Information Technology Reference
In-Depth Information
8.5.2 Example of Simulating Incubation in a Lexical
Decision Task
Yaniv and Meyer [ 40 ] asked human subjects to perform two tasks sequentially. First,
in a rare-word association task, they showed subjects word definitions that were
weakly associated with their definienda. The subjects had a limited time to find a
definiendum for each definition. If the subject found the definiendum, they were
transferred to a lexical decision task, where they had to classify briefly presented
strings of letters as 'word' or 'non-word'. If the subject did not produce a definiendum,
they were asked to rate their feeling of knowing (FOK) and then started the lexical
decision task also. The results of the second task showed that those definitions that
allowed for the retrieval of the correct definienda by the subjects or generated high
FOKs from the subjects produced priming in the lexical decision task (i.e., faster
reaction times).
According to the EII theory, a rare-word association trial produces a simultane-
ous search at the explicit and the implicit levels (Principle #2). Because the target
association is rare in this task, explicit memory search is not likely to yield a satis-
factory solution within the allotted time (because the existing set of hard constraints
does not necessarily lead to solutions in this case). In contrast, according to EII,
implicit memory search is more likely to retrieve the desired association if given
enough time, because soft constraint satisfaction can allow partial match that can
be iteratively improved. However, implicit memory search is often cut short by the
experimenter who then asks the subject to take part in lexical decision trials (for
those subjects who did not produce a definiendum). At the beginning of the lexical
decision trials, implicit knowledge is still in the same state as it was at the end of
the corresponding rare-word association trial. Hence, if the association was retrieved
or nearly retrieved during the rare-word association trial (i.e., with high FOK), the
preceding memory search is relevant and the target word (related to the definiendum
in question) is thus primed for the lexical decision trial (i.e., leading to faster reaction
times). In contrast, the recognition of unrelated words (distractors) is not affected by
the previous state of implicit knowledge, because the rare-word association trial was
irrelevant to these words.
This conceptual explanation by EII led to a detailed computational model that
produced simulations in line with Yaniv andMeyer's [ 40 ] results. The results of 3,000
simulations with a CLARION-based model are shown in Fig. 8.2 , which capture the
corresponding human data [ 15 ].
8.5.3 Modeling Insight Within EII
According to Wallas [ 38 ], insight, the sudden emergence into consciousness of a
potential solution, is another major stage of creative problem solving, normally fol-
lowing the incubation stage. According to EII, insight results from the transferring
Search WWH ::




Custom Search