"Seven...

The paper is concerned with the mechanisms through which children learn the meaning of words. Accordingly, the authors suggest that there are two ways in which the process occurs. The first is through explicit instruction, the second is implicit via the use of verbal contexts. The paper describes a study conducted on a set of children divided into five age groups of twenty-five children each. Each child was tasked with finding the meaning of an artificial word as it appeared in six differing contexts. The results showed that there were two categories of children, according to the way in which they interpreted the artificial words. The first category of children, mostly the younger ones, consisted of those unable to differentiate between the meaning of the single word and the meaning of the entire sentence. That is, they attributed to the entire sentence, their interpretation of the word. An interpretation of this sort is described by the authors as a “sentence-core concept.” As the word was encountered in its varying contexts, the children simply applied this sentence-core concept to fit the new sentence by a process the authors call “assimilation.” The second category of children was able to make the distinction between word meaning and sentence meaning. However, their interpretation of the words lacked situational focus. These interpretations are described as “simple holophrastic concepts.” Because of the nebulous nature of the interpretations, in this case, when the word was encountered in its different contexts the children were able to “expand” it to fit the new contexts. This is termed “holophrastic expansion.” In some cases, there was a more rigid expansion that the authors describe as “pluralization.” The authors mention two other forms of interpretation based on holophrastic expansion: “generalization by juxtaposition,” whereby the children applied a concept from one context to a second context by rationalizing based on spatial relations, and “generalization by chain,” whereby the children performed the same action, rationalizing, however, based on temporal relations. The paper provides statistics describing the categorization of the children, then concludes by suggesting that the data also suggested that there is a link between semantic and grammatical perception.
The age groups involved in the study cited in the paper range from 81/2 to 131/2 years of age. SNePS as a KR&R system, with the learning algorithms, is designed to be a bit more mature in the mechanisms by which it acquires new words. The rationalization process, the level of an inferencing mechanism expected for the CVA project certainly exceeds the levels exhibited by at least half of the children citied in the study. On the relationship between “meaning and structure” that the paper suggests exists, for SNePS this is nothing that a good grammar cannot take care of. In short, the CVA project is concerned with certain levels of competence that a significant percentage of the subjects of the study in this paper are not suitably endowed with at their age level. The paper did indicate that the older children, as expected, did perform better than the younger children, insofar as they did not ascribe to the whole sentence the meaning of the artificial words encountered. But, it was also stated that their interpretations of the words were not “circumscribed.” These children would be candidates for a CVA-SNePS induced curriculum that would assist them better acquire new words from context.

all words found in a lexicon amount to their numeric equivalent in the alphabet. of what practical use this information is is anyone's guess...