Written message production--论文代写范文精选

如果领域知识被认为是一个源文本生产和细化,生产一个文本认为至少有三个其他类型的知识，语言知识，务实知识和程序性知识，触发和管理流程应用于其他三种知识。下面的paper代写范文进行详述。

Written message production as a tool for modeling activities 
In this section our aim is to discuss the characteristics and activities of the message producer. This approach is rather unusual, as it supposes that the act of producing a written message for someone else is by itself a situation of implicit (self-) learning. This learning situation rests on the interaction between domain knowledge and the activity of language production, which allows the mediation of this knowledge. In the next section, we would like to briefly discuss the principal dimensions of the activity of text composition (section 2.3.1.). After this discussion, we try to elaborate on the nature of the interaction between domain knowledge and language production. This interaction will be discussed from two sides: (1) the effect of domain knowledge on the production of a message (section 2.3.2.), as well as (2) the effect of the production of a message on domain knowledge (section 2.3.3.).

Knowledge areas and processes involves in text composition 
If domain knowledge (as in Long Term memory, in a situation model, a mental model or an event model) is considered as a source for text production and elaboration, producing a text supposes at least three other kinds of knowledge: (1) linguistic knowledge: for expressing this knowledge; (2) pragmatic knowledge: allowing a writer to elaborate a representation of a potential and missing reader and (3) procedural knowledge: to trigger and to manage processes applied on the three other kinds of knowledge.

More than twenty years ago, Hayes and Flower (1980) proposed a first model (one probably should use the term "blue print") about the nature and the architecture of writing processes and the implied representations underlying text writing activity. They distinguished three major processes, corresponding to the three main stages of writing: Planning, Translating and Reviewing. The dynamics of the activity are postulated as sequential, with possible recursion, according to the considered phase of production (implying a process called Monitor). Within this model, Planning serves to establish an outline of the content of the text to be written (things to say, to whom, in what order, how), which implies retrieving content from memory (Generating), and Organizing this content in function of the instruction and the writer's communicative and processing goals (Goal setting). 

The Translating process serves to semantically develop each part of the plan and translation into linguistically elaborated sentences. Reviewing, finally, implies rereading what has been written and eventually modifying (Editing) the text content and/or form. According to this ‘classic’ model, the production of a text is described as a progressive transformation of a multidimensional reserve of domain knowledge into a linear linguistic trace. Obviously, many factors affect this transformation and its efficiency, making it important to be able to handle (in working memory) many constraints and processes at the same time (Cf. Flower and Hayes, 1980; Kellogg, 1996), and evoking questions about the effects on the process of the amount and quality of different types of available knowledge. With respect to this last point, a crucial part of written language production is played by the writer’s level of expertise of the domain, facilitating not only the production of wellstructured text content, but also the planning process as a whole.

The effect of domain expertise on text production: “ce qui se conçoit bien s’énonce clairement” 
Beginners and experts do not only differ in the quantity of prior knowledge that they possess, but also in the organization of that knowledge. All knowledge acquisition, and hence the development of expertise in a given domain, requires and depends on a reorganization (restructuring) of existing knowledge, represented (e.g. in Long Term Memory), or activated (e.g. in Working Memory), relevant to the domain in question. For example, ADDINEgan and Schwartz (1979)ADDIN compared the recall of diagrams of electronic circuits by novices and experts in the domain. They showed that the content of the products of experts was much more structured, at the semantic level, than that of novices. This phenomenon seemed to be related to a better structure or organization of this knowledge in memory by the experts. 

The effect of the development of expertise, which has been examined across a great number of tasks as different as chess playing (ADDINChase and Simon, 1973; De Groot, 1965)ADDIN, playing Go (ADDINReitman, 1976)ADDIN, or computer programming (ADDINMc Keithen, Reitman, Rueter and Hirtle, 1981)ADDIN, involves the creation of a great number of semantic relations or associations (chunks - units of knowledge at some higher level of conceptual abstraction - ADDINMiller, 1956; Simon, 1974ADDIN), between the different units of knowledge that characterize the domain in question. The development of expertise can essentially be described in terms of a growth of the number and size of "chunks", which allows an expert to simultaneously access and process a greater number of information units united by a similar conceptual category, and even to create new conceptual categories. This development does not only allow the activation of a very elaborated representation in Working Memory, but also, according to ADDINChi, Glaser and Farr (1988)ADDIN, the establishment in Long Term Memory of more and more hierarchically structured knowledge units. 

In this light, Caillies, Denhiere and Jhean-Larose (1999) suggest that the knowledge of advanced subjects in a domain is organized in a hierarchical goal-subgoal structure, whereas that of intermediates and beginners is organized in a temporocausal chain. Hence, to be an expert in (or being more familiar with) a domain, supposes the existence of a knowledge representation that is denser, in terms of the number of knowledge units, but also contains a greater number of semantic relations between those knowledge units. The hierarchic organization of the units supposes that the lower level units are attached to higher level units, consisting of groups of related concepts that form cohorts of semantic categories ADDIN(Rosch, 1973; Rosch, Gray, Johnson and Boyes-Braem, 1976; Rosch and Mervis, 1975)ADDIN.

Structure of knowledge and structuring of text content According to Caccamise (1987), the structure of domain knowledge is more hierarchically organized when the domain is more familiar to the writer (as seen above), and this has two types of effects on the activity of writing a text: the products assembled in Working Memory will be (1) more extended, because of the greater number of associations between chunks, and (2) already organized in terms of semantic or conceptual relationships. These two characteristics considerably limit the necessity for additional writing processes of organizing knowledge and they permit to generate text content that is already highly structured and hierarchically organized. From this perspective, Caccamise (1987) had 16 psychology students compose texts on the basis of different topics, manipulating the familiarity and generality of the topics. 

To find an effect of the topic manipulation on the textual level and the underlying processes, Caccamise analyzed in detail the hierarchical organization of the texts produced in each condition. First, the text content was cut into idea units, and their organization was analyzed in terms of chunks (grouping of units of several related ideas). Chunks could be organized in terms of different hierarchical levels, and could eventually be clustered (groups of related chunks). By listing text content in terms of the number of idea units, the number of chunks, the number of idea units per chunk, the hierarchical level of chunks, the number of clusters, etc., the author was able to show that in the case of greater domain expertise, text content can be characterized not only in terms of a greater number of idea units, but also in terms of a stronger hierarchical organization. 

This organization displayed a greater number of chunks and clusters, related in terms of superordination and subordination. This facilitation was to a great extent due to the structural characteristics of the domain knowledge that could directly furnish the organization of semantic relationships used during text planning and content elaborationADDIN (Schneider, Korkel and Weinert, 1989). More precisely, the hierarchical organization of knowledge by experts allowed them to access greater and more extensive amounts of knowledge, which are already related semantically or conceptually, for further processing by the cognitive system.(论文代写)

51Due网站原创范文除特殊说明外一切图文著作权归51Due所有；未经51Due官方授权谢绝任何用途转载或刊发于媒体。如发生侵犯著作权现象，51Due保留一切法律追诉权。(论文代写)
更多论文代写范文欢迎访问我们主页 www.51due.com 当然有论文代写需求可以和我们24小时在线客服 QQ:800020041 联系交流。-X(论文代写)