Subjects were exposed to sentences containing “direct” and “indirect” uses of names and definite descriptions. On a subsequent recognition test incorrect rejections tended to be of sentences involving indirect uses, and false alarms to sentences involving direct uses. This finding is contrary to the predictions of models that suggest indiscriminate substitution of names for descriptions, as do those of Anderson and Bower, and Rumelhart and Norman. The implication is that models of semantic memory must incorporate distinct intensional and extensional representations (...) to avoid semantic distortion. (shrink)

Definite descriptions, I shall argue, have two possible functions. 1] They are used to refer to what a speaker wishes to talk about, but they are also used quite differently. Moreover, a definite description occurring in one and the same sentence may, on different occasions of its use, function in either way. The failure to deal with this duality of function obscures the genuine referring use of definite descriptions. The best known theories of definite descriptions, those of Russell and Strawson, (...) I shall suggest, are both guilty of this. Before discussing this distinction in use, I will mention some features of these theories to which it is especially relevant. (shrink)

The paper explores certain phenomena which arise in stories, conversations, and human activity in general when the plans of two individuals are formed and carried out in an interactive situation. A notation system for representing interacting plans is introduced and applied in the analysis of a small portion of “Hansel and Gretel.” The analysis illustrates how a single actor plan can be modified by the needs of cooperative interaction with others and how cooperative interactive episodes can be transformed and used (...) deceptively by one party in achieving his or her own covert goals. (shrink)

It seems to me that the ingredients of most theories both in Artificial Intelligence and in Psychology have been on the whole too minute, local, and unstructured to account–either practically or phenomenologically–for the effectiveness of common-sense thought. The "chunks" of reasoning, language, memory, and "perception" ought to be larger and more structured; their factual and procedural contents must be more intimately connected in order to explain the apparent power and speed of mental activities.

The goal of this thesis is to model some of the cognitive structures and processes involve d in how people decide what to say in purposeful conversation. The main concern is to show how a speaker's knowledge of his/her hearer influences what s/he says. Utterances in such dialogues, where speakers can be presumed to be speaking for reasons, can best be viewed as the performance of "speech acts" (e.g., requesting). By modeling the process of deciding what to say as one (...) of problem-solving, we show that the same mechanisms used to plan non-linguistic actions are applicable to planning of speech acts. The methodology of this research employs a computer program that integrates beliefs, goals, problem-solving, and reasoning, with the use of language. The program does this by planning speech acts when it determines that to achieve its situational goals, it needs to influence someone else's behavior. The system is able to plan request speech actrions to influence its user's goals, and inform speech acts to influence his/her beliefs. In addition, it is shown that speech acts can be used to achieve subgoals of other such acts. For instance, in addition to requesting that the user perform some action, the system can also decide to supply him/her with information necessary to performing that action -- information that it believes s/he does not already have. Our model is extendable to the more general problem of planning multiparty speech acts -- ones where actions requested by the speaker involve acts to be performed by parties other than the hearer. [MORE]. (shrink)