The two participants can recall clearly how many points have been heard, but
they fail to recall all the detailed content of those propositional units in the long
complex sentence. It also proves that information is stored in the brain asidea units
(propositional units in a smaller scale) instead of individual words and phonemes,
which echoes Anderson’s (1976), Kintsch’s (1974) and Field’s (2008)findings. On
the other hand, there is a limit in terms of people’sworking memory. The natural
limit of working memory provides test designers with implication for writing test
items. For example, writing items on adjacent information units is not advisable, for
test-takers can hardly parse multiple propositional units at the same time.
8.7.3 Schema
According to Carrell, readers’comprehension is dependent upon linguistic sche-
mata, rhetoric schemata and content schemata (Carrell 1983). Linguistic schemata
and rhetoric schemata belong to schemata of different formality levels that refer to
readers’linguistic knowledge of vocabulary, idioms, grammar, and knowledge of
text types, discourse structure and text organization respectively. As a counterpart
to formal schemata, content schemata refer to readers’knowledge of the subject
matter and content of a text.
In the current project, we use schema to categorize participants’protocols related
to their background knowledge of the topical area and their linguistic knowledge.
For example, one participant reported in his TAP:
The following word is“examples”, or maybe I change it (the answer) into definition.
Because generally speaking, there must be definitionfirst and then examples.
This extract clearly shows the participant is dependent upon his previous schema
on discourse structure pattern of expository texts; therefore, the schema of discourse
structure embedded in his mind assists him in coping with the gap-filling task.
There is another example:
Zhuo: That is why I didn’t understand it. I thought differ can’t follow a person. Something
differs from or, a differs from b...I didn’t know that you can use somebody differ in
blabla...
Likewise, Zhuo’s protocol shows her comprehension breaks down while the
linguistic form of the new information contradicts to her prior linguistic knowledge
and then she fails to integrate the new information into her existing schema.
There is a quite conspicuous example of the impact of schema on participants’
selection of information for processing:
Researcher: So you memorized what passive learners do?
Zhou: I don’t know. Maybe because I am a passive learner, so it’s more familiar to me with
their ways of learning. I don’t know.150 8 Linking Task Demands, Cognitive Processes...