Transcript Language within our grasp:

Language within our grasp:
Gesture, mirror neurons, and meaning
Shared meaning and symbol grounding
• One mystery we have alluded to is ‘symbol grounding’: how
communicators manage to synchronize their meanings
– What keeps your ‘cat’ meaning the same as my ‘cat’?
• Wittgenstein’s emphasis on shared ‘forms of life’ as the ground of
meaning was intended to buttress the idea that only shared action
could bring different systems into synchrony of meaning
Biological meaning
• We can imagine (though Wittgenstein did not) that some early
shared actions would be synchronized by biological necessity
– For example, we all understand the basic biological needs, and
can easily parse actions related to seeking food, water, sex,
avoiding death and pain etc.
– Thus, concepts related to these activities may be transparently
(innately) obvious to us and therefore easy to communicate
about
– Most natural animal communication systems are about such
things
The more innateness, the better
• For the purposes of meaning synchronization, especially in
pre-linguistic history (and perhaps pre-linguistic
development, and non-linguistic environments?) the more
‘semantics’ we can ascribe to innate factors, the smaller the
mystery of symbol grounding
– The more shared meaning we start with, the easier it is
to imagine we could add more by building on what we
have
Mirror neurons
• Mirror neurons were discovered in singlecell recording in area F5: ventral premotor
cortex
44/45
PRODUCTION
22
RECEPTION
Mirror neurons
• Mirror neurons were discovered in single-cell recording in area
F5: ventral [= lower] premotor cortex
• They discharge during active movements of the hand and/or
mouth
• They are sensitive to different purposes
– Some discharge during grasping; some during (specific kinds
of) holding; some during reaching; some during tearing
• Surprisingly, they were also found to discharge during
observation of related movements: hence their name ‘mirror
neurons’
• Later it was found they also discharge when the animal hears
sounds associated with the purposes to which they are sensitive
Mirroring as communication
• These are very general ‘semantic’ encoders
– They fire with presentation of the action from many
perspective, near and far
– They are sensitive to the apparent purpose of the action
• In this way, they represent the actions and (thereby)
represent a common thread of understanding between an
actor and an observer
• A shared link of understanding is what defines a
communicative act, so some have claimed that mirror
neurons are an innate (gestural) communicative
mechanism
What’s it to humans?
• Many have claimed that the proper analogue to
primate F5 is Broca’s area (BA44/45)
– Cytoarchitectonically [in terms of the distribution of
cell structures] this seems to be true
– Broca’s area has been implicated in execution of arm
movements; and during mental imagining and
manipulation of hand movements
– fMRI of grasp observation has found left hemisphere
activity in BA45
What’s it to psycholinguists?
• These findings mesh very nicely with a ‘mind
reading’ theory of language
• Recall that ‘mind reading’ was associated with the
same region: ventral prefrontal cortex
• Some of the work of parsing and attaching
meaning to action in social contexts can be
underlain by mirror neurons
What’s it to psycholinguists?
• These findings mesh very nicely with the motor
theory of speech perception, and with the McGurk
effect)
• Recall that under these theories, speech perception
just is a motor mirroring: to be able to parse
phonology is to be able to parse the gestures of the
articulators
What’s it to psycholinguists?
• These findings mesh very nicely with the finding that
Broca’s area is implicated in language production in other
modalities than speech: i.e. in sign language
• And with the fact that the main deficit after lesion of the
arcuate fasciculus is in essence a deficit in mirroring: an
inability to repeat words
– This is because motor neurons constitute a prelinguistic mapping between between perception and
production
Syntax revisited
• As we have already discussed several times,
syntax and the semantics of action are closely
related
• Syntax assigns and handles roles to subjects and
objects, mirroring in many ways the way the
action unfolded
• The sentence “John hit Mary with his hand”
encodes who who hit, who was hit, and what was
used to carry out the hitting
• These are the kinds of things the mirror to which
the mirror neurons may be sensitive
Verb arguments
• Moreover, to use syntax correctly we need to be able to
understand what kind of arguments go with what kinds
of verbs
– You can ‘grasp a rock’ but you cannot ‘grasp a mountain’;
you can ‘pinch some sand’ but you cannot (usually) ‘pinch
some trees’
– You can also focus attention on the grasping method (‘Joe’s
hand touched her leg.’) or the grasping person (‘Joe touched
her leg’)
– If these distinctions are (at least in early language evolution)
mapped on to pre-existing semantic distinctions, it become
easier to comprehend how language might be able to encode
them
Does mimesis underlie language?