Relating Mental Activity to Cognition
Edwin Hutchins, Cognition_in_the_Wild
Ship navigation involves lots of
numbers . Numbers have to be processed in order to find out where the ship is
and especially to determine where it will be. It is easy to assume that
navigators must be good at arithmetic . When I looked
closely at the practice of navigation , however , I found the navigators
engaged in very few arithmetic tasks. How can that be?
It must be evident by now that the
computations performed by the navigation system are not equivalent to the
cognitive tasks facing the individual members of the navigation team. It is
possible to describe the computations performed by the navigation team without
recourse to the cognitive abilities or activities of the individual members of
the team.
The navigation system combines
one-dimensional constraints to fix a ship's position . The members of the navigation
team read scales and translate spoken representations into written ones. The navigation
system computes distance from rate and time , while the members of the team
imagine four –digit numbers as two -digit numbers .
The
computations that are performed by the navigation system are a side effect of
the cognitive activity
of the members of the navigation team. The tools of the trade both define the
tasks that are faced by the navigators and, in their operation , actually carry
out the computations .
As we have seen, the very same
computation can be implemented many ways, each implementation placing vastly
different cognitive demands on the task performer .
I argued
above that the naive notion of these tools as amplifiers of cognitive activity
was mistaken . Is a written procedure an amplifier of memory ? Not if the task
performer never knew the procedure .
Then , and
always , the functional system that performs the task is a constellation of
structured representational media that are brought into coordination with one
another .
These
tools permit us to transform difficult tasks into ones that can be done by
pattern matching , by the manipulation of simple physical systems, or by mental
simulations of manipulations of simple physical systems. These tools are
useful precisely because the cognitive processes required to manipulate them
are not the computational processes accomplished by their manipulation .
The computational constituents of the problem have been built into the physical
structure of the tools .
