Page 115 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

3. Embodied Cognition 103

Once again, Rizzolatti and Arbib in Language Within Our Grasp [16] provide the
empirical evidence which supports Liberman’s views. This time they find it in mirror
neurons located in Broca’s area. These neurons, they claim, represent the link be-
tween sender and receiver that Liberman postulated in his theory.
Broca’s area is most commonly thought of as an area for speech. PET data indi-
cate that Broca’s area might also become active during the execution of hand or arm
movement, during mental imagery of hand grasping movement and during tasks
involving hand mental rotations. They find further evidence indicating that individ-
uals recognize actions made by others because the neural pattern elicited in their
premotor areas during action observation is similar to that which is internally gener-
ated to produce that action.
In the case of action observation, there is a strong spinal cord inhibition that
selectively blocks motoneurons involved in the observed action execution. Some-
times, however, for example when the observed action is of particular interest, the
premotor system will allow a brief prefix of the movement to be exhibited. This pre-
fix will be recognized by the other individual. This fact will affect both the actor and
the observer. The actor will recognize an intention in the observer, and the observer
will notice that its involuntary response affects the behavior of the actor. When this
occurs, a primitive dialogue between observer and actor is established. This dialogue
forms the core of language. Advertising industry recognized the importance of these
motor prefixes, as evidenced by their use of electric field sensor attached to thumb
muscle to indicate buyer’s interest in a product. It turns out that “grabbing” reflex
betrays the buyer even if he pretends to display lack of interest with his disinterested
facial expression.
Rizzolatti and Arbib argue that (1) the mimetic capability inherent to F5 and Bro-
ca’s area had the potential to produce various types of closed systems related to the
different types of motor fields present in that area (hand, mouth, and larynx); (2) the
first open system to evolve en route to human speech was a manual gestural system
that exploited the observation and execution matching system described earlier; and
(3) that this paved the way for the evolution of the open vocalization system we
know as speech. Congruent with this view is the fact that mimetic capacity, a natural
extension of action recognition, is central to human culture such as dances, games,
and tribal rituals and that the evolution of this capacity was a necessary precursor to
the evolution of language. It has been shown that a newborn can mimic her mother’s
facial expression as early as an hour after birth.
Finally, Esther Thelen [17] claims that in principle there is no difference between
the processes engendering walking, reaching, and looking for hidden objects and
those resulting in mathematics and poetry. In her pioneering psychological studies
of reaching, grasping, and walking, she dismisses the accepted theory that phyloge-
netically old “central pattern generators” (CPG) in brain guide the child’s acquisi-
tion of these motor skills. She then proceeds to dismantle the concept of central
pattern generators in brain with a series of elegant experiments, starting with
walking. A normal infant transitions to weight supporting stepping behaviors at 8e
10 months. Thelen discovered that 7-month-old infants who normally did not step