What precisely is the universal nature of the human syntactic parser, such that it copes easily with some embedded structures, yet fails so dramatically on others (e.g., classic double center-embeddings)? A theory is proposed in the form of an architecture for parsing based on two simple ideas. The first is that human short-term memory is an indexing structure which can give rise to interference effects (storage limitations) when contents overlap with respect to the indices. For parsing, the contents are syntactic structures, and the indices are potential structural relations. The second idea is that the capacity of STM is the minimum capacity required to support the basic functions of parsing. The theory successfully accounts for the contrasts between over 50 difficult and acceptable constructions from English, French, German, Hebrew, Japanese, Mandarin, and Spanish. The theory has independent psychological and computational motivation, and is a functional part of a broader cognitive ...

The principles of readability are in every style manual. Readability formulas are in every word processor. What is missing is the research and theory on which they stand.

Coordinate structures have traditionally been analyzed as having no internal structure other than the sequencing of their members. * In particular, the possibility that the members of coordinate structures may themselves be coordinate structures has not been widely recognized. Those who have recognized the possibility of such embedding of coordinate structures have assumed that there are no limits on the depth of such embedding, just as there are no limits on the depth of embedding in subordinate structures. However, coordinate-structure embedding in English occurs only in order to prevent coordinate structures from containing distinct connectives (e.g. and and or), distinct junctures (breaks) between members, and sequences of members in which the first is introduced by a connective while the second is not. In order to prevent these conditions from arising, the depth of coordinate-structure embedding does not have to exceed 2. This limitation on coordinate-structure embedding must be dealt with by the grammars of natural languages; it is not simply a performance limitation. Standard generative mechanisms (finite-state, phrasestructure, and transformational grammars, including recently proposed analyses of coordinate structures based on the Minimalist Program) do not provide an adequate account of this limitation. On the other hand, a theory of constraint satisfaction such as Optimality Theory, in which ranked constraints select the optimally structured outputs for given inputs made up of members, connectives and junctures, does do so.

We investigate linguistic features that correlate with the readability of texts for adults with intellectual disabilities (ID). Based on a corpus of texts (including some experimentally measured for comprehension by adults with ID), we analyze the significance of novel discourselevel features related to the cognitive factors underlying our users ’ literacy challenges. We develop and evaluate a tool for automatically rating the readability of texts for these users. Our experiments show that our discourselevel, cognitively-motivated features improve automatic readability assessment. 1

this document is Thesis 1 Human sentence processing is a computation over a grammar whose progress proceeds towards less-uncertain, lower-energy states

Abstract — Necessary for furthering the development of more powerful evolutionary design systems, capable of scaling to evolving more sophisticated and complex artifacts, is the ability to meaningfully and objectively compare these systems by applying complexity measures to the artifacts they evolve. Previously we have proposed measures of modularity, reuse and hierarchy (MR&H), here we compare these measures to ones from the fields of Complexity, Systems Engineering and Computer Programming. In addition, we propose several ways of combining the MR&H measures into a single measure of structure and organization. We compare all of these measures empirically as well as on three sample objects and find that the best measures of complexity are two of the proposed measures of structure and organization. I.

The recent deluge of published sutdies employing sentences or connected discourse as the unit of study has left the question of whether the two types of materials are essentailly similar or importantly different unsolved. An understanding of this issue is crucial to theory, since the existence of major psychological differences between the comprehension, learning, and memory of sentence lists and connected discourse would make a unified theory covering both types of materials exceedingly difficult to formulate. While offering no final resolution of the issue, the present paper examines the evidence, considers the implications of several major theoretical developments, and poses questions amenable to experimentation. It is hoped that the paper will serve as a springboard to a higher level of understanding of how people process these two common types of experimental materials. Sentences

An experiment in the computer generation of coherent discourse was successfully conducted to test a hypothesis about the transitive nature of syntactic dependency relations among elements of the English language. The two primary components of the experimental computer program consisted of a phrase structure generation grammar capable of generating grammatical nonsense, and a monitoring system which would abort the generation process whenever it was apparent that the dependency structure of a sentence being generated was not in harmony with the dependency relations existing in an input source text. The final outputs of the system were coherent paraphrases of the source text. An implication of the hypothesis is that certain types of dependency relations are invariant under a variety of linguistic transformations. Potential applications include automatic kernelizing, question answering, automatic essay writing, and automatic abstracting systems. The question of the validity of transitive dependency models for languages other than English should be explored.

Speech produced by subjects during neuropsychological exams can provide markers other than test performance, via spoken language characteristics that discriminate between subject groups. We present preliminary results on the utility of such markers, automatically derived from spoken responses to narrative recall tests, in discriminating between healthy elderly and subjects with Mild Cognitive Impairment (MCI). Given the audio and transcript of the retellings, a range of markers were automatically derived, including (among others) pause frequency and grammatical complexity. Certain spoken language derived markers, which do not measure the fidelity of the retelling to the original narrative, show statistically significant differences between the group means, when calculated either manually or automatically. 1.

This is the pre-publication manuscript. The published version may slightly differ.