Reasons for Beliefs in Understanding: Applications of Non-Monotonic Dependencies to Story Processing
نویسنده
چکیده
Many of the inferences and decisions which contribute to understanding involve fallible assumptions. When these assumptions are undermined, computational models of comprehension should respond rationally. This paper crossbreeds AI research on problem solving and understanding to produce a hybrid model ("reasoned understanding"). In particular, the paper shows how nonmonotonic dependencies [Doyle791 enable a schemabased story processor to adjust to new information requiring the retraction of assumptions.
منابع مشابه
A New Logic of Beliefs: Monotonic and Non-Monotonic Beliefs - Part 1
We present here a new formalization of beliefs, which has a simpler semantics than the previous formalizations, and develop an inference method for it by generalizing the resolution method. The usual prepositional formulas are embedded in our logic as a special type of belief formulas. One can obtain a non-monotonic logic of beliefs by applying, say, circumscription to the basic belieflogic dev...
متن کاملOnline Processing of English Wh-Dependencies by Iranian EFL Learners
To be able to reach the level of ultimate attainment in the second language, learners need to acquire not only the grammar of the L2 but also the language processing mechanisms involved in the comprehension of sentences in real time. Contrary to its importance, very little is known yet about online L2 processing. This study examines whether advanced Iranian learners of English reactivate disloc...
متن کاملEpistemic Conditional Logics
This paper develops two conditional logics that allows one to reason about the belief of another agent (which reasons in conditional logic himself). The desirable properties of such logics that reason about the beliefs of another non-monotonic agent are discussed and the two logics proposed here are shown to have those properties.
متن کاملProcessing Demands Impact 3-Year-Olds’ Performance in a Spontaneous-Response Task: New Evidence for the Processing-Load Account of Early False-Belief Understanding
Prior to age four, children succeed in non-elicited-response false-belief tasks but fail elicited-response false-belief tasks. To explain this discrepancy, the processing-load account argues that the capacity to represent beliefs emerges in infancy, as indicated by early success on non-elicited-response tasks, but that children's ability to demonstrate this capacity depends on the processing de...
متن کاملThieal Nderstandin Ief Conflict In
Belief conflict patterns (BCPs) are knowledge structures representing the understander’s moral attitude toward problematic interpretations of the events in a story. These structures are used to model interest in stories by contrasting the understanding of stories to the system’s beliefs about the characters and what they have done. Once recognized, BCPs provide a framework for interpreting the ...
متن کامل