; how a
does not believe that it is a human, but may feel the presence of a human.
person experiences an object is determined by a set of
Even if this is not close to being achieved, the
dynamic possibilities concerning that object. The patient
realization of human presence is a leading motivation for
still associates a set of possibilities with familiar persons
developing
(e.g., how they will look like if the patient changes her
engineering. Philosophical inquiry into presence and
position or they move), but lack the possibilities of
related problems has been motivated and fueled by
communication and affective relatedness. This lack of the
psychiatric and neuropsychological studies of delusional
possibilities results in the patient experiencing familiar
misidentification syndromes. Robotics and engineering
persons in a distinctively impersonal way. Ratcliffe’s
can provide further fuel. On the other hand, philosophy
account, unlike orthodoxy, implies that the content s of
and those fields can help to develop telepresence devices
visual experience are in part constituted by affects.
by offering a rich set of frameworks in which presence or
Noë ([7]; also see [37]) develops an account of presence
telepresence
devices
in
robotics
and
the feeling of presence is conceptualized and investigated.
in general on the basis of his own variant of enactivism. He argues that there are varieties of presence, and how an
References
object is present depends on one’s implicit knowledge of sensorimotor dependencies characteristic of that object. For
example,
the
visual
content
and
character
of
experience of an object is determined by our implicit knowledge of systematic patterns of changes of retinal stimulation. The more remote an object is, it becomes less present, because it is associated with a smaller number of changing patterns. On these accounts, the lack of the feeling of presence from which the patient with Capgras delusion suffers can be explained in terms of a change in the structure or content of her experience, which is caused by an affective or cognitive change. The neurophysiological underpinning of these accounts is not yet in the offing. Prediction error system may play an important role for this purpose ([39], [40], [41], [42]). Prediction error is an important part of learning mechanism in organisms. Prediction error signals are triggered by discrepancies between what is perceived or experienced and what is expected. When a prediction error signal exists, beliefs update in order to minimize it (perhaps, in Bayesian fashions) . Delusional patients form expectations, anomalous
but
they
experience.
are They
disconfirmed respond
to
by
their
aberrant
prediction error signaling by forming delusions. The feeling of presence may be the result of successful suppression of prediction error signals ([42]).
4. Concluding Remarks An interesting new model of telepresence device is a realistic humanlike android by which the operator and the interlocutor can adequately communicate and interact ([43]). Such an android, if advanced, may be treated as
[1] M. Minsky, “Telepresence,” Omni, 2(9), pp. 45-52, 1980. [2] K. M. Lee, “Presence, Explicated,” Communication Theory, 14(1), pp. 27-50, 2004. [3] F. Biocca, “The Cyborg’s Dilemma: Progressive Embodiment in Virtual Environments,” Journal of Computer-Mediated Communication, 3(2), 1997, accessed Oct 20, 2013, fro m http://onlinelibrary.wiley.com/doi/10.1111/j.1083 -61 01.1997.tb00070.x/full [4] M. Lombard and T. Ditton, “At the Heart of It All: The Concept of Presence,” Journal of Computer-Mediated Communication, 3(2), 1997, accessed Oct 20, 2013, from http://onlinelibrary.wiley.com/doi/10.1111/j.1083 -61 01.1997.tb00072.x/full [5] M. V. Sanchez-Vives and M. Slater, “From Presence to Consciousness through Virtual Reality,” Nature Reviews Neuroscience, 6(4), pp. 332-39, 2005. [6] J. Steuer, “Defining Virtual Reality: Dimen sions Determining Telepresence,” Journal of Communication, 42(4), pp. 73-93, 1992. [7] A. Noë, Varieties of Presence, MIT Press, Cambridge, MA, 2012. [8] J. Dokic, “Perceptual Recognition and the Feeling of Presence,” in Perceiving the World, B. Nanay, ed., pp. 33-53, Oxford University Press, Oxford , 2010. [9] M. Matthen, “Two Visual Systems and the Feeling of Presence in Perception,” in Action and Consciousness: Sensorimotor Dynamics and Two Visual Systems, N. Gangopadhyay, M. Madary, and F. Spicer, ed., pp. 107-124, Oxford University Press, New York, 2010. [10] B. P. McLaughlin, “Monothetic Delusions and Existential Feelings,” in Delusion and Self-Deception: Affective and Motivational Influences on Belief Formation , T. Bayne and J. Fernández, ed., pp. 139-164, Psychology Press, New York, 2009. [11] M. Ratcliffe, “Understanding Existential Changes in Psychiatric Illness: The Indispensability of Phenomenology,” in Psychiatry as Cognitive
Neuroscience, M. Broome and L. Bortolotti, ed., pp. 223-244, Oxford University Press, Oxford, 2009. [12] M. Ratcliffe, “The Phenomenological Role of Affect in the Capgras Delusion,” Continental Philosophy Review, 41(2), pp. 195-216, 2008. [13] M. Ratcliffe, “Understanding Existential Changes in Psychiatric Illness: The Indispensability of Phenomenology,” in Psychiatry as Cognitive Neuroscience, M. Broome and L. Bortolotti, ed., pp. 223-244, Oxford University Press, Oxford, 2009. [14] H. D. Ellis and A. W. Young, “Accounting for Delusional Misidentification,” British Journal of Psychiatry, 157, pp. 239-248, 1990. [15] C. Fine, J. Craigie, and I. Gold, “Damned If You Do; Damned If You Don’t: The Impasse in Cognitive Accounts of Capgras Delusion,” Philosophy, Psychiatry, & Psychology, 12(2), pp.143-151, 2005. [16] R. M. Bauer, “The Cognitive Psychophysiology of Prosopagnosia,” in Aspects of Face Processing, H. Ellis, M. Jeeves, F. Newcombe, and A. Young, ed., pp. 253-267, Martinus Nijhoff, Dordrecht, 1986. [17] R. M. Bauer, “Autonomic Recognition of Names and Faces in Prosopagnosia: A Neuropsychological Application of the Guilty Knowledge Test, ” Neuropsychologia, 22(4), pp. 457-469, 1984. [18] H. D. Ellis and M. B. Lewis, “Capgras Delusion: A Window on Face Recognition.” Trends in Cognitive Sciences, 5(4), pp. 149-156, 2000. [19] H. D. Ellis, M. B. Lewis, H. F. Moselhy, and A. W. Young. “Automatic without Autonomic Responses to Familiar Faces: Differential Components of Covert Face Recognition in a Case of Capgras Delusion,” Cognitive Neuropsychiatry, 5(4), pp. 255-269, 2000. [20] W. Hirstein and V. S. Ramachandram, “Capgras Syndrome: A Novel Probe for Understanding the Neural Representation of the Identity and F amiliarity of Persons,” Proceedings of the Royal Society: Biological Sciences, B264(1380), pp. 437-444, 1997. [21] G. Young, “In What Sense ‘‘Familiar ’’? Examining Experiential Differences within Pathologies of Facial Recognition,” Consciousness and Cognition , 18(3), pp. 628-38, 2009. [22] B. A. Maher, “Anomalous Experience in Everyday Life: Its Significance for Psychopathology,” Monist, 82, pp. 547-70, 1999. [23] B. A. Maher, “Anomalous Experience and Delusional Thinking: The Logic of Explanations,” in Delusional Beliefs, T. F. Oltmann and B. A. Maher, ed., pp. 15-33, Wiley, New York, 1988. [24] I. Gold and J. Hohwy, “Rationality and Schizophrenic Delusion,” Mind & Language, 15(1), pp. 146-167, 2000. [25] M. Coltheart, The 33rd Sir Fredrick Bartlett Lecture. “Cognitive Neuropsychiatry and Delusional Belief,” The Quarterly Journal of Experimental Psychology , 60(8), pp. 1041-1062, 2007. [26] M. Davies and M. Coltheart, “Introduction: Pathologies of Belief,” in Pathologies of Belief, M. Coltheart and M. Davies, ed., pp. 1-46, Blackwell, Malden, MA, 2000. [27] E. Pacherie, “Perception, Emotion, and Delusions: The Case of the Capgras Delusion, ” in Delusion and Self-Deception: Affective and Motivational Influences on Belief Formation , T. Bayne and J.
Fernández, ed., pp. 107-125, Psychology Press, New York, 2009. [28] A. M. Aimola Davies and M. Davies, “Explaining Pathologies of Belief,” in Psychiatry as Cognitive Neuroscience: Philosophical Perspectives, M. R. Broome and L. Bortolotti, ed., pp. 285-323, Oxford University Press, Oxford, 2009. [29] M. Coltheart, P. Menzies, and J. Sutton, “Abductive Inference and Delusional Belief. ” Cognitive Neuropsychiatry, 15(1), pp. 261-287, 2010. [30] R. McKay, “Delusional Inference,” Mind & Language, 27(3), pp. 330-35, 2012. [31] T. Stone and A. W. Young, “Delusions and Brain Injury: The Philosophy and Psychology of Belief. ” Mind & Language, 12(3-4), pp. 327-64, 1997. [32] D. Tranel, H. Damasio, and A. R. Damasio, “Double Dissociation between Overt and Covert Face Recognition,” Journal of Cognitive Neuroscience , 7(4), pp. 425-432, 1995. [33] T. Bayne and E. Pacherie, “Bottom-up or Top-down? Campbell’s Rationalist Account of Monothematic Delusions,” Philosophy, Psychiatry, & Psychology, 11(1), pp.1-11, 2004. [34] J. Campbell, “Rationality. Meaning and the Analysis of Delusion,” Philosophy, Psychiatry, & Psychology, 8(2/3), pp. 89-100, 2001. [35] G. Young, “Beliefs, Experiences and Misplaced Being: An Interactionist Account of Delusional Misidentification,” Phenomenology and the Cognitive Sciences, 10(2), pp. 195-215, 2011. [36] G. Young, “Restating the Role of Phenomenal Experience in the Formation and Maintenance of the Capgras Delusion,” Phenomenology and the Cognitive Sciences, 7(2), pp. 177-189, 2008. [37] J. K. O’Regan and A. Noë, “A Sensorimotor Approach to Vision and Visual Consciousness,” Behavioral and Brain Sciences, 24(5), pp. 939–73, 2001. [38] P. R. Corlett, D. C. D’Souza, and J. H. Krystal, “Capgras Syndrome Induced by Ketamine in a Healthy Subject,” Biological Psychiatry 68(1), e1–e2, 2010. [39] P. R. Corlett, J. H. Krystal, J. R. Taylor, and P. C. Fletcher, “Why Do Delusions Persist?,” Frontiers of Human Neuroscience, 3(12), pp. 1-9, 2009. [40] P. R. Corlett, G. K. Murray, G. D. Honey, M. R. F. Aitken, D. R. Shanks, T. W. Robbins, E. T. Bullmore, A. Dickinson, and P. C. Fletcher, “Disrupted Prediction Error Signal in Psychosis: Evidence for an Associative Account of Delusions,” Brain, 130(Pt 9), pp. 2387-2400, 2007. [41] P. C. Fletcher and C. D. Frith, “Perceiving is Believing: A Bayesian Approach to Explaining the Positive Symptoms of Schizophrenia,” Nature Reviews Neuroscience, 10(1), pp. 48-58, 2009. [42] A. K. Seth, K. Suzuki, and H. D. Critchley, “An Interoceptive Predictive Coding Model of Consci ous Presence,” Frontiers in Consciousness Science , 2(395), pp. 1-16, 2011. [43] S. Nishio, H. Ishiguro, and N. Hagita, “Geminoid: Teleoperated Android of an Existing Person, ” in Humanoid Robots: New Developments, A. C. de Pina Filho, A.C. ed., pp. 343-352, I-Tech Education and Publishing, Vienna, 2007.