Time underlies many interesting human behaviors. Thus, the question of how to represent time in connectionist models is very important. One approach is to represent time implicitly by its effects on processing rather than explicitly (as in a spatial representation). The current report develops a proposal along these lines first described by Jordan (1986) which involves the use of recurrent links in order to provide networks with a dynamic memory. In this approach, hidden unit patterns are fed back to themselves; the internal representations which develop thus reflect task demands in the context of prior internal states. A set of simulations is reported which range from relatively simple problems (temporal version of XOR) to discovering syntactic/semantic features for words. The networks are able to learn interesting internal representations which incorporate task demands with memory demands; indeed, in this approach the notion of memory is inextricably bound up with task processing. These representations reveal a rich structure, which allows them to be highly context-dependent while also expressing generalizations across classes of items. These representations suggest a method for representing lexical categories and the type/token distinction.

Pinker, S. & Bloom, P. (1990). Natural language and natural selection. Behavioral and Brain Sciences 13

Abstract not found

Contents 1 Language Complexity in Generative Grammar 3 Part I The Descriptive Complexity of Strongly Context-Free Languages 11 2 Introduction to Part I 13 3 Trees as Elementary Structures 15 4 L 2 K;P and SnS 25 5 Definability and Non-Definability in L 2 K;P 35 6 Conclusion of Part I 57 DRAFT 2 / Contents Part II The Generative Capacity of GB Theories 59 7 Introduction to Part II 61 8 The Fundamental Structures of GB Theories 69 9 GB and Non-definability in L 2 K;P 79 10 Formalizing X-Bar Theory 93 11 The Lexicon, Subcategorization, Theta-theory, and Case Theory 111 12 Binding and Control 119 13 Chains 131 14 Reconstruction 157 15 Limitations of the Interpretation 173 16 Conclusion of Part II 179 A Index of Definitions 183 Bibliography DRAFT 1<

While there is no interesting limitation on the degree of right-embedding in acceptable sentences, center-embedding is quite severely restricted. Similarly, while there is no interesting bound on the number of nouns that can occur in acceptable noun compounds, there is a very low bound on the number of causative morphemes that can occur in the verb compounds of agglutinative languages. Turning to the clause-final verb clusters of West Germanic languages, we find another similar bound. A cluster including verbs from one embedded clause may beacceptable, but clusters formed from the verbs of two or three or even more deeply embedded clauses are much more awkward (regardless of whether the subject-verb dependencies are crossing or nested). And in languages that allow multiple wh-extractions from a single clause, extractions of more than one element with a given case quickly become unacceptable. More careful experimental study of the nature of these limitations is needed, in a range of languages, but here a preliminary attempt is made to subsume them all under a single generalization, a version of the familar idea that the human parsing

Recent developments in connectionist architectures for symbolic computation have made it possible to investigate parsing in a connectionist network while still taking advantage of the large body of work on parsing in symbolic frameworks. The work discussed here investigates syntactic parsing in the temporal synchrony variable binding model of symbolic computation in a connectionist network. This computational architecture solves the basic problem with previous connectionist architectures, while keeping their advantages. However, the architecture does have some limitations, which impose constraints on parsing in this architecture. Despite these constraints, the architecture is computationally adequate for syntactic parsing. In addition, the constraints make some significant linguistic predictions. These arguments are made using a specific parsing model. The extensive use of partial descriptions of phrase structure trees is crucial to the ability of this model to recover the syntactic st...

Abstract not found

Recent llnguistic theories cast surface complexity cs the result of interactl. ng subsys- tems of constraints. For instance, thc ID/LP grammar fonntdism separa,tcs constreints on immediate dominance from those on linear order. Shitbet (1983) has shown how to carry out direct paring of ID/LP grammars. His algorithm uses ID mid LP constraints directly in langmrge processing, without expmidhg them into a context-free "object gra.mmar." This report exmnines the computational diculty of ID/LP parsing. ShicbeFs purported O(]GI . n ) runtime bound mderestlmates the difficulty of ID/LP parsing; the worst-case runtime of his algorithm is exponential in grammar size. A reduction of the vertex-cover problem proves that ID/LP parsing is NP-complete. The growth of intcrna.1 data struc- tures is the source of difficulty in Shieber's algorithm. The computational and linguistic implicatimxs of these results are discussed. Despite the potentid for combinatorial explo- sion, Shicber's algorithm roma.ins better than the alternative of pa'rs]ng an expanded object gramma.r.

this article represents a step toward just such a rethinking of the linguistic endeavor.

We report the results from three cross-modal lexical decision experiments investigating antecedent priming effects in Japanese. In the first two experiments we examined antecedent reactivation at the preverbal trace position in long-distance scrambling sentences. We found an interaction between the participants' working memory (WM) span and antecedent priming. For the High Span Group, the magnitude of antecedent priming at the trace position was significantly larger than at the earlier control position; for the Low Span Group, on the other hand, there was no such difference. In a third experiment, we examined whether similar reactivation effects cotfid be observed for argument expressions that are not base-generated adjacent to the verb. Contrary to scrambled objects, subject NPs in canonically ordered sentences were not reactivated at the preverbal test position in either of the two participant groups. We argue that the priming effect observed in the High Span Group supports a trace-based account of long-distance scrambling. The degree of complexity of the experimental sentences was such, however, that they exceeded the memory span of the Low Span Group. We conclude that argument traces access their antecedents irrespective of the position of their subcategorizers.