Philosophy and Cognitive Science
Canisius College
What kind
of movement or behavior is involved in neonate imitation? What exactly is the newborn infant
doing when it responds to seeing gestures on another person's face? This question is closely related to
some other questions, such as whether neonate imitation is possible, and
whether it is truly imitation. Piaget,
of course, thought that this sort of "invisible imitation" was not
possible for infants less than 8-12 months of age.
The intellectual mechanisms of the [child under 8 months] will not allow him to imitate movements he sees made by others when the corresponding movements of his own body are known to him only tactually or kinesthetically, and not visually (as, for instance, putting out his tongue) .... Thus since the child cannot see his own face, there will be no imitation of movements of the face at this stage.... For imitation of such movements to be possible, there must be co-ordination of visual schemas with tactilo-kinesthetic schemas..." (Piaget 1962, pp. 19, 45)
You can see here, however, that it
is possible [video]. Meltzoff and Moore have done the controlled experiments
to demonstrate that this is not reflex action or some kind of release mechanism
that is involved. Specifically,
they have shown that
None of these things are
consistent with this kind of behavior being reflex.
If we rule out reflex movement, what else might explain this behavior? Recently Marcel Kinsbourne (2002) and Bennett Bertenthal (private correspondence) have proposed that neonate imitation may be the result of motor priming and the lack of any inhibitory mechanisms (due to inadequate development of prefrontal cortex) in the newborn human to prevent the imitative reaction.[1] Kinsbourne writes:
Prefrontally injured patients exhibit excessive, unwanted imitation. Infants are also prefrontally inadequate, because the requisite relatively late-occurring neural maturation has not yet taken place. Infants' imitation may be [the result of] uninhibited enactive perception. ... The act of perception activates the infant's own innate corresponding movement patterns. (2002, pp. 311, 316).
This explanation would not require the infant to be self-consciously aware of what it was doing. It could be a case of simple, mechanical, contagion. But, according to Bertenthal, this interpretation does not necessarily imply that the behavior does not represent "true imitation." It would depend
on how you define imitation, and whether or not it must be accessible to conscious processing. I think the recent empirical literature shows very clearly that not all forms of adult imitation are conscious (e.g., imitation of meaningful vs meaningless gestures). For this reason, my preference is to suggest the possibility of different forms of imitation rather than to define only some forms of imitation as "true imitation."
The motor priming
hypothesis is an interesting and possible explanation. The testing that would confirm it,
however, has not yet been carried out.
Let's
consider Meltzoff's explanation.
Meltzoff and his colleagues are actually in some agreement with Piaget
on the logic of how to explain this behavior. According to Piaget, the infant is not capable of imitation
because it has not yet developed an adequate body schema, and, following a very
old empiricist doctrine, it is unable to perceive in a cross-modal way - that
is, the infant is not able to translate what it sees into a proprioceptive
sense of its own bodily possibilities.
Meltzoff turns this logic around.
The fact that infants are able
to imitate counts as evidence for an innate body schema and for innate
capacities for cross-modal perception. This forms the first part of Meltzoff's
answer to the question of what the infant is doing: neonate imitation is a controlled
motor response that can be explained in terms of body schemas, cross-modal
perceptions, and motor representations that generate mental representations of
the other person's face, and possible imitative responses.
The
second part of Meltzoff's answer has to do with the context in which this kind
of imitation happens, and the use to which the infant seems capable of putting
it. Specifically, Meltzoff argues
that this is the basis and the beginning of social cognition. It is not simply motor activity, but an
intersubjective event. He defends
the idea that
the imitation of human actions is the first bridge between the infant
and others, and that imitation serves the dual functions of differentiating the
broad class of 'others' into inviduals and providing an early means of
communication with them. (1999, p. 49)
I have to admit that
I'm sympathetic to this account, and indeed I co-authored an article with
Meltzoff several years ago defending some of its details. Now, however, I want to reconsider it
in light of some other empirical evidence developed in a quite different kind
of study. My concern does not involve doubts about either of the two parts of
the answer - let's call them the sensory-motor part (specified in terms of
body-schema processes, proprioception, intermodal code, and so forth) and the
intersubjective part (specified in terms of social cognition, communication,
and theory of mind). My question
is about how these two parts go together.
There are, in fact, three ways to think about this issue.
The
first approach is what I will call the motor theory of social cognition. It has recently been called by some of
its proponents 'motor chauvinism' and described as follows:
They develop this last
thought along the following lines:
Direct information transmission between people, such as speech, arm gestures or facial expressions, is mediated through the motor system, which provides a common code for communication ... . It could be that the same computational processes underlie action, the perception of action, and social cognition. [...] Although the behavior of others in response to our actions is more noisy and non-linear than the response of our arm to a motor command, computationally they are not fundamentally different. (Wolpert, Grahramani, and Flanagan , 2001, p. 493).
On this account, social cognition
is reducible to instrumental movement. I see another person's instrumental
movement and my own motor system reverberates as if I were making that
instrumental movement. This reverberation is the mechanism of social
cognition. Reaching to pick up a
glass is more or less equivalent to reaching to shake someone's hand. Understanding the facial gesture of
another person is nothing more than activating the same computational process
responsible for my own comparable facial gestures.
Let me
note two things about this approach.
First, there is very good neurological evidence that supports it.[2] Second, if you are familiar with the
current debate between simulationists and theory-theorists, you will recognize
that the motor theory of social cognition provides important support to the
simulationist model. Meltzoff is more on the theory theory side of this debate
(see especially Gopnik and Meltzoff, 1998), and this may help us to see the
difference between his model and the motor theory of social cognition.[3]
The
second approach to consider, then, is the one taken by Meltzoff and his
colleagues. Motor events are
related to intersubjective events according to the following formula: motor
action, which is volitional and intentional, is used by the infant in a
cognitive way to gain access to the other person's intentions and mental
states. The child uses imitation, for example, to test out the identity of the
other person (Meltzoff and Moore, 1992; 1994). "[I]nfants have a code for interpreting that the other
is 'like me' right from the earliest phases of infancy. Self and other can be related because
their bodily actions can be compared in commensurate terms" (Meltzoff,
1999, p. 58). My own capacity for
instrumental action allows me to infer the intentions of others (Meltzoff,
1995). In a theory-theory
perspective, it is indeed a form of inference from something not explicitly social
(i.e., instrumental movement) to something characterized by social
relations. The infant is portrayed
as thinking in the following way: "I intend to produce these acts, the
adult performs these same acts, they are not chance events, therefore the adult
intends his acts" (Meltzoff, 1999, p. 62). In this way, instrumental actions come to have
intersubjective significance.
So,
to summarize the first two approaches:
Motor (simulation) theory of social cognition:
Social cognition = reducible to instrumental motor
performance
Theory theory of social cognition (Meltzoff):
Social cognition = instrumental motor performance + mental
inference
I want to offer evidence for a third approach, an alternative that proposes a different relation between motor action and social cognition.[4] On this view, the movement involved in neonate imitation is not equivalent to instrumental movement - a movement that needs to be transformed into socially significant movement via simulation or inference. Rather, imitation, like gesture, is expressive movement - and something that is intrinsically social from the very beginning.
Evidence for this comes from experiments conducted on gesture in a subject (IW) who lacks proprioception and the sense of touch from the neck down. Lacking proprioception, IW is without a functioning body schema (Gallagher and Cole, 1995), and has profound difficulties with locomotive and instrumental movements. IW is required to use visual feedback and cognitive effort to control his movements. In contrast to such motor problems, IW's gestural movements are normal across a number of important parameters.
For controlled locomotive and instrumental movement, for example, to walk or to reach for a glass, IW must make use of conscious visual and cognitive strategies. For example, he needs to see his hand and consciously think through what he must do with his hand in order to reach and grasp a glass and to move it to his mouth. By contrast, his conversational hand gestures, when he sees his hands, appear normal for launch, timing, morphokinesis (shape), and, to an extent, topokinesis (location of one hand relative to the other and relative to other parts of his body) in ways that appear not to involve conscious control. Furthermore, when IW is prevented from seeing his hands, his gestures remain normal for launch, timing, and morphokinesis, but they decline in regard to topokinesis (Cole, Gallagher, and McNeill, 1998; 2002).
These experiments lead us to conclude that, normally, body-schematic
motor processes are important in the normal control of some aspects of
gesticulation and expressive movements, especially those that involve
topokinetic precision. In IW, this specific function depends on visual control,
similar to his control of locomotive and instrumental movements.
Importantly, however, the other aspects of his gesticulation are not reducible to the same kind of motor control. Following David McNeill's (1992) theory that gesture is an integrated part of language, then IW's control of gesture in regard to launch, timing, and morphokinesis, as revealed when he was unable to see his hands, is independent of body-schematic processes, proprioception, or even the cognitive effort required for his instrumental movements), and is rather dependent on linguistic and communicative functions. Gesture is a form of language, and more generally a form of socially expressive movement that is not fully explained in terms of body schemas, proprioception, intermodal perception, or any additional conscious mental control.
Gesture, as a form of expressive movement, is irreducible to instrumental or locomotive movement, and is clearly distinct from reflex movement. Nor is it explainable in terms of motor priming, since the movement that may be elicited from motor priming can be meaningless and non-expressive.
My question (and I do not propose it as a definitive answer - but certainly as a hypothesis to be tested) is whether we could consider neonate imitation (the imitation of a gesture) to be a form of expressive movement - and as such, not reducible to reflex, or to intentional body-schematic processes, motor representations, or motor priming.[5] Let me note two things. First, we should not think of expression in terms of externalizing something that is internal. It is not as if the infant has a conceptual understanding or a thought that they attempt to externalize by imitative gesture. What Merleau-Ponty says with repect to language, I would generalize to expressive action, including gesture and imitation:
What, then, does language [gesture or expressive action] express, if it does not express thoughts? It presents or rather it is the subject's taking up of a position in the world of his meanings" (Merleau-Ponty, 1962, p. 193).
Second, the implications of this view include the idea that the human infant is born with an innate capability for expression. In other words, the infant is born not simply with a body-schematic capacity for motor control that serves as an ancillary to social interaction; the infant is born already capable of action that is intrinsically social. Social cognition is built up, not from elements of non-social movements, but from action that is already social.
Cole, J.D. et al. (1998) Gestures after total deafferentation of the bodily and spatial senses. In Oralité et Gestualité: Communication Multi-modale, Interaction (Santi, S. et al., eds) pp. 65-69. Paris: L. Harmattan,
Gallese, V. (2001) The 'shared manifold' hypothesis: from mirror neurons to empathy. J. Conscious. Stud. 8, 33-50
Gallese, V. and Goldman, A. (1998) Mirror neurons and the simulation theory of mind reading. Trends Cogn. Sci. 2, 493-501
Wolpert, D.M. et al. (2001) Perspectives and problems in motor learning. Trends Cogn. Sci. 5, 487-494
Blakemore, S-J. and Decety, J. (2001) From the perception of action to the understanding of Intention. Nat. Rev. Neurosci. 2, 561-567
Decety, J. and Grèzes, J. (1999) Neural mechanisms subserving the perception of human actions. Trends Cogn. Sci. 3, 172-178
Grèzes, J. and Decety, J. (2001) Functional anatomy of execution, mental simulation, and verb generation of actions: a meta-analysis. Hum. Brain Mapp. 12, 1-19
Ruby, P. and Decety, J. (2001) Effect of subjective perspective taking during simulation of action: a PET investigation of agency. Nat. Neurosci. 4, 546-550
Decety, J. et al. (2002) A PET exploration of the neural mechanisms involved in reciprocal imitation. NeuroImage 15, 265-272
[1] This kind of answer is quite consistent with what
Mark Johnson proposes in terms of two distinct processes: "...the first process consists of
a system accessed via the subcortical visuomotor pathway (but possibly also
involving some of the deeper, earlier developing, cortical layers) that is
responsible for the preferential tracking of faces in newborns. However, the
influence of this system over behavior declines (possibly due to inhibition by
developing cortical circuits) during the second month of life. The second
brain system depends upon cortical maturity, and exposure to faces over the
first month or two, and begins to control infant orienting preferences from
around two or three months of age" (113). Johnson, Mark (1997). Developmental Cognitive
Neuroscience. Blackwell. For Kinsbourne, the idea is that another person's gesture activates mirror neurons and
since the infant's inhibitory mechanisms are not yet developed, the infant
cannot help but imitate.
[2]The authors cite research on mirror neurons, which
respond both to self-generated actions and the actions of others. They also
note that 'human neuroimaging and magnetic stimulation studies have shown that
areas associated with action are also active during imitation and observation.'
Decety and their colleagues, for example [6-10], have shown that there are
shared motor representations for action, observation of another person's
action, imitation and mental simulation of action. That is, these different
activities activate the same brain areas (the SMA, the dorsal premotor cortex,
the supramarginal gyrus, and the superior parietal lobe) (Blakemore and Decety,
2001; Decetyand Grèzes, 1999; Grèzesand Decety, 2001; Ruby and Decety, 2001;
Decety, J. et al. 2002).
[3] This is not as clear as I have indicated. Meltzoff often describes what seems
more like a simulationist model, even in developing his theory-theory
model. Furthermore, Meltzoff is
working closely with Decety, who has developed some of the evidence often cited
by simulationists.
[4] This view is consistent with an integrative theory of
social cognition (outlined in Cole, Gallagher, and McNeill, 2002, and
Gallagher, Cole, and McNeill, 2002) - indeed, it is a simple extension of that
theory - and with an alternative approach to the theory of mind debate, based
on Trevarthan's notion of primary intersubjectivity (Gallagher, 2001) - a
practice of mind that is more basic than either simulation or theory stances.
[5]Kinsbourne (2002) does suggest that
infants imitate because they attend to the visual - and they especially attend
to others like themselves - not to mechanical devices that lack biological
motion. Imitation "underlies
affiliation both to individuals and to the group. If so, affiliation has a neurobiological rudiment, mediated
by imitation."