FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W142286355> ?p ?o ?g. }

• W142286355 abstract "Synthetic Brain Imaging of English Past Tense Inflection Gert Westermann (gwestermann@brookes.ac.uk) Nicolas Ruh (nruh@brookes.ac.uk) Department of Psychology, Oxford Brookes University Oxford OX3 0BP, UK Abstract Brain imaging studies of English past tense inflection have found dissociations between regular and irregular verbs, but no coherent picture has emerged to explain how these dissociations arise. Here we use synthetic brain imaging on a neural network model to provide a mechanistic account of how regional dissociations between regular and irregular processing in adults can emerge in a single mechanism system with experience-dependent structural development. We show that these dissociations arise from a combination of different statistical properties of verbs relating to frequency, relationships to other verbs, and phonological complexity. The model generates predictions about the patterning of active brain regions for different verbs that can be tested in future brain imaging studies. Keywords: English past tense; verb inflection; connectionist modeling, synthetic brain imaging, neuroconstructivism Introduction The question of how verb inflections, particularly the English past tense, are represented and processed in the brain has been the subject of intense debate over the past 20 years. This is because the past tense contains regular forms (e.g., look-looked) that seem to obey a linguistic mental rule, and irregular forms (e.g., see-saw) that show properties of associative storage. One view, the Dual-Mechanism or Words-and-Rules Theory (e.g., Pinker, 1999; Pinker & Ullman, 2002) holds that the apparent processing differences between regular and irregular forms are indeed caused by qualitatively different underlying mechanisms: a rule for regulars, and associative storage for irregulars. Another view, closely linked to connectionist approaches to cognitive processing, argues that all forms are processed in a single associative system and that apparent dissociations between regulars and irregulars emerge on the basis of different statistical properties of verbs that affect their ease of processing (e.g., frequency, phonological complexity, number of similar verbs with a similar past tense (friends, e.g., sing and ring), number of similar verbs with a different past tense (enemies, e.g., sing and bring)) or on the basis of different reliance on semantic vs. phonological factors (Joanisse & Seidenberg, 1999; McClelland & Patterson, 2002; Seidenberg & Arnoldussen, 2003; Westermann, 1998). A large amount of empirical and computational work has aimed to provide evidence for either view. One methodology that has been used in several studies is brain imaging with positron emission topography (PET), functional magnetic resonance imaging (fMRI), magnetoencephalography (MEG) and evoked response potentials (ERP). These studies have found differences in brain activation patterns when participants inflect regular and irregular verbs, respectively, leading some researchers to claim that these data provide support for a dual mechanism system, with the rule component and the associative mental lexicon located in different brain regions (Dhond et al., 2003; Jaeger et al., 1996; Lavric et al., 2001). For example, in a seminal study by Jaeger et al. (1996) using PET, participants were asked to generate past tense forms of visually presented monosyllabic verb stems. Results showed that although many brain regions were activated by all verbs, production of regulars selectively activated left dorsolateral prefrontal cortex and left anterior cingulate cortex. Irregulars led to higher overall activation and involved occipital visual processing areas. These systematic differences between both verb types were interpreted by the authors as strong evidence for the dual- mechanism account of inflection. In another study, Dhond et al. (2003) asked participants to covertly generate past tense forms of visually presented verb stems while imaging their brains using MEG. Dhond et al. also found that generation of regulars and irregulars activated many areas in common, but that processing of regulars led to greater activation in left inferior prefrontal areas (Broca’s area), and processing of irregulars preferentially activated left occipitotemporal cortex as well as right dorsolateral prefrontal cortex. These results were interpreted as indicating that regulars activated rule-based grammar regions and irregulars activated areas involved in associative retrieval of forms, corresponding directly to the dual-mechanism theory. However, the results of these and other studies have been controversial. One problem is that specific methodological choices can strongly affect results. For example, because of the low temporal resolution of PET, Jaeger et al. (1996) used a block design in which all regular verbs and all irregular verbs were presented together. However, this design introduces the confound that participants could develop strategies for regular but not for irregular verbs, suggesting that differences between both verb types should be found independently of the nature of the underlying processing mechanisms (Seidenberg & Hoeffner, 1998). Several more recent imaging studies have investigated the possibility that observed activation differences between regular and irregular verbs are due to statistical factors of" @default.
• W142286355 created "2016-06-24" @default.
• W142286355 creator A5024569430 @default.
• W142286355 creator A5063598524 @default.
• W142286355 date "2009-01-01" @default.
• W142286355 modified "2023-09-23" @default.
• W142286355 title "Synthetic Brain Imaging of English Past Tense Inflection" @default.
• W142286355 cites W1209649067 @default.
• W142286355 cites W1553508775 @default.
• W142286355 cites W1556645839 @default.
• W142286355 cites W1601164948 @default.
• W142286355 cites W1604879017 @default.
• W142286355 cites W1966314496 @default.
• W142286355 cites W2021237916 @default.
• W142286355 cites W2026065029 @default.
• W142286355 cites W2035896002 @default.
• W142286355 cites W2037320790 @default.
• W142286355 cites W2052176889 @default.
• W142286355 cites W2079108857 @default.
• W142286355 cites W2098386076 @default.
• W142286355 cites W2099222747 @default.
• W142286355 cites W2101911181 @default.
• W142286355 cites W2124485288 @default.
• W142286355 cites W2141047823 @default.
• W142286355 cites W2147289774 @default.
• W142286355 cites W2148905302 @default.
• W142286355 cites W2154517380 @default.
• W142286355 cites W2158910967 @default.
• W142286355 hasPublicationYear "2009" @default.
• W142286355 type Work @default.
• W142286355 sameAs 142286355 @default.
• W142286355 citedByCount "1" @default.
• W142286355 countsByYear W1422863552012 @default.
• W142286355 crossrefType "journal-article" @default.
• W142286355 hasAuthorship W142286355A5024569430 @default.
• W142286355 hasAuthorship W142286355A5063598524 @default.
• W142286355 hasConcept C111472728 @default.
• W142286355 hasConcept C138885662 @default.
• W142286355 hasConcept C154945302 @default.
• W142286355 hasConcept C15744967 @default.
• W142286355 hasConcept C159403335 @default.
• W142286355 hasConcept C169760540 @default.
• W142286355 hasConcept C169900460 @default.
• W142286355 hasConcept C188147891 @default.
• W142286355 hasConcept C2776397901 @default.
• W142286355 hasConcept C2776626494 @default.
• W142286355 hasConcept C2779281236 @default.
• W142286355 hasConcept C41008148 @default.
• W142286355 hasConcept C41895202 @default.
• W142286355 hasConcept C8521452 @default.
• W142286355 hasConcept C89611455 @default.
• W142286355 hasConceptScore W142286355C111472728 @default.
• W142286355 hasConceptScore W142286355C138885662 @default.
• W142286355 hasConceptScore W142286355C154945302 @default.
• W142286355 hasConceptScore W142286355C15744967 @default.
• W142286355 hasConceptScore W142286355C159403335 @default.
• W142286355 hasConceptScore W142286355C169760540 @default.
• W142286355 hasConceptScore W142286355C169900460 @default.
• W142286355 hasConceptScore W142286355C188147891 @default.
• W142286355 hasConceptScore W142286355C2776397901 @default.
• W142286355 hasConceptScore W142286355C2776626494 @default.
• W142286355 hasConceptScore W142286355C2779281236 @default.
• W142286355 hasConceptScore W142286355C41008148 @default.
• W142286355 hasConceptScore W142286355C41895202 @default.
• W142286355 hasConceptScore W142286355C8521452 @default.
• W142286355 hasConceptScore W142286355C89611455 @default.
• W142286355 hasIssue "31" @default.
• W142286355 hasLocation W1422863551 @default.
• W142286355 hasOpenAccess W142286355 @default.
• W142286355 hasPrimaryLocation W1422863551 @default.
• W142286355 hasRelatedWork W111288913 @default.
• W142286355 hasRelatedWork W1139857265 @default.
• W142286355 hasRelatedWork W1532341290 @default.
• W142286355 hasRelatedWork W1985471064 @default.
• W142286355 hasRelatedWork W199295433 @default.
• W142286355 hasRelatedWork W2046909339 @default.
• W142286355 hasRelatedWork W2097412036 @default.
• W142286355 hasRelatedWork W2107141631 @default.
• W142286355 hasRelatedWork W2110042646 @default.
• W142286355 hasRelatedWork W2123329009 @default.
• W142286355 hasRelatedWork W2282193652 @default.
• W142286355 hasRelatedWork W2334073986 @default.
• W142286355 hasRelatedWork W2337728750 @default.
• W142286355 hasRelatedWork W2577263306 @default.
• W142286355 hasRelatedWork W2751783252 @default.
• W142286355 hasRelatedWork W2783011109 @default.
• W142286355 hasRelatedWork W2885850199 @default.
• W142286355 hasRelatedWork W2896966696 @default.
• W142286355 hasRelatedWork W2924351899 @default.
• W142286355 hasRelatedWork W2991000877 @default.
• W142286355 hasVolume "31" @default.
• W142286355 isParatext "false" @default.
• W142286355 isRetracted "false" @default.
• W142286355 magId "142286355" @default.
• W142286355 workType "article" @default.