FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2108157079> ?p ?o ?g. }

• W2108157079 abstract "A Simple Categorisation Model of Anaphor Resolution Andrew J. Stewart ( Andrew.Stewart@Unilever.com) Frederic Gosselin (gosselif@psy.gla.ac.uk) Department of Psychology,52 Hillhead Street, University of Glasgow, Glasgow G12 2QQ, United Kingdom. Abstract In this paper we examine the way in which approaching the task of anaphor resolution as a categorisation problem can shed light on the possible mechanisms underlying pronoun resolution. We formulate a model of anaphor resolution data within SLIP (Strategy Length & Internal Practicability) (Gosselin & Schyns, 1997, 1999), a general categorisation framework. We chiefly focus on pronominal anaphors in this paper but we also report the results of modelling repeat name anaphor reading time data collected by Stewart, Pickering and Sanford (in press). The success of adopting the redefinition of anaphor resolution as a categorisation problem suggests that problems faced by the cognitive system that have been considered unique to particular processing domains might be understood at a more cognitively general level. Introduction In this article we bring together work on categorisation and work on psycholinguistics. We adopt a particular psycholinguistic phenomenon as a case study and examine it within a categorisation framework. We illustrate what a categorisation perspective can offer psycholinguistics in terms of theoretical apparatus. We examine the performance of a model formulated within the SLIP (Strategy Length & Internal Practicability) categorisation framework (Gosselin & Schyns, 1997, 1999), and show that it can account for human behaviour in pronoun resolution, a problem common in language processing. We begin by reviewing existing work on pronoun resolution. Then we move on to our proposal which redefines the task of pronoun resolution as a categorisation problem. Following this we turn to outlining the SLIP framework. Finally, we discuss the consequences of redefining pronoun resolution as a categorisation problem and examine the correspondence between our model’s predictions and experimental data. Existing Psycholinguistic Work on Pronoun Resolution Anaphors are expressions that refer back to characters mentioned in a text. One example of an anaphor is a pronoun. Consider the fragment of sentence (A) up to but including the pronoun ’he’. (A) John blamed Bill because he had damaged John’s car. This pronoun could refer to either character. Based on the information conveyed by the pronoun itself, the only restriction is that it refers to a singular male character. As both potential antecedents match on these features the sentence could plausibly continue like sentence (A) or (B): (B) John blamed Bill because he didn’t really like Bill. In (A) the pronoun is coreferential with the character ’Bill’, while in (B) it is coreferential with the character ’John’. There are a number of cues available in the text to facilitate the process of identifying the appropriate pronominal referent. Grammatical role cues One cue is the grammatical positions occupied by the potential antecedents. The word ’John’ occupies the grammatical subject position, while ’Bill’ occupies the grammatical object position. A number of psychological theories, e.g. Subject Assignment Strategy (Stevenson, Nelson, & Stenning, 1995) and Parallel Function Strategy (Sheldon, 1974), predict a preference to interpret the referentially ambiguous pronoun in the above examples as coreferential with the grammatical subject (although for different reasons). Note that in the examples discussed in this paper the character occupying the grammatical subject position is also the first mentioned character. Gernsbacher (Gernsbacher & Hargreaves, 1988; Gernsbacher, 1989) proposed that the first mentioned character occupies a privileged position in the reader’s discourse model. A similar first mention privilege has been observed in other tasks (e.g. Neath, 1993; Neath & Knoedler, 1994). One of the consequences of the first mention preference found in language comprehension is that later in a sentence it is relatively easy to refer to the first mentioned character. Gender cues Additional to grammatical position information, other cues may also be present. Consider sentences (C) and (D) below. (C) John blamed Mary because she broke the window. (D) John blamed Mary because he was in a bad mood. The gender differentiation between the two characters serves as an additional (strong) cue as to which character the pronoun can refer. However, even under conditions where gender information can unambiguously identify the appropriate pronominal referent, there is much evidence to suggest that the system does not immediately take advantage of this (Stevenson & Vitkovitch, 1986; MacDonald & MacWinney, 1990; Tyler & Marslen- Wilson, 1982). It appears that gender information is treated simply as another cue, not in any way qualitatively distinct from other factors." @default.
• W2108157079 created "2016-06-24" @default.
• W2108157079 creator A5011469121 @default.
• W2108157079 creator A5029220878 @default.
• W2108157079 date "2000-01-01" @default.
• W2108157079 modified "2023-09-26" @default.
• W2108157079 title "A Simple Categorisation Model of Anaphor Resolution." @default.
• W2108157079 cites W1742781230 @default.
• W2108157079 cites W1964248296 @default.
• W2108157079 cites W1966881866 @default.
• W2108157079 cites W1973664604 @default.
• W2108157079 cites W1984919177 @default.
• W2108157079 cites W1985918194 @default.
• W2108157079 cites W2000793767 @default.
• W2108157079 cites W2005127381 @default.
• W2108157079 cites W2016120610 @default.
• W2108157079 cites W2033752890 @default.
• W2108157079 cites W2034840133 @default.
• W2108157079 cites W2040025108 @default.
• W2108157079 cites W2106029544 @default.
• W2108157079 cites W2151666357 @default.
• W2108157079 cites W2154986064 @default.
• W2108157079 cites W2166895198 @default.
• W2108157079 hasPublicationYear "2000" @default.
• W2108157079 type Work @default.
• W2108157079 sameAs 2108157079 @default.
• W2108157079 citedByCount "1" @default.
• W2108157079 crossrefType "proceedings-article" @default.
• W2108157079 hasAuthorship W2108157079A5011469121 @default.
• W2108157079 hasAuthorship W2108157079A5029220878 @default.
• W2108157079 hasConcept C120665830 @default.
• W2108157079 hasConcept C121332964 @default.
• W2108157079 hasConcept C12713177 @default.
• W2108157079 hasConcept C138268822 @default.
• W2108157079 hasConcept C138885662 @default.
• W2108157079 hasConcept C154945302 @default.
• W2108157079 hasConcept C15744967 @default.
• W2108157079 hasConcept C169760540 @default.
• W2108157079 hasConcept C169900460 @default.
• W2108157079 hasConcept C188147891 @default.
• W2108157079 hasConcept C192209626 @default.
• W2108157079 hasConcept C204321447 @default.
• W2108157079 hasConcept C2778551981 @default.
• W2108157079 hasConcept C41008148 @default.
• W2108157079 hasConcept C41895202 @default.
• W2108157079 hasConcept C89267518 @default.
• W2108157079 hasConceptScore W2108157079C120665830 @default.
• W2108157079 hasConceptScore W2108157079C121332964 @default.
• W2108157079 hasConceptScore W2108157079C12713177 @default.
• W2108157079 hasConceptScore W2108157079C138268822 @default.
• W2108157079 hasConceptScore W2108157079C138885662 @default.
• W2108157079 hasConceptScore W2108157079C154945302 @default.
• W2108157079 hasConceptScore W2108157079C15744967 @default.
• W2108157079 hasConceptScore W2108157079C169760540 @default.
• W2108157079 hasConceptScore W2108157079C169900460 @default.
• W2108157079 hasConceptScore W2108157079C188147891 @default.
• W2108157079 hasConceptScore W2108157079C192209626 @default.
• W2108157079 hasConceptScore W2108157079C204321447 @default.
• W2108157079 hasConceptScore W2108157079C2778551981 @default.
• W2108157079 hasConceptScore W2108157079C41008148 @default.
• W2108157079 hasConceptScore W2108157079C41895202 @default.
• W2108157079 hasConceptScore W2108157079C89267518 @default.
• W2108157079 hasIssue "22" @default.
• W2108157079 hasLocation W21081570791 @default.
• W2108157079 hasOpenAccess W2108157079 @default.
• W2108157079 hasPrimaryLocation W21081570791 @default.
• W2108157079 hasRelatedWork W1519642432 @default.
• W2108157079 hasRelatedWork W1525768962 @default.
• W2108157079 hasRelatedWork W1525940556 @default.
• W2108157079 hasRelatedWork W2006694948 @default.
• W2108157079 hasRelatedWork W2015004387 @default.
• W2108157079 hasRelatedWork W208087179 @default.
• W2108157079 hasRelatedWork W2244000429 @default.
• W2108157079 hasRelatedWork W2464295032 @default.
• W2108157079 hasRelatedWork W2576897708 @default.
• W2108157079 hasRelatedWork W2585956289 @default.
• W2108157079 hasRelatedWork W2621777979 @default.
• W2108157079 hasRelatedWork W2766012002 @default.
• W2108157079 hasRelatedWork W2766311268 @default.
• W2108157079 hasRelatedWork W2767738682 @default.
• W2108157079 hasRelatedWork W2777339208 @default.
• W2108157079 hasRelatedWork W3040294733 @default.
• W2108157079 hasRelatedWork W3043050326 @default.
• W2108157079 hasRelatedWork W624689230 @default.
• W2108157079 hasRelatedWork W179487609 @default.
• W2108157079 hasRelatedWork W2526543789 @default.
• W2108157079 hasVolume "22" @default.
• W2108157079 isParatext "false" @default.
• W2108157079 isRetracted "false" @default.
• W2108157079 magId "2108157079" @default.
• W2108157079 workType "article" @default.