Information Technology Reference
In-Depth Information
information of heuristics is gathered only from one row (or one column/box).
And visual module doesn't take much time to encode three digits because it
does not need to shift focus to another dimension. The gathering digits in only
one dimension make participants encode them soon, only one second, as shown
in Table 3. While in complex tasks, heuristic related information disposes on
at least two dimensions and costs more time to encode, and the participants
have to shift focus to another dimension after encoding one. It takes about 1.5
seconds before retrieving all information for heuristics, 0.5 second more than the
simple one, as shown in Table 4. Only related dimensions and valid information
are encoded in the model because encoding other irrelative information will cost
more and is not supported in the model. It is also supposed that the order
of encoding in 4*4 Sudoku is row, column and box. Only valid information in
heuristics is presented in the model.
Goal Module
. Before getting the answer, the state of controlling and transform-
ing cognitive processes is changed by the complexity of heuristics. The basic pro-
cedure of control state in the goal module includes visualizing the information
from input modality through eyes, encoding and integrating the information,
and then retrieving the heuristics to get the answer, pressing key and speaking
the answer out in the end. The complexity of heuristics differs in dimension
number and digits distribution. For simple tasks, the states in the goal module
are simpler than the complex ones in dimension encoding, representation, infor-
mation integration and memory retrieval, as listed in Tables 3 and 4. The tasks
are easy for adult participants because goal space is fixed in 1, 2, 3, 4 and only
different in their orders. By goal status oriented, participants might only attend
the valid information in related dimension and retrieved heuristics to answer the
question.
Retrieval Module
. As mentioned above, the retrieval module retrieves infor-
mation from declarative memory. For the simple tasks, the digits in heuristics
are only in one dimension and the memory burden is least, so the retrieval mod-
ule is fired one time; for the complex ones, heuristic related digits are in two or
three dimensions and the brain maintained the former encoded digits. In this
case, retrieval module is activated two or three times.
Imaginal Module
. The imaginal module in ACT-R is response for transforming
problem representations, such as a change to retrieve digits from one dimension
to another. So it is related to the operation of retrieval module where it is
mainly determined by valid information of heuristics. Note that although there
are some irrespective digits, they are not recalled, such as '3' in part (a) of
Figure 2. One interpret action is that human being always attends and selects
useful information actively in heuristic strategy.
Procedural Module
. In ACT-R architecture, the procedural module takes a
role of conductor or control center in dealing with the communication of in-
formation among modules. The production rules implement the basic units
of procedural knowledge. The scale of production system is decided by the
