@article{LCNL:1, author={Todd R. Haskell and Maryellen C. MacDonald and Mark S. Seidenberg}, year={2003}, title={Language learning and innateness: Some implications of compounds research.}, journal={Cognitive Psychology}, volume={47}, pages={119-163}, comments={The fact that an eater of rats is a rat-eater, not a *rats-eater has been repeatedly cited as evidence for the level-ordering theory of the lexicon and the Pinker theory of the past tense. We show that the facts about rat-eating are not as commonly portrayed and that constraints on plurals as modifiers in compounds are graded in nature, arising from phonological and semantic factors.}, abstract={In noun compounds in English, the modifying noun may be singular (mouse-eater) or an irregularly inflected plural (mice-eater), but regularly inflected plurals are dispreferred (*rats-eater). This phenomenon has been taken as strong evidence for dual-mechanism theories of lexical representations, which hold that regular (rule-governed) and irregular (exception) items are generated by qualitatively different and innately specified mechanisms. Using corpus analyses, behavioral studies, and computational modeling, we show that the rule-versus-exceptions approach makes a number of incorrect predictions. We propose a new account in which the acceptability of modifiers is determined by a constraint satisfaction process modulated by semantic, phonological, and other factors. The constraints are acquired by the child via general purpose learning algorithms, based on noun compounds and other constructions in the input. The account obviates the regular/irregular dichotomy while simultaneously providing a superior account of the data.}, language={English}, keywords={language learning; language development; compound nouns; lexical representations; dual mechanism theory; rule learning; innateness; morphology; adjectives; phonology; models; English; Grammar; Nouns; Cognitive Hypothesis Testing; Human Information Storage; Inflection; Syntax}, issn={00100285}, URL={http://lcnl.wisc.edu/publications/archive/1.pdf}, } @article{LCNL:2, author={Mark S. Seidenberg and Maryellen C. MacDonald and Jenny R. Saffran}, year={2003}, title={Are there limits to statistical learning?}, journal={Science}, volume={300}, pages={53-54}, comments={The Seidenberg, MacDonald, and Saffran (2002) article elicited a response from Marcus and Berendt, included here along with our response.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/2.pdf}, } @article{LCNL:3, author={Mark S. Seidenberg and Maryellen C. MacDonald and Jenny R. Saffran}, year={2002}, title={Does grammar start where statistics stop?}, journal={Science}, volume={298}, pages={553-554}, comments={A "perspective" on an article by Pena et al. claiming to have demonstrated distinct statistical and grammatical mechanisms in learning language. Summary: not so fast.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/3.pdf}, } @article{LCNL:4, author={Mark S. Seidenberg and Alan Petersen and Maryellen C. MacDonald and David C. Plaut}, year={1996}, month={01}, title={Pseudohomophone effects and models of word recognition. }, journal={Journal of Experimental Psychology: Learning, Memory, and Cognition}, volume={22}, number={1}, pages={48-62}, abstract={Two experiments examined factors that influence the processing of pseudohomophones (nonwords such as brahe orjoak, which sound like words) and nonpseudohomophones (such as brone andjoap, which do not sound like words). In Experiment 1, pseudohomophones yielded faster naming latencies and slower lexical-decision lateneies than did nonpseudohomophones, replicating results of R. S. McCann and D. Besner (1987) and R. S. McCann, D. Besner, and E. Davelaar (1988). The magnitude of the effect was related to subjects' speed in lexical decision but not naming. In Experiment 2, both immediate and delayed naming conditions were used. There was again a significant pseudohomophone effect that did not change in magnitude across conditions. These results indicate that pseudohomophone effects in the lexical-decision and naming tasks have different bases. In lexical decision, they reflect the pseudohomophone's activation of phonological and semantic information associated with words. In naming, they reflect differences in ease of articulating familiar versus unfamiliar pronunciations. Implications of these results concerning models of word recognition are discussed, focusing on how pseudohomophone effects can arise within models that do not incorporate word-specific representations, such as the M. S. Seidenberg and J. L. McClelland (1989) model.}, language={English}, keywords={word frequency, pronunciation of pseudohomophones, college students, implications for models of word recognition; Phonology; Pronunciation; Word Recognition; Words (Phonetic Units); Oral Reading}, issn={02787393}, URL={http://lcnl.wisc.edu/publications/archive/4.pdf}, } @article{LCNL:5, author={Mark S. Seidenberg and Maryellen C. MacDonald}, year={1999}, month={09}, title={A probabilistic constraints approach to language acquisition and processing.}, journal={Cognitive Science}, volume={23}, number={4}, pages={569-588}, abstract={Provides an overview of a probabilistic constraints framework for thinking about language acquisition and processing. The generative approach characterizes knowledge of language (i.e., competence grammar) and then asks how this knowledge is acquired and used. The approach is performance oriented: the goal is to explain how people comprehend and produce utterances and how children acquire this skill. Use of language involves exploiting multiple probabilistic constraints over various types of linguistic and nonlinguistic information. Acquisition is the process of accumulating this information, which begins in infancy. The constraint satisfaction processes that are central to language use are the same as the bootstrapping processes that provide entry to language for the child. Framing questions about acquisition in terms of models of adult performance unifies the two topics under a set of common principles and has important consequences for arguments concerning language learnability.}, language={English}, keywords={probabilitic constraints; language acquisition; language processing; Language Development}, issn={03640213}, URL={http://lcnl.wisc.edu/publications/archive/5.pdf}, } @article{LCNL:6, author={Jelena Mirković and Maryellen C. MacDonald and Mark S. Seidenberg}, year={2005}, month={02}, title={Where does gender come from? Evidence from a complex inflectional system.}, journal={Language and Cognitive Processes}, volume={20}, number={1}, pages={139-168}, comments={Much of the debate about the proper treatment of inflectional morphology has focused on English, which has a dreadfully impoverished inflectional system. More complex system, such as the one for Serbian, which encodes number, gender, and case, are difficult to even describe in rules. We present a connectionist model in which acquiring this system is treated as a statistical learning problem. Gender, in this system, is a graded property of words, which is acquired through exposure to a large vocabulary. This contrasts with other theories which employ explicit gender nodes.}, abstract={Although inflectional morphology has been the focus of considerable debate in recent years, most research has focused on English, which has a much simpler inflectional system than in many other languages. We have been studying Serbian, which has a complex inflectional system that encodes gender, number, and case. The present study investigated the representation of gender. In standard theories of language production, gender is treated as an abstract syntactic feature segregated from semantic and phonological factors. However, we describe corpus analyses and computational models which indicate that gender is correlated with semantic and phonological information, consistent with other cross-linguistic studies. The research supports the idea that gender representations emerge in the course of learning to map from an intended message to a phonological representation. Implications for models of speech production are discussed.}, language={English}, keywords={inflectional morphology; syntactic feature; semantic factors; phonological factors; gender; Inflection; Language; Morphology (Language); Phonology; Syntax; Gender Identity}, issn={01690965}, URL={http://lcnl.wisc.edu/publications/archive/6.pdf}, } @article{LCNL:7, author={Maryellen C. MacDonald and Neal J. Pearlmutter and Mark S. Seidenberg}, year={1994}, month={10}, title={The lexical nature of syntactic ambiguity resolution.}, journal={Psychological Review}, volume={101}, number={4}, pages={676-703}, comments={A short overview of the bigger picture.}, abstract={Ambiguity resolution is a central problem in language comprehension. Lexical and syntactic ambiguities are standardly assumed to involve different types of knowledge representations and be resolved by different mechanisms. An alternative account is provided in which both types of ambiguity derive from aspects of lexical representation and are resolved by the same processing mechanisms. Reinterpreting syntactic ambiguity resolution as a form of lexical ambiguity resolution obviates the need for special parsing principles to account for syntactic interpretation preferences, reconciles a number of apparently conflicting results concerning the roles of lexical and contextual information in sentence processing, explains differences among ambiguities in terms of ease of resolution, and provides a more unified account of language comprehension than was previously available.}, language={English}, keywords={lexical & syntactic ambiguities, resolution & processing in sentence & language comprehension; Comprehension; Language; Lexical Access; Stimulus Ambiguity; Syntax; Sentence Comprehension}, issn={0033295X}, URL={http://lcnl.wisc.edu/publications/archive/7.pdf}, } @article{LCNL:96, author={Silvia P. Gennari and Maryellen C. MacDonald and Bradley R. Postle and Mark S. Seidenberg}, year={2007}, title={Context-dependent interpretation of words: Evidence for interactive neural processes}, journal={NeuroImage}, volume={35}, publisher={Elsevier}, pages={1278-1286}, URL={http://lcnl.wisc.edu/publications/archive/96.pdf}, } @article{LCNL:8, author={Robert Thornton and Maryellen C. MacDonald and Jeniffer E. Arnold}, year={2000}, title={The concomitant effects of phrase length and informational content in sentence comprehension.}, journal={Journal of Psycholinguistic Research}, volume={29}, pages={195-203}, abstract={Recent evidence suggests that phrase length plays a crucial role in modification ambiguities. Using a self-paced reading task with 64 college students, these results were extended by examining the additional pragmatic effects that length manipulations may exert. The construction used was the mixed-phase modification ambiguity, which consists of a verb phrase (VP), followed by an noun phrase (NP), and a pronoun phrase (PP) which could modify the VP (distant site) or the NP (local site). The results demonstrate that length not only modulates modification preferences directly, but that it also necessarily changes the informational content of a sentence, which itself affects modification preferences. These findings suggest that the same length manipulation affects multiple sources of constraints, both structural and pragmatic, which can each exert differing effects on processing.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/8.pdf}, } @article{LCNL:9, author={Maryellen C. MacDonald and Amit Almor and Victor W. Henderson and Daniel Kempler and Elaine S. Andersen}, year={2001}, month={07}, title={Assessing working memory and language comprehension in Alzheimer's Disease.}, journal={Brain and Language}, volume={78}, number={1}, pages={17-42}, abstract={Studies of language impairments in patients with Alzheimer's disease have often assumed that impairments in linguistic working memory underlie comprehension deficits. Assessment of this hypothesis has been hindered both by vagueness of key terms such as "working memory" and by limitations of available working memory tasks, in that many such tasks either seem to have little relationship to language comprehension or are too confusing or difficult for Alzheimer's patients. Four experiments investigated the usefulness of digit ordering, a new task assessing linguistic working memory and/or language processing skill, in normal adults and patients with probable Alzheimer's disease. The digit ordering task was shown to be strongly correlated with the degree of dementia in Alzheimer\'s patients. The task correlated with measures of language processing on which patients and normal controls performed differently. The results are interpreted as indicating that linguistic representations, linguistic processing, and linguistic working memory are intertwined, such that a deficit of one (e.g., working memory) cannot be said to "cause" a deficit in the other. The implications of this approach are explored in terms of task demands in comprehension and memory measures, and interpretation of previous results in the literature.}, language={English}, keywords={digit ordering; linguistic working memory; language comprehension; language processing skill; Alzheimer's disease; dementia; assessment; Alzheimers Disease; Digit Span Testing; Short Term Memory; Verbal Comprehension; Verbal Ability; Verbal Memory}, issn={0093934X}, URL={http://lcnl.wisc.edu/publications/archive/9.pdf}, } @article{LCNL:10, author={Maryellen C. MacDonald and Morten H. Christiansen}, year={2002}, month={01}, title={Reassessing working memory: A comment on Just & Carpenter (1992) and Waters & Caplan (1996). }, journal={Psychological Review}, volume={109}, number={1}, pages={35-54}, abstract={M. A. Just and P. A. Carpenter's (1992) capacity theory of comprehension posits a linguistic working memory functionally separated from the representation of linguistic knowledge. G. S. Waters and D. Caplan’s (1996) critique of this approach retained the notion of a separate working memory. In this article, the authors present an alternative account motivated by a connectionist approach to language comprehension. In their view, processing capacity emerges from network architecture and experience and is not a primitive that can vary independently. Individual differences in comprehension do not stem from variations in a separate working memory capacity; instead they emerge from an interaction of biological factors and language experience. This alternative is argued to provide a superior account of comprehension results previously attributed to a separate working memory capacity.}, language={English}, keywords={working memory; linguistic knowledge; theory of comprehension; language comprehension; connectionist approach; Connectionism; Language Development; Short Term Memory; Theories}, issn={0033295X}, URL={http://lcnl.wisc.edu/publications/archive/10.pdf}, } @article{LCNL:11, author={Robert Thornton and Maryellen C. MacDonald and Mariela Gil}, year={1999}, month={11}, title={Pragmatic constraint on the interpretation of complex noun phrases in Spanish and English. }, journal={Journal of Experimental Psychology: Learning, Memory & Cognition}, volume={25}, number={6}, pages={1347-1365}, abstract={Four experiments examined the role of a pragmatic constraint, the modifiability of noun phrases (NPs), in the modification of complex NPs. Experiment 1 demonstrated that NPs that had received relatively specific prior modification were less likely to take additional modification than NPs with less specific modification. This effect was obtained in both Spanish and English using 2 off-line tasks. Experiments 2 and 3 demonstrated on-line modifiability effects for both languages using a self-paced reading task. The results further suggest that although these Spanish and English speaking college students may have opposing modification preferences, modifiability constrained their interpretations in the same direction. The results of Experiment 4 suggest that discrepancies between the off-line results from Experiment 1 and the on-line results from Experiment 3 may be due to task differences. Implications are discussed in relation to current models of sentence processing.}, language={English}, keywords={role of pragmatic constraint & modifiability of noun phrases, interpretation of ambiguous & complex noun phrases, English & Spanish speaking college students; Nouns; Phrases; Pragmatics; Stimulus Ambiguity; Stimulus Complexity}, issn={02787393}, URL={http://lcnl.wisc.edu/publications/archive/11.pdf}, } @article{LCNL:12, author={Amit Almor and Maryellen C. MacDonald and Daniel Kempler and Elaine S. Andersen and Lorraine K. Tyler}, year={2001}, month={02}, title={Comprehension of long distance number agreement in probable Alzheimer's Disease.}, journal={Language and Cognitive Processes}, volume={16}, pages={35-63}, abstract={Two cross-modal naming experiments examined the role of working memory in processing sentences and discourses of various lengths. In Exp 1, 10 memory impaired patients (aged 76-89 yrs) with probable Alzheimer's disease (AD) and 10 healthy elderly controls showed similar sensitivity to violations of subject-verb number agreement in a short sentence condition and similar degradation to this sensitivity in a long sentence condition. Performance in neither length condition correlated with performance on working memory tasks. In Exp 2, the same 10 AD Ss were less sensitive than the 10 controls to pronoun-antecedent number agreement violations in a short discourse condition, but neither group was affected by additional length. In this experiment, performance in both the short and the long conditions correlated with working memory performance. These results show that grammatical and discourse dependencies pose different memory and processing demands, and that these differences are not simply due to differences in the amount of intervening material between dependent words. Furthermore, while the working memory deficits characteristic of AD do not interfere with on-line grammatical processing within sentences, they do compromise on-line discourse processing across sentences.}, language={English}, keywords={role of working memory in processing sentences & observing grammar rules 76-89 yr olds with memory impairments & probable Alzheimer's disease; Alzheimers Disease; Grammar; Semantic Memory; Sentence Comprehension; Short Term Memory; Geriatric Patients; Sentence Structure}, issn={01690965}, URL={http://lcnl.wisc.edu/publications/archive/12.pdf}, } @article{LCNL:13, author={Robert Thornton and Maryellen C. MacDonald}, year={2003}, title={Plausibility and grammatical agreement.}, journal={Journal of Memory and Language}, volume={48}, pages={740-759}, abstract={Three experiments examined plausibility effects on the production and comprehension of subject-verb agreement. In a production task, participants were given a verb and sentence preamble and asked to create a complete passive sentence. The preambles contained two nouns (e.g., the album by the classical composers). The plausibility of the verb was manipulated so that either (a) both nouns could be plausible passive subjects (e.g., praised, as both albums and composers can plausibly be praised) or (b) only the head noun could be a plausible subject (e.g., played, as only albums can plausibly be played). The comprehension task was self-paced reading with the same materials. The results from both methodologies demonstrated robust plausibility effects. There were higher agreement error rates in production and longer RTs at the verb in comprehension when both nouns were plausible subjects than when only the head was plausible. Implications for current production models are considered and an alternative account is presented that is motivated by current comprehension models and other recent production data.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/13.pdf}, } @article{LCNL:14, author={Todd R. Haskell and Maryellen C. MacDonald}, year={2003}, title={Conflicting cues and competition in subject-verb agreement.}, journal={Journal of Memory and Language}, volume={48}, pages={760-778}, abstract={Traditional theories of agreement production assume that verb agreement is an essentially syntactic process. However, recent work shows that agreement is subject to a variety of influences both syntactic and non-syntactic, which raises the question of how these different sources of information are integrated during agreement production. We propose an account of agreement production in which several information sources contribute activation to singular and plural verb forms. Conflict between cues leads to competition which can in turn magnify the influence of subtle cues. Three fragment completion experiments tested key predictions of this constraint satisfaction approach. Experiment 1 demonstrated competition effects on verb choice and sentence initiation latencies. Experiments 2and 3 demonstrated that conflicts between semantic and grammatical cues allow morphological regularity to exert a small but detectable effect on agreement. These results suggest that the constraint-satisfaction framework may provide a productive approach for understanding agreement production.}, language={English}, keywords={Agreement; Number; Language production}, URL={http://lcnl.wisc.edu/publications/archive/14.pdf}, } @inproceedings{LCNL:15, author={David S. Race and Maryellen C. MacDonald}, year={2003}, title={The use of "that" in the production and comprehension of object relative clauses.}, booktitle={Proceedings of the 25th Annual Meeting of the Cognitive Science Society}, abstract={We explore the interplay between production and comprehension by investigating why producers insert or omit the function word "that" in Object Relative Clauses, and how this choice affects comprehension. We present data from three experiments which suggests that producers insert "that" to alleviate production difficulty and in doing so create a distributional pattern of "that" use. Comprehenders are shown to be sensitive to these patterns. Implications for the interaction of comprehension and production processes are discussed.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/15.pdf}, } @article{LCNL:16, author={Lynne M. Stallings and Maryellen C. MacDonald and Padraig G. O'Seaghdha}, year={1998}, month={10}, title={Phrasal ordering constraints in sentence production: Phrase length and verb disposition in heavy-NP shift }, journal={Journal of Memory and Language}, volume={39}, number={3}, pages={392-417}, abstract={Heavy-NP shift is the tendency for speakers to place long or “heavy” noun phrase direct objects at the end of a sentence rather than in the canonical postverbal position. Three experiments using several task variations confirmed that length of the noun phrase influenced the ordering of the noun phrase and prepositional phrase during production. We also found that heavy-NP shift was strongly constrained by the “shifting disposition” of individual verbs. Verbs that do not require their complements (e.g., sentential complements) to appear in an adjacent position yielded more shifting during production than did verbs that more frequently appear adjacent to their complements. Analyses of decision/preparation times suggested that shifted and unshifted structures competed for selection. These findings point to the simultaneous activation of lexically derived syntactic representations and ordering options in sentence planning. A multiple constraints framework provides a means of reconciling the existence of competition among ordering options with incremental sentence construction.}, language={English}, keywords={phrase length & verb disposition in heavy direct object noun phrase as phrasal ordering constraints in cued recall task, sentence production, college students; Cued Recall; Phrases; Sentence Structure; Speech Development; Nouns; Verbs}, issn={0749596X}, URL={http://lcnl.wisc.edu/publications/archive/16.pdf}, } @article{LCNL:17, author={Mark S. Seidenberg and Laura M. Gonnerman}, year={2000}, month={09}, title={Explaining derivational morphology as the convergence of codes.}, journal={Trends in Cognitive Sciences}, volume={4}, number={9}, pages={353-361}, comments={A theoretical discussion piece in which we propose that derivational morphology can be treated as a statistical phenomenon resulting from the convergence of orthographic, phonological, semantic, and other codes across words, rather than a discrete level of representation.}, abstract={Language users have a remarkable ability to create, produce, and comprehend complex words. Words such as UNDERCUT and BAKERY appear to be composed of units, traditionally called morphemes, that recombine in rule-like ways to form other words, such as UNDERLINE and CANNERY. However, morphological systems are quasiregular: they are systematic and productive but admit many seemingly irregular forms. Thus, BAKERY is related to BAKE and CANNERY to CAN but what is the GROCE in GROCERY? There is no bread in SWEETBREADS, liver in DELIVER, corn in CORNER, or ginger in GINGERLY. Such words exhibit partial regularities concerning the correspondences between form and meaning, the treatment of which has important implications for linguistic and psycholinguistic theories. This article describes an approach to morphological phenomena called the convergence theory, in which morphology is a graded, interlevel representation that reflects correlations among orthography, phonology, and semantics.}, language={English}, keywords={derivational morphology; convergence theory; morphological systems; linguistics; psycholinguistics; complex words; orthography; phonology; semantics; morphemes; Morphology (Language); Words (Phonetic Units); Language; Theories}, issn={13646613}, URL={http://lcnl.wisc.edu/publications/archive/17.pdf}, } @inproceedings{LCNL:18, author={Marc F. Joanisse and Mark S. Seidenberg}, year={1999}, title={Impairments in verb morphology after brain injury: A connectionist model.}, booktitle={Proceedings of the National Academy of Sciences of the United States of America}, pages={7592-7597}, comments={A small model of past tense processing in which the production of the correct past tense is a constraint satisfaction problem utilizing a conjunction of phonological and semantic information.}, abstract={The formation of the past tense of verbs in English has been the focus of the debate concerning connectionist vs. symbolic accounts of language. Brain-injured patients differ with respect to whether they are more impaired in generating irregular past tenses (TAKE–TOOK) or past tenses for nonce verbs (WUG–WUGGED). Such dissociations have been taken as evidence for distinct "rule" and "associative" memory systems in morphology and against the connectionist approach in which a single system is used for all forms. We describe a simulation model in which these impairments arise from damage to phonological or semantic information, which have different effects on generalization and irregular forms, respectively. The results provide an account of the bases of impairments in verb morphology and show that these impairments can be explained within connectionist models that do not use rules or a separate mechanism for exceptions.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/18.pdf}, } @article{LCNL:19, author={Keith Rayner and Barbara R. Foorman and Charles A. Perfetti and David Pesetsky and Mark S. Seidenberg}, year={2001}, title={How Psychological Science Informs the Teaching of Reading. }, journal={Psychological Science in the Public Interest Monograph}, volume={2}, number={2}, pages={31-74}, comments={American Psychological Society. A review of psychological research bearing on how reading should be taught; illustrates the deeply unfortunate disconnect between reading research and educational practice that has occurred over the past 25 years.}, abstract={This monograph discusses research, theory, and practice relevant to how children learn to read English. After an initial overview of writing systems, the discussion summarizes research from developmental psychology on children's language competency when they enter school and on the nature of early reading development. Subsequent sections review theories of learning to read, the characteristics of children who do not learn to read (i.e., who have developmental dyslexia), research from cognitive psychology and cognitive neuroscience on skilled reading, and connectionist models of learning to read. The implications of the research findings for learning to read and teaching reading are discussed. Next, the primary methods used to teach reading (phonics and whole language) are summarized. The final section reviews laboratory and classroom studies on teaching reading. From these different sources of evidence, two inescapable conclusion emerge: (a) Mastering the alphabetic principle (that written symbols are associated with phonemes) is essential to becoming proficient in the skill of reading, and (b) methods that teach this principle directly are more effective than those that so not (especially for children who are at risk in some way for having difficulty learning to read). Using whole-language activities to supplement phonics instruction does help make reading fun and meaningful for children, but ultimately, phonics instruction is critically important because it helps beginning readers understand the alphabetic principle and learn to read. Thus, elementary-school teachers who make the alphabetic principle explicit are most effective in helping their students become skilled, independent readers.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/19.pdf}, } @inproceedings{LCNL:20, author={Marc F. Joanisse and Mark S. Seidenberg}, year={1998}, title={Functional bases of phonological universals: A connectionist approach.}, location={Berkeley, CA.}, booktitle={Proceedings of the 18th Annual Meeting of the Berkeley Linguistics Society}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/20.pdf}, } @article{LCNL:21, author={Marc F. Joanisse and Mark S. Seidenberg}, year={1998}, title={Specific language impairment: a deficit in grammar or processing?}, journal={Trends in Cognitive Sciences}, volume={2}, number={7}, pages={240-247}, comments={This review article discusses two competing hypotheses concerning developmental language impairments (SLI): one holds that SLI is caused by impairments in components of UG; the other that SLI is secondary to information processing deficits which affect the acquisition of several aspects of language. In this article we discuss how a phonological deficit could account for the impairments in both inflectional morphology and syntax (e.g., the resolution of pronouns) that are typically observed in such children. This account also correctly predicts that children with SLI will tend to be phonological dyslexics. The Joanisse and Seidenberg (2003) Brain and Language article presents some relevant modeling results and Joanisse et al. (2000) provide relevant behavioral data.}, abstract={Specific Language Impairment (SLI) is observed in children who fail to acquire ageappropriate language skills but otherwise appear to be developing normally. There are two main hypotheses about the nature of these impairments. One assumes that they reflect impairments in the child’s innate knowledge of grammar. The other is that they derive from information-processing deficits that interfere with several aspects of language learning. There is considerable evidence that SLI is associated with impaired speech processing; however, the link between this deficit and the kinds of grammatical impairments observed in these children has been unclear. We suggest that the link is provided by phonology, a speech-based code that plays important roles in learning linguistic generalizations and in working memory.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/21.pdf}, } @article{LCNL:22, author={James L. McClelland and Mark S. Seidenberg}, year={2000}, month={01}, title={Why do kids say goed and brang? Review of S. Pinker, Words and Rules.}, journal={Science}, volume={287}, number={5450}, pages={47-48}, comments={Because their little networks tell them so.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/22.htm}, } @article{LCNL:23, author={Debra Jared and Mark S. Seidenberg}, year={1991}, title={Does word identification proceed from spelling to sound to meaning?}, journal={Journal of Experimental Psychology: General}, volume={120}, number={4}, pages={358-394}, abstract={Six experiments addressed the role of phonological information in visual word recognition using a semantic-decision task. Exp 1 replicated Van Orden's (1987) finding that Ss make more false-positive errors on homophone foils than on spelling controls, indicating phonological activation of meaning. Exp 2 showed that only lower frequency words yield this effect when broader categories are used. In Exps 3 and 4, the homophony effect for lower frequency words remained, even though the stimuli included a large proportion of homophones, suggesting that activation of phonological information cannot be strategically inhibited. Exps 5 and 6 examined effects of homophony on targets that were correct category exemplars and yielded similar results. These studies indicate that in skilled readers, phonological information contributes to the activation of word meaning only for low-frequency words.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/23.pdf}, } @article{LCNL:24, author={Catherine McBride-Chang and Franklin R. Manis and Mark S. Seidenberg and Rebecca G. Custodio and Lisa M. Doi}, year={1993}, title={Print exposure as a predictor of word reading and reading comprehension in reading disabled and normally achieving children.}, journal={Journal of Educational Psychology}, volume={85}, number={2}, pages={230-238}, abstract={The relation of print exposure, measured by a revised version of A. E. Cunningham and K. E. Stanovich's (1990) Title Recognition Test (TRT), to word reading and reading comprehension was examined in disabled and nondisabled readers, Grades 5-9. In disabled readers, the TRT was a significant predictor of word reading when phonological skill was accounted for but not when orthographic ability was added to the regression equation, suggesting that the TRT overlaps considerably with orthographic skill. The TRT significantly predicted nondisabled readers' word reading after both phonological and orthographic skills were accounted for. The TRT contributed significantly to reading comprehension once variance was partialed from higher order reading processes for disabled readers only. The TRT's power to predict comprehension may be ascribed to the effects of print exposure on automaticity of word recognition, knowledge, or familiarization with text structure.}, language={English}, keywords={print exposure as measured by revised Title Recognition Test & phonological & orthographic skills, word reading & reading comprehension, 5th-9th grade disabled & nondisabled readers; Reading Comprehension; Reading Disabilities; Reading Measures; Word Recognition; Elementary School Students; High School Students; Junior High School Students; Orthography; Phonology}, issn={00220663}, URL={http://lcnl.wisc.edu/publications/archive/24.pdf}, } @article{LCNL:25, author={Mark S. Seidenberg and David C. Plaut and Alan S. Petersen and James L. McClelland and Ken McRae}, year={1994}, month={12}, title={Nonword pronunciation and models of word recognition.}, journal={Journal of Experimental Psychology: Human Perception and Performance,}, volume={20}, number={6}, pages={1177-1196}, abstract={Nonword pronunciation is a form of generalization behavior that has been at the center of debates about models of word recognition, the role of rules in explaining behavior, and the adequacy of the parallel distributed processing approach. An experiment yielded data concerning the pronunciation of a large corpus of nonwords. The data were then used to assess 2 models of naming: a model developed by D. C. Plaut and J. L. McClelland (1993), which is similar to the one described by M. S. Seidenberg and J. L. McClelland (1989) but uses improved orthographic and phonological representations, and the grapheme-phoneme correspondence rules of M. Coltheart, B. Curtis, P. Atkins, and M. Haller's (1993) dual-route model. Both models generate plausible nonword pronunciations and match subjects' responses accurately. The dual-route model does so by using rules that generate correct output for most words but mispronounce a significant number of exceptions. The parallel distributed processing model does so by finding a set of weights that allow it to generate correct output for both "rule-governed" items and exceptions. Some ways in which the two approaches differ and other issues facing them are also discussed.}, language={English}, keywords={nonword pronunciation & models of word recognition, college students; Models; Pronunciation; Word Recognition; Phonology}, issn={00961523}, URL={http://lcnl.wisc.edu/publications/archive/25.pdf}, } @article{LCNL:26, author={Marc F. Joanisse and Franklin R. Manis and Patricia Keating and Mark S. Seidenberg}, year={2000}, title={Language Deficits in Dyslexic Children: Speech Perception, Phonology and Morphology.}, journal={Journal of Experimental Child Psychology}, volume={77}, pages={30-60}, comments={This is the preprint version of the article that appeared in the journal. It provides evidence for a speech perception deficit in children whose dyslexia is secondary to a more general language impairment ("SLI") but not in dyslexics whose spoken language is apparently normal. Thus the "phonological" deficit apparent in many dyslexics may arise from something other than a sensory deficit.}, abstract={We investigated the relationship between dyslexia and three aspects of language: speech perception, phonology and morphology. Reading and language tasks were administered to dyslexics aged 8-9 years, and two normal reader groups (age-matched and reading-level matched). Three dyslexic groups were identified: phonological dyslexics (PD), developmentally language impaired (LI), and globally delayed (delay-type dyslexics). The LI and PD groups exhibited similar patterns of reading impairment, attributed to low phonological skills. However, only the LI group showed clear speech perception deficits, suggesting that such deficits affect only a subset of dyslexics. Results also indicated phonological impairments in children whose speech perception was normal. Both the LI and PD groups showed inflectional morphology difficulties, with the impairment being more severe in the LI group. The Delay group’s reading and language skills closely matched those of younger normal readers, suggesting these children had a general delay in reading and language skills, rather than a specific phonological impairment. The results are discussed in terms of models of word recognition and dyslexia.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/26.pdf}, } @article{LCNL:27, author={Mark S. Seidenberg and James L. McClelland}, year={1990}, month={07}, title={More words but still no lexicon: Reply to Besner et al. (1990).}, journal={Psychological Review}, volume={97}, number={3}, pages={447-452}, abstract={The major points in the Besner, Twilley, McCann, and Seergobin (1990) critique of the Seidenberg and McClelland (1989) model are addressed. The model's performance differs from that of people in ways that are predictable from an understanding of the limitations of implementation. The principal limitations are the size of the training corpus and the phonological representation. The issue of pseudohomophone effects is discussed, and Besner et al's new data are shown to be consistent with the Seidenberg and McClelland account of lexical decision.}, language={English}, keywords={model of visual word recognition & naming, language development & reading & lexical decision, commentary reply; Language Development; Lexical Decision; Naming; Reading; Word Recognition; Models; Visual Perception}, issn={0033295X}, URL={http://lcnl.wisc.edu/publications/archive/27.pdf}, } @article{LCNL:28, author={Mark S. Seidenberg and Jeffrey L. Elman}, year={1999}, month={08}, title={Networks are not "hidden rules."}, journal={Trends in Cognitive Science}, volume={3}, number={8}, pages={288-289}, abstract={Comments on an article by G. F. Marcus et al (see record 1999-00199-002). In this letter we would like to raise a more general issue about the relationship between connectionist models and algebraic rules. The purpose of our simulation was to demonstrate that the sequential regularities implicit in the author's stimuli provided a sufficient basis for differentiating sequences that conformed to the 'algebraic rule' from ones that did not, and that having acquired this information, a network could generalize appropriately to novel stimuli. The authors objected to two aspects of this simulation: instead of the prediction task used in some other models, our model was trained to categorize stimuli as fitting a pattern or not doing so. We also used a supervised-learning procedure in which the network was provided with explicit feedback. In a sense, then, feedback to the model was structured on the basis of a rule; hence, the authors concluded that the model must have had the rule 'hidden' in it. However, his statement is a non sequitur. Merely training a network to categorize stimuli into two groups using explicit feedback does not cause it to formulate a rule. (PsycINFO Database Record (c) 2006 APA, all rights reserved)}, language={English}, keywords={abstract algebraic rules; language learning; infant development; Cognitive Development; Cognitive Hypothesis Testing; Generalization (Learning); Language Development}, issn={13646613}, URL={http://lcnl.wisc.edu/publications/archive/28.pdf}, } @article{LCNL:114, author={Daniel J. Acheson and Maryellen C. MacDonald}, year={2009}, title={Verbal Working Memory and Language Production: Common Approaches to the Serial Ordering of Verbal Information.}, journal={Psychological Bulletin}, volume={135}, number={1}, pages={50-68}, URL={http://lcnl.wisc.edu/publications/archive/114.pdf}, } @article{LCNL:115, author={Justine B. Wells and Morten H. Christiansen and David S. Race and Daniel J. Acheson and Maryellen C. MacDonald}, year={2009}, title={Experience and sentence comprehension: Statistical learning and relative clause comprehension}, journal={Cognitive Psychology}, volume={58}, pages={250-271}, URL={http://lcnl.wisc.edu/publications/archive/115.pdf}, } @article{LCNL:116, author={James Keidel and Keith Kluender and Rick Jenison and Mark Seidenberg}, year={2007}, title={Does grammar constrain statistical learning?}, journal={Psychological Science}, volume={18}, number={10}, pages={922-923}, comments={Bonatti et al. (Psych Sci, 2007) showed that adult learners picked up on statistical properites of consonants but not vowels in an artificial language learning study. They claimed that this finding could not be explained by a general statistical learning mechanism. We showed that the asymmetry reflects properties of French, the native language of the subjects. (Their response was to abandon the original claim--that the effects were not due to any statistical property of French--in favor of the idea that UG is responsible for the observed statistical asymmetry.)}, abstract={We're right, they're wrong (there is no abstract because this is a commentary paper)}, URL={http://lcnl.wisc.edu/publications/archive/116.pdf}, } @article{LCNL:117, author={Jennifer Bruno and Franklin Manis and Patricia Keating and Anne Sperling and Jonathan Nakamoto and Mark S. Seidenberg}, year={2007}, title={Auditory word identification in dyslexic and normally achieving readers}, journal={Journal of Experimental Child Psychology}, volume={97}, pages={125-154}, URL={http://lcnl.wisc.edu/publications/archive/117.pdf}, } @article{LCNL:30, author={Michael W. Harm and Mark S. Seidenberg}, year={2001}, title={Are There Orthographic Impairments In Phonological Dyslexia?}, journal={Cognitive Neuropsychology}, volume={18}, pages={71-92}, comments={Modeling and behavioral research that contradict the sometime claim that acquired phonological dyslexia also involves an orthographic deficit. Our models show how the critical data could arise with intact orthographic processing and the behavioral data show the same effects in normals.}, abstract={Two hypotheses have been advanced concerning the basis of acquired phonological dyslexia. According to the dual-route model, the pattern derives from impaired grapheme-phoneme conversion. According to the phonological impairment hypothesis, it derives from impaired representation and use of phonology. Effects of graphemic complexity and visual similarity observed in studies by D. Howard and W. Best (1996), orthographic effects on phoneme counting (R. S. Berndt et al, 1996), and data from patient LB (J. Derouesné and M. F. Beauvois, 1985) have been taken as evidence for an orthographic impairment in phonological dyslexia and therefore against the impaired phonology hypothesis (M. Coltheart, 1996). The authors present a computational simulation, results of 2 behavioral studies (with students as participants), and a critical-analysis of the MJ and LB data, which suggest that the "orthographic" deficits in such patients arise from phonological impairments that interact with orthographic properties of stimuli.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/30.pdf}, } @article{LCNL:31, author={Mark S. Seidenberg and James L. McClelland}, year={1989}, title={A distributed developmental model of word recognition and naming.}, journal={Psychological Review}, volume={96}, number={4}, pages={523-568}, abstract={A parallel distributed processing model of visual word recognition and pronunciation is described. The model consists of sets of orthographic and phonological units and an interlevel of hidden units. Weights on connections between units were modified during a training phase using the back-propagation learning algorithm. The model simulates many aspects of human performance, including (a) differences between words in terms of processing difficulty, (b)pronunciation of novel items, (c) differences between readers in terms of word recognition skill, (d) transitions from beginning to skilled reading, and (e) differences in performance on lexical decision and naming tasks. The model's behavior early in th learning phase corresponds to that of children acquiring word recognition skills. Training with a smaller number of hidden units produces output characteristic of many dyslexic readers. naming is simulated without pronunciation rules, and lexical decisions are simulated without accessing word-level representations. The performance of the model is largely determined by three factors: the nature of the input, a significant fragment of written English; the learning rule, which encodes the implicit structure of the orthography in the weights on connections; and the architecture of the system, which influences the scope of what can be learned.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/31.pdf}, } @article{LCNL:32, author={Keith Rayner and Barbara R. Foorman and Charles A. Perfetti and David Pesetsky and Mark S. Seidenberg}, year={2002}, month={03}, title={How should reading be taught?}, journal={Scientific American}, volume={286}, number={3}, pages={84-91}, comments={An accessible discussion of the controversies over how reading should be taught.}, abstract={Educators have long argued over the best way to teach reading to children. The research, however, indicates that a highly popular method is inadequate on its own}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/32.pdf}, } @article{LCNL:33, author={Marc F. Joanisse and Mark S. Seidenberg}, year={2005}, title={Imaging the past: Neural activation in frontal and temporal regions during regular and irregular past tense processing.}, journal={Cognitive, Affective, and Behavioral Neurosciences.}, volume={5}, number={3}, pages={282-296}, comments={fMRI evidence bearing on theories of the past tense. "Irregular" past tenses such as SLEPT actually pattern with "regular" past tenses such as STEPPED. Hence the brain does not organize this knowledge in terms of rule-governed forms and exceptions.}, abstract={This article presents fMRI evidence bearing on dual-mechanism vs. connectionist theories of inflectional morphology. Ten participants were scanned at 4 Tesla as they covertly generated the past tenses of real and nonce (nonword) verbs presented auditorily. Regular past tenses (e.g, walked, wugged) and irregular past tenses (e.g., took, slept) produced similar patterns of activation in the posterior temporal lobe in both hemispheres. In contrast, there was greater activation for regular past tenses in left and right IFG compared to irregular past tenses. Similar previous results have been taken as evidence for the dual-mechanism theory of the past tense (Pinker & Ullman, 2002). However, additional analyses indicated that irregulars that are phonologically similar to regulars (e.g., slept, fled, sold) produced the same level of activation as regular forms, and significantly more than irregulars that are not phonologically similar to regulars (e.g., took, gave). Thus, activation patterns were predicted by phonological characteristics of the past tense rather than the rule-governed vs. exception distinction that is central to the words-and-rules framework. The results are consistent with a constraint satisfaction model in which phonological, semantic and other probabilistic constraints jointly determine the past tense, with different degrees of involvement for different verbs.}, language={English}, keywords={neural activation; frontal regions; temporal regions; functional MRI; dual mechanism; morphology; inferior frontal gyrus; Frontal Lobe; Magnetic Resonance Imaging; Neural Plasticity; Temporal Lobe; Cerebral Cortex; Gyrus Cinguli}, issn={15307026}, URL={http://lcnl.wisc.edu/publications/archive/33.pdf}, } @article{LCNL:34, author={Mark S. Seidenberg}, year={2005}, title={Connectionist models of word reading.}, journal={Current Directions in Psychological Science.}, volume={14}, number={5}, pages={238-242}, comments={A brief nontechnical overview of our reading models. Includes "The First Law of Modeling".}, abstract={Connectionist models of word reading attempt to explain the computational mechanisms underlying this important skill. The goal is an integrated theory of reading and its brain bases, with the computational model as the interface between the two. The models are governed by computational principles that differ considerably from naïve intuitions but nonetheless account for many aspects of normal and impaired (dyslexic) reading.}, language={English}, keywords={reading; connectionist models; dyslexia}, URL={http://lcnl.wisc.edu/publications/archive/34.pdf}, } @article{LCNL:35, author={Eamon Strain and Karalyn Patterson and Mark S. Seidenberg}, year={1995}, title={Semantic effects in single-word naming.}, journal={Journal of Experimental Psychology: Learning, Memory, and Cognition}, volume={21}, number={5}, pages={1140-1154}, abstract={Three experiments demonstrated that, for lower frequency words, reading aloud is affected not only by spelling-sound typicality but also by a semantic variable, imageability. Participants were slower and more error prone when naming exception words with abstract meanings (e.g., scarce) than when naming either abstract regular words (e.g., scribe) or imageable exception words (e.g., soot). It is proposed that semantic representations of words have the largest impact on translating orthography to phonology when this translation process is slow or noisy (i.e., for low-frequency exceptions) and that words with rich semantic representations (i.e., high-imageability words) are most likely to benefit from this interaction.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/35.pdf}, } @article{LCNL:36, author={Debra Jared and Mark S. Seidenberg}, year={1990}, title={Naming multisyllabic words.}, journal={Journal of Experimental Psychology: Human Perception and Performance}, volume={16}, number={1}, pages={92-105}, abstract={The process of reading multisyllabic words aloud from print was examined in 4 experiments. Experiment 1 used multisyllabic words that vary in terms of the consistency of component spelling-sound correspondences. The stimuli were regular, regular inconsistent, and exception words analogous to the monosyllabic items used in previous studies. Both regular inconsistent and exception words produced longer naming latencies than regular words. In Experiment 2 these differences between word types were found to be limited to lower frequency items. Experiment 3 showed that effects of number of syllables on naming latency are also limited to lower frequency words. In the final experiment, consistency effects were obtained for both higher and lower frequency words when the stimulus display forced subjects to use syllabic units. Thus, frequency modulates the effects of two aspects of lexical structure--consistency of spelling-sound correspondences and number of syllables. The results suggest that the naming of multisyllabic words draws on some of the same knowledge representations and processes as monosyllabic words; however, naming does not require syllabic decomposition. The results are discussed in the context of current models of naming.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/36.pdf}, } @article{LCNL:37, author={Michael W. Harm and Mark S. Seidenberg}, year={1999}, title={Reading acquisition, phonology, and dyslexia: Insights from a connectionist model.}, journal={Psychological Review}, volume={106}, pages={491-528}, comments={A spelling-sound model applied to issues concerning learning to read and dyslexia. Some cool simulations of the effects of phonological impairment on reading and illustrations of attractor dynamics.}, abstract={The development of reading skill and bases of developmental dyslexia were explored using connectionist models. Four issues were examined: the acquisition of phonological knowledge prior to reading, how this knowledge facilitates learning to read, phonological and non phonological bases of dyslexia, and effects of literacy on phonological representation. Compared with simple feedforward networks, representing phonological knowledge in an attractor network yielded improved learning and generalization. Phonological and surface forms of developmental dyslexia, which are usually attributed to impairments in distinct lexical and nonlexical processing “routes,” were derived from different types of damage to the network. The results provide a computationally explicit account of many aspects of reading acquisition using connectionist principles.}, language={English}, keywords={connectionist model of phonology & reading acquisition & dyslexia; Connectionism; Dyslexia; Models; Phonology; Reading Development}, issn={0033295X}, URL={http://lcnl.wisc.edu/publications/archive/37.pdf}, } @article{LCNL:38, author={Marc F. Joanisse and Mark S. Seidenberg}, year={2003}, title={Phonology and syntax in specific language impairment: Evidence from a connectionist model.}, journal={Brain and Language}, volume={86}, pages={40-56}, comments={This article presents computational modeling evidence linking a phonological deficit to impaired processing of an aspect of grammar, pronominal anaphora. The results provide a causal demonstration of how a perceptual deficit that affects phonological representation could have secondary effects on grammatical processing. This is consistent with the conclusion that "grammatical" deficits in SLI do not reflect anomalous development within grammatical modules but rather are sequelae of more basic information processing deficits.}, abstract={Difficulties in resolving pronominal anaphora have been taken as evidence that Specific Language Impairment (SLI) involves a grammar-specific impairment. The present study explores an alternative view, that grammatical deficits in SLI are sequelae of impaired speech perception. This perceptual deficit specifically affects the use of phonological information in working memory, which in turn leads to poorer than expected syntactic comprehension. This hypothesis was explored using a connectionist model of sentence processing that learned to map sequences of words to their meanings. Anaphoric resolution was represented in this model by recognizing the semantics of the correct antecedent when a bound pronoun was input. When the model was trained on distorted phonological inputs—simulating a perceptual deficit—it exhibited marked difficulty resolving bound anaphors. However, many other aspects of sentence comprehension were intact; most importantly, the model could still resolve pronouns using gender information. In addition, the models deficit was graded rather than categorical, as it was able to resolve pronouns in some sentences, but not in others. These results are consistent with behavioral data concerning syntactic deficits in SLI. The model provides a causal demonstration of how a perceptual deficit could give rise to grammatical deficits in SLI.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/38.pdf}, } @article{LCNL:39, author={Anne J. Sperling and Zhong-Lin Lu and Franklin R. Manis and Mark S. Seidenberg}, year={2005}, month={07}, title={Deficits in perceptual noise exclusion in developmental dyslexia.}, journal={Nature Neuroscience.}, volume={8}, number={7}, pages={862-863}, comments={A brief paper reporting an elegant experiment that was part of Anne Sperling's USC thesis. Strong evidence against the magnocellular deficit account of dyslexia; strong evidence that dyslexics are impaired in suppressing perceptual noise, in both magno and parvo channels.}, abstract={Connectionist models have made significant contributions to understanding developmental phenomena, mainly by providing novel computational accounts of behavioral emergence and change. What is the fate of such models given the increasing interest in and information about the biological bases of development? We consider this issue with respect to the classical idea of a critical period for acquiring language. The standard view is that neurobiological developments on a strict maturational timetable create limits on language learning capacity. Computational analysis suggests the opposite: that learning itself creates neurobiological conditions underlying the “closing” of the critical period. The critical period example suggests how connectionist models can continue to provide a necessary level of analysis intermediate between behavior and brain.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/39.pdf}, } @article{LCNL:40, author={Mark S. Seidenberg and Jeffrey L. Elman}, year={1999}, month={04}, title={Do infants learn grammar with algebra or statistics?}, journal={Science}, volume={284}, pages={433}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/40.htm}, } @article{LCNL:41, author={Caroline E. Bailey and Franklin R. Manis and William C. Pedersen and Mark S. Seidenberg}, year={2004}, title={Variation among developmental dyslexics: Evidence from a printed-word-learning task.}, journal={Journal of Experimental Child Psychology}, volume={87}, pages={125-154}, abstract={Complex words consist of morphemic subunits that can recombine to form other words. Thus midnight is standardly analyzed as consisting of the prefix mid- and stem night, which also occur in words such as midstream and nightly. A considerable body of empirical and theoretical research suggests that morphological structure governs the representation of words in memory and that many words are decomposed into morphological components in processing. We investigated an alternative approach in which morphology arises from the interaction of semantic, phonological, and orthographic codes. Five cross-modal lexical decision experiments show that the magnitude of priming (e.g., for pairs such as teacher-teach) is affected by the degree of semantic and phonological overlap between words. Crucially, items that are only moderately similar produce intermediate facilitation effects (e.g., latelylate). This pattern is observed both for words standardly treated as morphologically related (e.g., teacher-teach) and for morphologically unrelated words that exhibit similar degrees of semantic and phonological overlap (e.g., snarl-sneer). The results can be understood in terms of connectionist models employing distributed representations rather than discrete morphemes.}, language={English}, keywords={Reading; Dyslexia; Subgroups; Phonological skill; Word identification; Models of reading}, URL={http://lcnl.wisc.edu/publications/archive/41.pdf}, } @article{LCNL:42, author={Morten H. Christiansen and Joseph Allen and Mark S. Seidenberg}, year={1998}, month={04}, title={Learning to segment speech using multiple cues: A connectionist model.}, journal={Language and Cognitive Processes}, volume={13}, pages={221-268}, comments={This article describes an initial attempt to model the word segmentation problem using a conjunction of probabilistic cues. The model is a demonstration of an approach more than a realistic solution to the problem: to keep the simulations manageable, it was given capacities that infants lack (e.g., knowledge of segmental phonology). Also, the trained model correctly identified only about 45% of the words in the speech stream--but then again how accurate is your average 4 month old?}, abstract={Considerable research in language acquisition has addressed the extent to which basic aspects of linguistic structure might be identified on the basis of probabilistic cues in caregiver speech to children. This type of learning mechanism presents classic learnability issues: There are aspects of language for which the input is thought to provide no evidence, and the evidence that does exist tends to be unreliable. These issues are addressed in the context of the specific problem of learning to identify lexical units in speech. A simple recurrent network was trained on a phoneme prediction task. The model was explicitly provided with information about phonemes, relative lexical stress, and boundaries between utterances. Individually these sources of information provide relatively unreliable cues to word boundaries and no direct evidence about actual word boundaries. After training on a large corpus of child-directed speech, the model was able to use these cues to reliably identify word boundaries. The model shows that aspects of linguistic structure not overtly marked in the input can be derived by efficiently combining multiple probabilistic cues. Connectionist networks provide a plausible mechanism for acquiring, representing, and combining such probabilistic information.}, language={English}, keywords={connectionist model of speech segmentation learning; Connectionism; Oral Communication; Verbal Learning; Models}, issn={01690965}, URL={http://lcnl.wisc.edu/publications/archive/42.pdf}, } @article{LCNL:43, author={Michael W. Harm and Mark S. Seidenberg}, year={2004}, title={Computing the Meanings of Words in Reading: Cooperative Division of Labor Between Visual and Phonological Processes}, journal={Psychological Review}, volume={111}, number={3}, pages={662-720}, comments={A major extension of the "triangle" model to the computation of meaning; addresses controversy about role of direct vs. phonologically-mediated mechanisms in reading.}, abstract={Are words read visually (by means of a direct mapping from orthography to semantics) or phonologically (by mapping from orthography to phonology to semantics)? We addressed this longstanding debate by examining how a large-scale computational model based on connectionist principles would solve the problem and comparing the model's performance to people's. In contrast to previous models, the present model employs an architecture in which meanings are jointly determined by the two components, with the division of labor between them affected by the nature of the mappings between codes. The model is consistent with a variety of behavioral phenomena, including the results of studies of homophones and pseudohomophones thought to support other theories, and illustrates how efficient processing can be achieved using multiple simultaneous constraints.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/43.pdf}, } @article{LCNL:44, author={Helen Bird and Matthew A. Lambon Ralph and Mark S. Seidenberg and James L. McClelland and Karalyn Patterson}, year={2003}, month={04}, title={Deficits in phonology and past-tense morphology: What's the connection?}, journal={Journal of Memory and Language}, volume={48}, pages={502-526}, comments={Evidence that impairments in generating regular past tenses such as WAITED derive from phonological deficits, not an impairment in the "rule" module.}, abstract={Neuropsychological dissociations between regular and irregular past tense verb processing have been explained in two ways: (a) separate mechanisms of a rule-governed process for regular verbs and a lexical-associative process for irregular verbs; (b) a single system drawing on phonological and semantic knowledge. The latter account invokes phonological impairment as the basis of poorer performance for regular than irregular past tense forms, due to greater phonological complexity of the regular past. In 10 nonfluent aphasic patients (aged 46-83 yrs), the apparent disadvantage for the production of regular past tense forms disappeared when phonological complexity was controlled. In a same-different judgment task on spoken words, all patients were impaired at judging regular stem and past-tense verbs like man/manned to be different, but equally poor at phonologically matched non-morphological discriminations like men/mend. These results indicate a central phonological deficit that is not limited to speech output nor to morphological processing; under such a deficit, distinctions lacking phonological salience, as typified by regular past tense English verbs, become especially vulnerable.}, language={English}, keywords={neuropsychological dissociation; phonological impairment; phonological complexity; regular & irregular past tense verb processing; Morphology (Language); Neurolinguistics; Phonology; Verbal Fluency; Verbs; Aphasia}, issn={0749596X}, URL={http://lcnl.wisc.edu/publications/archive/44.pdf}, } @article{LCNL:45, author={Franklin R. Manis and Mark S. Seidenberg and Lisa M. Doi and Catherine McBride-Chang and Alan Peterson}, year={1996}, title={On the basis of two subtypes of developmental dyslexia.}, journal={Cognition}, volume={58}, pages={157-195}, abstract={This study examined whether there are different subtypes of developmental dyslexia. The subjects were 51 dyslexic children (reading below the 30th percentile in isolated word recognition), 51 age-matched normal readers, and 27 younger normal readers who scored in the same range as the dyslexics on word recognition. Using methods developed by Castles and Coltheart (1993), we identified two subgroups who fit the profiles commonly termed "surface" and "phonological" dyslexia. Surface subjects were relatively poorer in reading exception words compared to nonwords; phonological dyslexics showed the opposite pattern. However, most dyslexics were impaired on reading both exception words and nonwords compared to same-aged normal readers. Whereas the surface dyslexics' performance was very similar to that of younger normal readers, the phonological dyslexics' was not. The two dyslexic groups also exhibited a double dissociation on two validation tasks: surface subjects were impaired on a task involving orthographic knowledge but not one involving phonology; phonological dyslexics showed the opposite pattern. The data support the conclusion that there at least two distinct subtypes of developmental dyslexia. Although these patterns have been taken as evidence for the dual-route model, we provide an alternative account of them within the Seidenberg and McClelland (1989) connectionist model. The connectionist model accounts for why dyslexics tend to be impaired on both exception words and nonwords; it also suggests that the subtypes may arise from multiple underlying deficits. Performance on exception words and nonwords is not sufficient to identify the underlying basis of dyslexic behavior; rather, information about children's performance on other tasks and their remediation experiences must be taken into account as well.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/45.pdf}, } @article{LCNL:46, author={Michael W. Harm and Bruce D. McCandliss and Mark S. Seidenberg}, year={2003}, title={Modeling the successes and failures of interventions for disabled readers.}, journal={Scientific Studies of Reading}, volume={7}, pages={155-182}, comments={Using a computational model of reading to understand why some types of reading remediation work and others don't.}, abstract={A connectionist model of reading development previously used to simulate detailed aspects of developmental dyslexia (Harm and Seidenberg, 1999) was used to explore why certain classes of interventions designed to overcome reading impairments are more effective than others. Previous research has shown that interventions targeting the development of spelling-sound correspondences are more effective at promoting generalization skills than ones designed solely to increase phonological awareness. The simulations broadly replicate the patterns of success and failure found in the developmental literature and provide explicit computational insights into exactly why the interventions that include training on spelling-sound regularities are more effective than those targeting phonological development alone.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/46.pdf}, } @article{LCNL:47, author={Jason D. Zevin and Mark S. Seidenberg}, year={2004}, title={Age of acquisition effects in reading aloud: Tests of cumulative frequency and frequency trajectory.}, journal={Memory & Cognition}, volume={32}, number={1}, pages={31-38}, abstract={Several studies have reported that the age at which a word is learned affects skilled reading. This age-of-acquisition effect is potentially important for theories of reading and learning. The effect has been difficult to pin down, however, because the age at which a word is learned is correlated with many other lexical properties. Zevin and Seidenberg (2002) analyzed these phenomena, using connectionist models that distinguished between cumulative frequency (the total number of times a word is experienced) and frequency trajectory (the distribution of these experiences over time). The models prompted a reevaluation of the empirical literature on this topic. The present research tested and confirmed three behavioral predictions derived from these models. First, cumulative frequency has an impact on skilled word naming, more so than standard measures of frequency derived from such norms as those of Kucera and Francis (1967). Second, frequency trajectory affects age of acquisition: The timing of exposure to words affects how rapidly they are learned. However, frequency trajectory does not affect skilled reading aloud, because the consistencies in mapping between spelling and sound eventually wash out the effects of early differences in frequency of exposure. Thus, in skilled performance, the timing of exposure to words is less important than the amount of exposure. The results clarify the conditions under which age-dependent learning effects occur in reading aloud.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/47.pdf}, } @article{LCNL:48, author={Mark S. Seidenberg and Aimee Arnoldussen}, year={2003}, month={06}, title={The brain makes a distinction between hard and easy stimuli: Comments on Baretta et al.}, journal={Brain and Language}, volume={85}, number={3}, pages={527-530}, comments={Can neuroimaging provide decisive evidence concerning competing theories of past tense morphology? And what about German, anyway?}, abstract={The Beretta et al. study tested an invalid prediction concerning connectionist models of inflectional morphology and the study exhibits a confound between type of stimulus (regular, irregular) and processing difficulty (easy, hard) that invalidates their conclusions. Harder stimuli produced greater activation across a broader range of brain areas, as in previous studies, but the data have no bearing on the rules vs. connections debate.}, language={English}, keywords={event-related potentials; morphological inflection; German language; regular vs irregular nouns & verb inflections; Brain; Evoked Potentials; Inflection; Morphology (Language); Nouns; Verbs}, issn={0093934X}, URL={http://lcnl.wisc.edu/publications/archive/48.pdf}, } @article{LCNL:49, author={Anne J. Sperling and Zhong-Lin Lu and Franklin R. Manis and Mark S. Seidenberg}, year={2003}, title={Selective magnocellular deficits in dyslexia: a "phantom contour" study.}, journal={Neuropsychologia}, volume={41}, pages={1422-1429}, abstract={A technique by Rogers-Ramachandran and Ramachandran [Vis. Res. 38 (1998) 71-77] was adapted to evaluate magnocellular (M) and parvocellular (P) visual processing efficiency, with identical task structure, in 19 normal and 19 dyslexic children. A battery of phonological, orthographic and cognitive tasks was administered to assess reading ability and component reading skills in both groups. For the visual processing experiment, children identified shapes created by patterns of dots flickering in counter-phase. The dots were black and white in the M condition, versus isoluminant red and green in the P condition. A staircase procedure determined the children's threshold flicker rate for shape identification. Dyslexics displayed selectively slower visual processing in the M condition but not in the P condition. Across all subjects, performance in the M condition was correlated with measures of orthographic skill, consistent with previous findings linking M processing and orthographic skill. Within the dyslexic group, processing in the M condition was negatively correlated with level of phonological awareness. The results are not consistent with the argument that dyslexics with phonological impairments suffer from deficits across all sensory modalities, as those children with the poorest phonological awareness displayed magnocellular processing well within the normal range.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/49.pdf}, } @inproceedings{LCNL:50, author={Marc F. Joanisse and Mark S. Seidenberg}, year={1997}, title={[i e a u] and Sometimes [o]: Perceptual and computational constraints on vowel inventories.}, location={Stanford, CA.}, booktitle={Proceedings of the 15th Annual Conference of the Cognitive Science Society.}, abstract={Common vowel inventories of languages tend to be better dispersed in the space of possible vowels than less common or unattested inventories. The present research explored the hypothesis that functional factors cause this preference. Connectionist models were trained on different inventories of spoken vowels, taken from a naturalistic corpus. The first experiment showed that networks trained on well-dispersed five-vowel sets like [i e a o u] learned the inventory more quickly and generalized better to novel stimuli, compared to those trained on less dispersed vowel sets. Experiments 2-3 examined how effects due to ease of perception are modulated by factors related to production. Languages tend to prefer front vowel contrasts over back vowels because the latter tend to be produced with more variability. This caused networks trained on an [i e a u] inventory to perform better than those trained on [i a o u]. Thus both acoustic separation of vowels and variability in how they are realized in speech affect ease of learning and generalization. The results suggest that acoustic and articulatory factors can explain apparent phonological universals.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/50.pdf}, } @article{LCNL:51, author={Mark S. Seidenberg}, year={1997}, month={03}, title={Language acquisition and use: Learning and applying probabilistic constraints.}, journal={Science}, volume={275}, pages={1599-1604}, abstract={What kinds of knowledge underlie the use of language and how is this knowledge acquired? Linguists equate knowing a language with knowing a grammar. Classic "poverty of the stimulus" arguments suggest that grammar identification is an intractable inductive problem and that acquisition is possible only because children possess innate knowledge of grammatical structure. An alternative view is emerging from studies of statistical and probabilistic aspects of language, connectionist models, and the learning capacities of infants. This approach emphasizes continuity between how language is acquired and how it is used. It retains the idea that innate capacities constrain language learning, but calls into question whether they include knowledge of grammatical structure.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/51.pdf}, } @article{LCNL:52, author={Mark S. Seidenberg and James H. Hoeffner}, year={1998}, title={Evaluating behavioral and neuroimaging evidence about past tense processing.}, journal={Language}, volume={74}, pages={104-122}, abstract={Jaeger, Lockwood, Kemmerer, Van Valin, Murphy, and Khalak (1996) reported an experimental study that provided reaction time and PET neuroimaging data said to support Pinker's (1991) theory of inflectional morphology in which rule-governed forms and exceptions are processed by separate mechanisms. The results were also taken as evidence against connectionist accounts in which a single processing system generates both types of forms. We provide a critical analysis of the study that yields three main conclusions: First, Jaeger et al.'s data do not provide strong evidence that rule-governed forms and exceptions are processed in separate brain regions. Second, there are problems with the design of the study that contaminate critical comparisons between conditions. The results therefore afford alternative interpretations related to experimentspecific factors rather than the regular-irregular distinction. Third, the dissociations between rulegoverned forms and exceptions observed in studies such as Jaeger et al.'s can be accommodated by the connectionist theory. We conclude by offering suggestions for future research that would overcome the major limitations of this study and provide more decisive evidence bearing on the issues.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/52.pdf}, } @article{LCNL:53, author={Eamon Strain and Karalyn Patterson and Mark S. Seidenberg}, year={2002}, month={01}, title={Theories of word naming interact with spelling-sound consistency.}, journal={Journal of Experimental Psychology: Learning, Memory, and Cognition}, volume={28}, number={1}, pages={207-214}, abstract={In a previous study (E. Strain, K. Patterson, & M. S. Seidenberg, 1995), the authors concluded that word naming is characterized by an interaction between spelling-sound typicality and word imageability, thus implicating a role for word meaning in the naming process. J. Monaghan and A. W. Ellis (2002) reject E. Strain et al.'s conclusion, arguing that it is age of acquisition (AoA) and not imageability that interacts with spelling-sound typicality. In this article, the authors question their alternative interpretation (a) by raising a number of conceptual and methodological issues germane to this debate and (b) by presenting new data that confirm a significant interaction between spelling-sound typicality and imageability in word-naming latencies, an interaction that is reliable when word AoA is controlled in a regression analysis.}, language={English}, keywords={theories; word naming; spelling-sound consistency; word meaning; spelling-sound typicality; word imageability; age of acquisition; Imagery; Naming; Spelling; Words (Phonetic Units)}, issn={02787393}, URL={http://lcnl.wisc.edu/publications/archive/53.pdf}, } @article{LCNL:54, author={Jason D. Zevin and Mark S. Seidenberg}, year={2002}, month={07}, title={Age of acquisition effects in reading and other tasks}, journal={Journal of Memory and Language}, volume={47}, number={1}, pages={1-29}, comments={Provides a new analysis of age of acquisition effects, in the context of a more general theory of how experience limits plasticity.}, abstract={Recent studies have suggested that age of acquisition (AoA) has an impact on skilled reading independent of factors such as frequency. This result raises questions about previous studies in which AoA was not controlled and about current theories in which it is not addressed. Analyses of the materials used in previous studies suggest that the observed AoA effects may have been due to other factors. We also found little evidence for an AoA effect in computational models of reading that used words that exhibit normal spelling–sound regularities. An AoA effect was observed, however, in a model in which early and late learned words did not overlap in terms of orthography or phonology. The results suggest that with other correlated properties of stimuli controled, AoA effects occur when what is learned about early patterns does not carry over to later ones. This condition is not characteristic of learning spelling–sound mappings but may be relevant to tasks such as learning the names for objects.}, language={English}, keywords={age of acquisition; skilled reading; word reading; Age Differences; Language Development; Reading; Reading Skills; Words (Phonetic Units)}, issn={0749596X}, URL={http://lcnl.wisc.edu/publications/archive/54.pdf}, } @article{LCNL:55, author={Mark S. Seidenberg}, year={1994}, title={Language and connectionism: The developing interface.}, journal={Cognition}, volume={50}, pages={385-401}, abstract={After a difficult initial period in which connectionism was perceived as either irrelevant or antithetical to linguistic theory, connectionist concepts are now beginning to be brought to bear on basic issues concerning the structure, acquisition, and processing of language, both normal and disordered / this article describes some potential points of further contact between connectionism and linguistic theory / consider how connectionist concepts may be relevant to issues concerning the representation of linguistic knowledge; the role of a priori constraints on acquisition; and the poverty of the stimulus argument / discuss whether these models contribute to the development of explanatory theories of language}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/55.pdf}, } @article{LCNL:56, author={David C. Plaut and James L. McClelland and Mark S. Seidenberg and Karalyn Patterson}, year={1996}, title={Understanding normal and impaired word reading: Computational principles in quasi-regular domains}, journal={Psychological Review}, volume={103}, pages={56-115}, abstract={A connectionist approach to processing in quasi-regular domains, as exemplified by English word reading, is developed. Networks using appropriately structured orthographic and phonological representations were trained to read both regular and exception words, and yet were also able to read pronounceable nonwords as well as skilled readers. A mathematical analysis of a simplified system clarifies the close relationship of word frequency and spelling-sound consistency in influencing naming latencies. These insights were verified in subsequent simulations, including an attractor network that accounted for latency data directly in its time to settle on a response. Further analyses of the ability of networks to reproduce data on acquired surface dyslexia support a view of the reading system that incorporates a graded division of labor between semantic and phonological processes, and contrasts in important ways with the standard dual-route account.}, language={English}, keywords={connectionist simulated modeling approach to processing in quasi-regular domains & naming latency & normal & impaired word reading; Connectionism; Naming; Reading; Words (Phonetic Units); Dyslexia; Simulation}, issn={0033295X}, URL={http://lcnl.wisc.edu/publications/archive/56.pdf}, } @article{LCNL:57, author={Ken McRae and Virginia R. de Sa and Mark S. Seidenberg}, year={1997}, month={06}, title={On the nature and scope of featural representations of word meaning.}, journal={Journal of Experimental Psychology: General}, volume={126}, number={2}, pages={99-130}, abstract={Behavioral experiments and a connectionist model were used to explore the use of featural representations in the computation of word meaning. The research focused on the role of correlations among features, and differences between speeded and untimed tasks with respect to the use of featural information. The results indicate that featural representations are used in the initial computation of word meaning (as in an attractor network), patterns of feature correlations differ between artifacts and living things, and the degree to which features are intercorrelated plays an important role in the organization of semantic memory. The studies also suggest that it may be possible to predict semantic priming effects from independently motivated featural theories of semantic relatedness. Implications for related behavioral phenomena such as the semantic impairments associated with Alzheimer's disease (AD) are discussed.}, language={English}, keywords={use of featural representations in computation of word meaning, college students, Canada, implications for semantic impairments in Alzheimer's disease; Cognitive Processes; Semantics; Word Meaning}, issn={00963445}, URL={http://lcnl.wisc.edu/publications/archive/57.pdf}, } @article{LCNL:58, author={Silvia P. Gennari}, year={2003}, title={Tense meanings and temporal interpretation.}, journal={Journal of Semantics}, volume={20}, pages={35-71}, abstract={For any theory of tense meanings, subordinate sentences are particularly problematic because embedded tenses do not seem to receive the same interpretations as their non-embedded counterparts. Previous approaches to this problem have often proposed some syntactic mechanism or sequence of tense rule that allows the embedded tense morphemes to receive interpretations that differ from those typically assumed for non-embedded tenses. In this paper, I explore an alternative view in which tenses are assumed to be uniformly defined for both independent and embedded occurrences. I argue that the problematic subordinate interpretations can be explained if appropriate definitions of tense meanings are provided and independent factors influencing the temporal interpretation are taken into account. Specifically, I suggest that tense meanings can be evaluation time sensitive and/or indexical, and that variations on these abstract properties account for cross-linguistic differences. Also, I suggest that tense meanings alone do not always yield the intended temporal interpretation of a sentence. Aktionsart properties and pragmatic inferences often determine the exact extension and location of the interval constructed. Thus, the interaction of tense meanings and general facts of the grammar, rather than sequence of tense specific mechanisms, conspire to explain temporal interpretation in all contexts.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/58.pdf}, } @article{LCNL:59, author={Silvia Gennari and David Poeppel}, year={2003}, title={Processing correlates of lexical semantic complexity.}, journal={Cognition}, volume={89}, pages={B27-41}, abstract={This paper explores how verb meanings that differ in semantic complexity are processed and represented. In particular, we compare eventive verbs, which denote causally structured events, with stative verbs, which denote facts without causal structure. We predicted that the conceptually more complex eventive verbs should take longer to process than stative verbs. Two experiments, a lexical decision task and a self-paced reading study, confirmed this prediction. The findings suggest that (a) semantic complexity is reflected in processing time, (b) processing verb meanings involves activating properties of the event structure beyond participants’ roles, and (c) more generally, lexical event structures, which subsume thematic roles, may mediate between syntactic knowledge and semantic interpretation.}, language={English}, keywords={Verb meaning; Semantic complexity; Verb processing; Event structure}, URL={http://lcnl.wisc.edu/publications/archive/59.pdf}, } @article{LCNL:60, author={Jason D. Zevin and David A. Balota}, year={2000}, title={Priming and attentional control of lexical and sublexical processes during naming.}, journal={Journal of Experimental Psychology: Learning, Memory and Cognition}, volume={26}, number={1}, pages={121-135}, abstract={A modified priming task was used to investigate whether skilled readers are able to adjust the degree to which lexicai and sublexical information contribute to naming. On each trial, participants named 5 low-frequency exception word primes or 5 nonword primes before a target. The low-frequency exception word primes should have produced a greater dependence on lexical information, whereas the nonword primes should have produced a greater dependence on sublexical information. Across 4 experiments, the effects of lexicality, regularity, frequency, and imageability were all modulated in predictable ways on the basis of the notion that the primes directed attention to specific processing pathways. It is argued that these results are consistent with an attentional control hypothesis.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/60.pdf}, } @article{LCNL:61, author={David A. Balota and Mark B. Law and Jason D. Zevin}, year={2000}, title={The Attentional Control of Lexical Processing Pathways: Reversing the Word-Frequency Effect.}, journal={Memory and Cognition}, volume={28}, pages={1081-1089}, language={English}, } @article{LCNL:62, author={Barbara C. Malt and Steven A. Sloman and Silvia Gennari and Meiyi Shi and Yuan Wang}, year={1999}, title={Knowing versus Naming: similarity and Linguistic Categorization of Artifacts.}, journal={Journal of Memory and Language.}, volume={40}, pages={230-262}, abstract={We argue that it is important to distinguish between categorization as object recognition and as naming because the relation between the two may not be as straightforward as has often been assumed. We present data from speakers of English, Chinese, and Spanish that support this contention. Speakers of the three languages show substantially different patterns of naming for a set of 60 common containers, but they see the similarities among the objects in much the same way. The observed patterns of naming therefore cannot arise only from the similarities that speakers of the three languages see among the objects. We also offer suggestions about how complexity in naming may arise, and the data provide some evidence consistent with these suggestions. Exploring how artifacts are named vs "known" may provide new insights into artifact categorization.}, language={English}, } @article{LCNL:63, author={Jennifer Merva Stedron and Sarah Devi Sahni and Yuko Munakata}, year={2005}, month={04}, title={Common mechanisms for working memory and attention: The case of perseveration with visible solutions.}, journal={Journal of Cognitive Neuroscience}, volume={17}, number={4}, pages={623-631}, abstract={Everyone perseverates at one time or another, repeating previous behaviors when they are no longer appropriate. Such perseveration often occurs in situations with working memory demands, and the ability to overcome perseveration has been linked to brain regions critical for working memory. Many theories thus explain perseveration in terms of working memory deficits. However, perseveration also occurs in situations without apparent working memory demands, in which the visible environment specifies appropriate behavior. Such findings appear to challenge working memory accounts of perseveration. To evaluate this challenge, a neural network model of a working memory account of perseveration was tested on tasks with visible solutions. With advances in the mechanisms that support working memory, networks became increasingly able to attend to relevant information in the environment. These developments led to improvements in performance on tasks with visible solutions, paralleling the developmental progression observed in infants. The simulations demonstrate how mechanisms of working memory can subserve perseveration and success on tasks with and without obvious memory demands. In both types of tasks, controlled processing occurs through the activation of task-relevant representations, which provide top.down biasing of other processing pathways. More generally, the simulations demonstrate how common mechanisms can support working memory and attention.}, language={English}, keywords={working memory; attention; perseveration; neural network model}, issn={0898929X}, URL={http://lcnl.wisc.edu/publications/archive/63.pdf}, } @article{LCNL:64, author={Silvia P. Gennari and Steven A. Sloman and Barbara C. Malt and W. Tecumseh Fitchd}, year={2002}, title={Motion events in language and cognition.}, journal={Cognition}, volume={83}, pages={49-79}, abstract={This study investigated whether different lexicalization patterns of motion events in English and Spanish predict how speakers of these languages perform in non-linguistic tasks. Using 36 motion events, we compared English and Spanish speakers’ linguistic descriptions to their performance on two non-linguistic tasks: recognition memory and similarity judgments. We investigated the effect of language processing on non-linguistic performance by varying the nature of the encoding before testing for recognition and similarity. Participants encoded the events while describing them verbally or not. No effect of language was obtained in the recognition memory task after either linguistic or non-linguistic encoding and in the similarity task after non-linguistic encoding. We did find a linguistic effect in the similarity task after verbal encoding, an effect that conformed to languagespecific patterns. Linguistic descriptions directed attention to certain aspects of the events later used to make a non-linguistic judgment. This suggests that linguistic and non-linguistic performance are dissociable, but language-specific regularities made available in the experimental context may mediate the speaker’s performance in specific tasks.}, language={English}, keywords={Whorfian hypothesis; Verb concept; Motion event; Lexicalization; English; Spanish; Recognition; Similarity; Path; Manner; Talmy}, URL={http://lcnl.wisc.edu/publications/archive/64.pdf}, } @article{LCNL:65, author={Barbara C. Malt and Steven A. Sloman and Silvia P. Gennari}, year={2003}, title={Universality and Language Specificity in Object Naming.}, journal={Journal of Memory and Language}, volume={49}, number={1}, pages={20-42}, abstract={Rather than having universal linguistic categories for sets of common objects, languages develop their own, idiosyncratic naming patterns for them. Accounting for these patterns requires reference not only to the understanding of stimulus properties by individual speakers of a language, but also to the linguistic and cultural histories of the language they speak. To better understand how these two sources of influence work together to produce linguistic categories, we examined the relations among linguistic categories for 60 common containers for speakers of English, Spanish, and Chinese. We discriminated among several possibilities that imply different relative contributions of the two sources of influence. No single type of relation dominated; the contributions of the two influences varied across different parts of the container domain. We suggest that perception of stimulus properties by individuals interacts with linguistic and cultural histories, but their interaction is constrained by structure in the stimulus space.}, language={English}, keywords={Naming; Categorization (linguistic); Linguistic diversity; Cross-linguistic comparison}, issn={0749596X}, } @article{LCNL:66, author={Silvia P. Gennari}, year={2004}, title={Temporal references and temporal relations in sentence comprehension.}, journal={Journal of Experimental Psychology: Learning, Memory and Cognition}, volume={30}, pages={877-890}, abstract={The author investigated the interpretation of temporal references during comprehension of sentences containing a main and subordinate clause. Experiments 1 and 2 examined state and event subordinate clauses, respectively, and showed that subordinate temporal references overlapping with or close to the time of the main clause event were read faster than nonoverlapping distant references. Experiment 3 examined temporal references in nonsubordinate main clauses and confirmed that temporal relations between main and subordinate clauses were established on-line in the previous experiments. Experiment 4 independently manipulated temporal overlap and distance and suggested that event and state clauses are processed according to distinct temporal parameters. The results are explained by the contingency relations that events and states establish with other discourse events.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/66.pdf}, } @article{LCNL:67, author={Michael T. Ullman and Elizabeth I. Pierpont}, year={2005}, title={Specific Language Impairment is not specific to language: the Procedural Deficit Hypothesis.}, journal={Cortex}, volume={41}, number={3}, pages={399-433}, abstract={Specific Language Impairment (SLI) has been explained by two broad classes of hypotheses, which posit either a deficit specific to grammar, or a non-linguistic processing impairment. Here we advance an alternative perspective. According to the Procedural Deficit Hypothesis (PDH), SLI can be largely explained by the abnormal development of brain structures that constitute the procedural memory system. This system, which is composed of a network of interconnected structures rooted in frontal/basal-ganglia circuits, subserves the learning and execution of motor and cognitive skills. Crucially, recent evidence also implicates this system in important aspects of grammar. The PDH posits that a significant proportion of individuals with SLI suffer from abnormalities of this brain network, leading to impairments of the linguistic and non-linguistic functions that depend on it. In contrast, functions such as lexical and declarative memory, which depend on other brain structures, are expected to remain largely spared. Evidence from an in-depth retrospective examination of the literature is presented. It is argued that the data support the predictions of the PDH, and particularly implicate Broca.s area within frontal cortex, and the caudate nucleus within the basal ganglia. Finally, broader implications are discussed, and predictions for future research are presented. It is argued that the PDH forms the basis of a novel and potentially productive perspective on SLI.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/67.pdf}, } @inproceedings{LCNL:68, author={James L. Keidel and Jason D. Zevin and Keith R. Kluender and Mark S. Seidenberg}, year={2003}, title={Modeling the role of native language knowledge in perceiving nonnative speech contrasts.}, booktitle={Proceedings of the 15th International Congress of Phonetic Sciences}, pages={2221-2224}, abstract={A novel connectionist model of speech perseption is presented which accounts for data regarding the perception of foreign speech sounds by native speakers of English. THe model was trained on a large corpus of English CV syllables, and then tested on isiZulu stimuli. Results were similar to those obtained in a study with human participants by Best et al. Importantly, the match between the model's performance and th ehuman data did not depend on the inclusion of articulatory information in the model.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/68.pdf}, } @inbook{LCNL:69, author={Joseph Allen and Mark S. Seidenberg}, year={1999}, title={The emergence of grammaticality in connectionist networks.}, publisher={Lawrence Earlbaum Associates}, location={Hillsdale, NJ}, editor={MacWhinney,Brian}, booktitle={Emergentist Approaches to Language: proceedings of the 28th Carnegie symposium on cognition}, pages={115-151}, comments={An important paper in which we show why "Colorless green ideas sleep furiously" can be judged grammatical without a grammar, and why grammaticality judgments can still be made by some "agrammatic" aphasics.}, abstract={In the course of discussing connectionist networks and the emergence of grammaticality, this chapter specifically discusses an alternative framework, grammaticality judgments, agrammatism, and simulating grammaticality judgments. The performance and results of such a connectionist simulation of language processing are presented and discussed. The model successfully acquired knowledge of language in the course of learning to comprehend and produce utterances.}, language={English}, keywords={connectionist model of language learning & emergence of grammaticality; Connectionism; Grammar; Language Development; Models; Neural Networks}, isbn={0805830103}, URL={http://lcnl.wisc.edu/publications/archive/69.pdf}, } @inbook{LCNL:70, author={Mark S. Seidenberg}, year={1999}, title={Visual word recognition.}, publisher={MIT Press.}, location={Cambridge, MA.}, booktitle={MIT Encyclopedia of Cognitive Science.}, language={English}, isbn={0262232006}, URL={http://lcnl.wisc.edu/publications/archive/70.pdf}, } @inbook{LCNL:71, author={Barbara C. Malt and Steven A. Sloman and Silvia P. Gennari}, year={2003}, title={Speaking versus thinking about objects and actions.}, publisher={MIT Press}, location={Cambridge, MA}, editor={Dedre Gentner and Susan Goldin-Meadow}, booktitle={Language in Mind: Advances in the Study of Language and Thought}, pages={81-112}, language={English}, isbn={0262571633}, } @inbook{LCNL:72, author={Silvia Gennari and Luisa Meroni and Stephen Crain}, year={2005}, title={Rapid relief of stress in dealing with ambiguity.}, journal={In J. Trueswell & M.K. Tanenhaus (Eds.) Approaches to studying world-situated language use. (pp. 245-259). Cambridge, MA: MIT Press.}, publisher={MIT Press}, location={Cambridge, MA}, editor={Trueswell,John and Tanenhaus,Michael K.}, booktitle={Approaches to studying world-situated language use.}, abstract={This study used a head mounted eye-movement recording system to investigate the influence of phonological stress on the interpretation of spoken sentences that described visual scenes. The test sentences contained the focus operator only as in (1)-(3) (where capital letters indicate marked stress and italics indicate default stress): (1) The mother only brought SOME MILK to the boy. (2) The mother only brought the boy some milk. (3) The mother only brought some milk to the boy. These sentences presuppose that a set of alternatives to the referent of the stressed constituent has been introduced in the context (the contrast set); and they assert that the focused entity has some unique property that members of the contrast set lack. Sentences (1)-(2) are distinguished from (3) both in the presupposition and in the assertion because of the location of stress. Reinhart (1999) claims that sentences like (1) tax the parser more than (2)-(3) because of an inherent preference in all three cases to compute a contrast set based on the sentence's default stress pattern. This initial computation must be subsequently abandoned, however, in sentence (1), in response to the shift of stress onto the Direct Object some milk. If so, sentences (2)-(3) should be easier than (1). An alternative hypothesis, based on empirical findings of on-line interpretation, makes the opposite prediction. Under this view, marked stress facilitates the computation of the appropriate contrast set in (1), while multiple contrast sets (one for each postverbal NP) may be computed in (2)-(3), due to the temporary ambiguities they contain, before the parser settles on the interpretation associated with default stress. To evaluate these hypotheses, we designed a study in which subjects were asked to judge the truth or falsity of spoken sentences like those exemplified in (1)-(3) against a visual scene, which contained possible contrast sets for each of the possible foci of the sentences. Results indicate that overall response accuracy was better and response time was shorter for sentences like (1) than for sentences like (2)-(3). Also, subjects fixated longer on the contrast set corresponding to the Direct Object in sentences like (1), as compared to other objects in the 3 scene, indicating that they did not find such sentences ambiguous. A different pattern of fixations was observed for sentences like (2)-(3). This pattern suggests that these sentences were temporarily ambiguous, forcing subjects to compute two contrast sets, one for each post-verbal NP. The findings suggest that the parser computes multiple interpretations of ambiguous sentences on-line, except when (marked) stress indicates how an ambiguity should be resolved.}, language={English}, isbn={0262701049}, URL={http://lcnl.wisc.edu/publications/archive/72.pdf}, } @conference{LCNL:73, author={Jelena Mirković and Maryellen C. MacDonald}, year={2003}, title={The role of morphophonological factors in agreement production: When singular and plural are both grammatical.}, organization={the 16th CUNY Conference on Human Sentence Processing}, location={Boston, MA.}, abstract={A central debate in language production research is the extent to which the computation of noun-verb agreement is an autonomous syntactic process or whether it is constrained by non-syntactic factors (e.g. Bock et al. 2001; Thornton & MacDonald, in press). The primary data for these alternatives have been the rates of agreement errors in fragment completion task (a speaker completes a complex noun phrase like "the key to the cabinets" with a verb that agrees with the local noun "cabinets" rather than the head noun "key"). Evidence for non-syntactic influences on agreement is mixed in these studies. Recently several researchers have identified constructions in which several grammatical options are available (Haskell & MacDonald, submitted; Hemforth & Konieczny, 2002). These constructions are potentially quite informative, because subtle non-syntactic effects may be more evident here than in cases where only one agreement pattern is grammatical. We investigate another case of this sort, subject-verb number agreement with certain quantifier phrases in Serbian such as "five cows", for which both singular and plural verbs are grammatical.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/73.ppt}, } @conference{LCNL:74, author={Daniel J. Acheson and Maryellen C. MacDonald}, year={2004}, title={Phonological Interference in Working Memory and Sentence Comprehension.}, organization={the 45th Annual Meeting of the Psychonomic Society}, location={Minneapolis, MS.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/74.pdf}, } @conference{LCNL:75, author={Daniel J. Acheson and Maryellen C. MacDonald}, year={2005}, title={A Role for Phonological Information in Relative Clause Processing.}, organization={the 17th Annual CUNY Conference on Human Sentence Processing}, location={Tucson, AZ}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/75.pdf}, } @conference{LCNL:76, author={Jelena Mirković and Mark S. Seidenberg and Marc F. Joanisse}, year={2002}, title={Morphology in an inflectionally rich language: Implications for the rules vs. connections debate.}, organization={the 43rd Annual Meeting of the Psychonomic Society}, location={Kansas City, KS}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/76.pdf}, } @conference{LCNL:77, author={Aimee Arnoldussen and Susan Bookheimer and Frank Manis and Mark S. Seidenberg}, year={2002}, title={Dyslexic subtypes: An fMRI study of reading pathways.}, organization={the Annual Meeting of the Cognitive Neuroscience Society}, location={San Francisco, CA}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/77.ppt}, } @conference{LCNL:78, author={Aimee Arnoldussen and Julia Evans and Mark S. Seidenberg}, year={2004}, title={Differential Effects of Slowing on Orthographic, Phonological and Semantic Processing in Children with Specific Language Impairment.}, organization={the annual meeting of the Cognitive Neuroscience Society}, location={San Francisco, CA.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/78.ppt}, } @conference{LCNL:79, author={David A. Medler and Aimee Arnoldussen and Jeffrey R. Binder and Rutvik Desai and Lisa L. Conant and Mark S. Seidenberg}, year={2004}, title={Activation of Semantic Regions by the Printed Word: An fMRI Study.}, organization={Cognitive Neuroscience Society Meeting}, location={San Francisco, CA}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/79.pdf}, } @conference{LCNL:80, author={Aimee Arnoldussen and Julia Evans and Mark S. Seidenberg}, year={2004}, title={Performance Matching: Comparing Children with Specific Language Impairment to Younger Children with Similar Abilities using fMRI.}, organization={Symposium on Research in Child Language Disorders}, location={Madison, WI}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/80.ppt}, } @inbook{LCNL:81, author={Mark S. Seidenberg and Jason D. Zevin}, year={2006}, title={Connectionist models in developmental cognitive neuroscience: Critical periods and the paradox of success.}, publisher={Oxford University Press.}, location={Oxford, UK}, editor={Munakata,Yuko and Johnson,Mark}, booktitle={Processes of Change in Brain and Cognitive Development. Attention and Performance XXI}, comments={A reinterpretation of the age-related decline in language learning capacity often attributed to a critical or sensitive period under maturational rather than experiential control. Drawing on computational models and data from both humans and song learning in zebra finches, we conclude that this decline has a different basis, which we term the Paradox of Success: success in learning a first language creates conditions that interfere with subsequent learning.}, abstract={Connectionist models have made significant contributions to understanding developmental phenomena, mainly by providing novel computational accounts of behavioral emergence and change. What is the fate of such models given the increasing interest in and information about the biological bases of development? We consider this issue with respect to the classical idea of a critical period for acquiring language. The standard view is that neurobiological developments on a strict maturational timetable create limits on language learning capacity. Computational analysis suggests the opposite: that learning itself creates neurobiological conditions underlying the "closing" of the critical period. The critical period example suggests how connectionist models can continue to provide a necessary level of analysis intermediate between behavior and brain.}, language={English}, isbn={0198568746}, URL={http://lcnl.wisc.edu/publications/archive/81.pdf}, } @inbook{LCNL:82, author={Mark S. Seidenberg}, year={1992}, title={Connectionism without tears.}, publisher={Oxford University Press}, location={New York, NY.}, editor={Davis,Steven}, booktitle={Connectionism: Theory and practice.}, pages={84-122}, abstract={my own work has been grounded in the belief that connectionism and linguistics have more in common than some of the more polemical critiques of the approach would suggest / discuss some of the issues that have tended to separate the two approaches and describe some potentially interesting points of contact (from the chapter) This chapter is followed by a commentary entitled "Connectionist Models in the Information Processing Paradigm" by Michael E. J. Masson.}, language={English}, isbn={0195076656}, URL={http://lcnl.wisc.edu/publications/archive/82.pdf}, } @article{LCNL:83, author={Todd R. Haskell and Maryellen C. MacDonald}, year={2005}, title={Constituent structure and linear order in language production: Evidence from subject verb agreement. }, journal={Journal of Experimental Psychology: Learning, Memory and Cognition}, volume={35}, number={5}, pages={891-904}, abstract={A number of studies have shown that structural factors play a much larger role than the linear order of words during the production of grammatical agreement. These findings have been used as evidence fora stage in the production process at which hierarchical relations between constituents have beenestablished (a necessary precursor to agreement), but before the final linear order of words is determined.The current article combines evidence from off-line ratings, online production studies, and a corpusanalysis in support of the view that linear order effects do exist. These findings have implications bothfor theories of agreement production and language production more generally.}, language={English}, keywords={subject verb agreement; syntactic structure; word order; language production; Grammar; Language; Sentence Structure; Syntax; Words (Phonetic Units); Verbs}, issn={02787393}, URL={http://lcnl.wisc.edu/publications/archive/83.pdf}, } @article{LCNL:84, author={Jason D. Zevin and Mark S. Seidenberg and Sarah W. Bottjer}, year={2004}, month={06}, title={Limits on reacquisition of song in adult zebra finches exposed to white noise.}, journal={Journal of Neuroscience}, volume={24}, pages={5849-5862}, abstract={Zebra finches (Taeniopygia guttata) learn a specific song pattern during a sensitive period of development, after which song changes littleor not at all. However, recent studies have demonstrated substantial behavioral plasticity in song behavior during adulthood under arange of conditions. The current experiment examined song behavior of adult zebra finches temporarily deprived of auditory feedback bychronic exposure to loud white noise (WN). Long-term exposure to continuousWNresulted in disruption of song similar to that observedafter deafening. When auditory feedback was restored by discontinuing WN, birds were either tutored using tape-recorded playback orhoused with adult conspecific tutors.Noevidence of learning new tutor syllables was observed, and recovery of pre-WN song patterns wasvery limited after restoration of hearing. However, many birds did reacquire some aspects of their pretreatment song, suggesting an adultform of learning thatmayretain some of the initial aspects of sensorimotor acquisition of song in which vocalizations are shaped to matcha stored template representation. The failure to learn novel song elements and the modest degree of recovery observed overall suggest alimit on plasticity in adult birds that have acquired species-typical song patterns and may reflect an important species difference betweenzebra finches and Bengalese finches.}, language={English}, keywords={sensitive period; behavioral plasticity; songbirds; vocalization; auditory learning; song learning}, URL={http://lcnl.wisc.edu/publications/archive/84.pdf}, } @inbook{LCNL:85, author={Maryellen C. MacDonald}, year={1999}, title={Distributional information in language comprehension, production, and acquisition: Three puzzles and a moral.}, publisher={Erlbaum}, location={Mahwah, NJ}, editor={MacWhinney,Brian}, booktitle={The Emergence of Language}, pages={177-196}, abstract={Discusses 3 interrelated findings in language comprehension, production, and acquisition. In each case, the puzzling results in one field appear to have solutions in another. The intricate relationships between these puzzles hold important implications for the nature of the human language faculties and for the isolationist research strategies that currently dominate psycholinguistic research. The author argues that a sensitivity to distributional information is an important link between comprehension, production, and acquisition. The puzzles addressed include findings in syntactic ambiguity resolution, exceptions to incremental speech production, and the acquisition of distributional information. The moral mentioned in the title is: If this general account is on the right track, then the acquisition, comprehension, and production processes have links between them that cannot be safely ignored.}, language={English}, keywords={distributional information & relationships between language comprehension & production & acquisition; Comprehension; Language Development; Psycholinguistics}, isbn={0805830103}, URL={http://lcnl.wisc.edu/publications/archive/85.pdf}, } @inbook{LCNL:86, author={Robert Thornton and Mariela Gil and Maryellen C. MacDonald}, year={1998}, title={Accounting for Cross-Linguistic Variation: A Constraint-Based Perspective}, publisher={Academic Press}, location={San Deigo, CA}, editor={Hillert,Dieter}, booktitle={Syntax and Semantics, Vol. 31: A crosslinguistic perspective.}, pages={211-225}, abstract={Much of the research on sentence processing has focused on discovering universal principles to explain parsing preferences. Recent cross-linguistic work, however, suggests that processing is sensitive to distributional information about individual languages. Along these lines, we explore a constraint-based approach to processing, in which cross-linguistic variation is explained by the interaction of language specific grammatical constraints with more general pragmatic principles. Specifically, we examine the role of pragmatic information in constraining the modification of complex noun phrases in English and Spanish. We first present data suggesting that, for both languages, initial comprehension is constrained in the same manner by pragmatic information. We then pursue an explanation of crosslinguistic differences in terms of pragmatic constraints on grammatical differences between the languages.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/86.pdf}, } @conference{LCNL:87, author={Elizabeth I. Pierpont and Mark S. Seidenberg}, year={2005}, title={Language and auditory sequence processing: individual differences in memory for serial order.}, organization={the 46th Annual Meeting of the Psychonomic Society}, location={Toronto, Ontario, Canada}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/87.pdf}, } @conference{LCNL:88, author={Daniel J. Acheson and Maryellen C. MacDonald}, year={2005}, title={A Phonological Similarity Advantage in Serial Recall.}, organization={the 46th Annual Meeting of the Psychonomic Society}, location={Toronto, ON}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/88.pdf}, } @article{LCNL:89, author={Jason D. Zevin and Mark S. Seidenberg}, year={2006}, month={02}, title={Consistency effects and individual differences in nonword naming: A comparison of current models.}, journal={Journal of Memory and Language}, volume={54}, number={2}, pages={145-160}, comments={Think the DRC model does better at reading nonwords than connectionist models? Read this paper and decide.}, abstract={The mechanisms underlying nonword pronunciation have a been a focus of debates over dual-route and connectionist models of reading aloud. The present study examined two aspects of nonword naming: spelling-sound consistency effects and variability in the pronunciations assigned to ambiguous nonwords such as MOUP. Performance of a parallel distributed processing model was assessed over multiple runs, representing multiple subjects with varying reading experience. The model provided a good account of behavioral data concerning these phenomena. In contrast, the Dual Route Cascaded model does not produce consistency effects and does not account for the alternative pronunciations that subjects produce. The results highlight the importance of considering multiple aspects of a phenomenon such as nonword naming in assessing computational models.}, language={English}, keywords={individual differences; pronunciation; nonword naming; spelling sound consistencey; Naming; Oral Reading; Words (Phonetic Units)}, issn={0749596X}, URL={http://lcnl.wisc.edu/publications/archive/89.pdf}, } @article{LCNL:90, author={Anne J. Sperling and Zhong-Lin Lu and Franklin R. Manis and Mark S. Seidenberg}, year={2006}, title={Motion perception deficits and reading impairment: It's the noise not the motion.}, journal={Psychological Science}, volume={17}, pages={1047-1053}, abstract={We tested the hypothesis that deficits on sensory processing tasks frequently associated with poor reading and dyslexia are the result of impairments in external noise exclusion, rather than motion perception or magnocellular processing. We compared the motion direction discrimination thresholds of adults and children with good or poor reading performance, using coherent motion displays embedded in external noise. Both adult and child poor readers had higher thresholds in the presence of high external noise, but did not differ from their respective peers in low external noise or when the signal was clearly demarcated. Adults' performance in high external noise correlated with general reading ability, whereas children's performance correlated with language and verbal abilities. The results support the hypothesis that noise exclusion deficits impair reading and language development, suggesting that the impact of such deficits on the development of reading skills changes with age.}, language={English}, URL={http://lcnl.wisc.edu/publications/archive/90.pdf}, } @inbook{LCNL:91, author={Mark S. Seidenberg and David C. Plaut}, year={2006}, title={Progress in understanding word reading: Data fitting versus theory building.}, publisher={Psychology Press.}, location={Hove, UK.}, editor={Andrews,Sally}, booktitle={From inkmarks to ideas: Current issues in lexical processing.}, comments={A comparison of the DRC and PDP approaches to reading. Pitfalls of fitting models to data.}, language={English}, isbn={1841696072}, URL={http://lcnl.wisc.edu/publications/archive/91.pdf}, } @article{LCNL:92, author={Karen Stevens Dagerman and Maryellen C. MacDonald and Michael W. Harm}, year={2006}, title={Aging and the Use of Context in Ambiguity Resolution: Complex Chnages From Simple Slowing.}, journal={Cognitive Science}, volume={30}, pages={311-345}, abstract={Older and younger adults' abilities to use context information rapidly during ambiguity resolution were investigated. In Experiments 1 and 2, younger and older adults heard ambiguous words (e.g., fires) in sentences where the preceding context supported either the less frequent or more frequent meaning of the word. Both age groups showed good context use in offline tasks, but only young adults demonstrated rapid use of context in cross-modal naming. A 3rd experiment demonstrated that younger and older adults had similar knowledge about the contexts used in Experiments 1 and 2. The experiment results were simulated in 2 computational models in which different patterns of context use were shown to emerge from varying a single speed parameter. These results suggest that age-related changes in processing efficiency can modulate context use during language comprehension.}, language={English}, keywords={Language comprehension; Cognitive aging; Ambiguity resolution; Computational modeling}, URL={http://lcnl.wisc.edu/publications/archive/92.pdf}, } @inproceedings{LCNL:94, author={Kim G. Daugherty and Maryellen C. MacDonald and Alan S. Petersen and Mark S. Seidenberg}, year={1993}, title={Why no mere mortal has ever flown out to center field but people often say they do.}, booktitle={Proceedings of the Fifteenth Annual Conference of the Cognitive Science Society}, pages={383-388}, abstract={The past tense has been the source of considerable debate concerning the role of connectionist models in explaining linguistic phenomena. In response to Pinker and Prince (1988), several connectionist models have been developed that compute a mapping between the present tense phonological form of a verb to a past tense phonological form. Most of these models cannot distinguish between homophones such as FLY-FLEW and FLY-FLIED (as in "flied out"). Kim, Pinker, Prince, & Prasada (1991) have suggested that the addition of semantic information to such nets will not provide an adequate solution to this homophony problem. They showed that English speakers use derivational status, rather than semantic information in generating past tenses. We provide evidence contradicting this account. Subjects' rated preferences for past tense forms are predicted by semantic measures; moreover, a simulation model shows that semantic distance provides a basis for learning the alternative past tenses for words such as FLY. We suggest a reconciliation of the two theories in which knowledge of "derivational status" arises out of semantic facts in the course of learning.}, URL={http://lcnl.wisc.edu/publications/archive/94.pdf}, } @inbook{LCNL:97, author={M. C. MacDonald and M. S. Seidenberg}, year={2006}, title={Constraint satisfaction accounts of lexical and sentence comprehension.}, publisher={Elsevier Inc.}, location={London}, editor={M. J. Traxler and M. A. Gernsbacher}, booktitle={Handbook of Psycholinguistics, 2nd Edition}, chapter={15}, pages={581-611}, URL={http://lcnl.wisc.edu/publications/archive/97.pdf}, } @article{LCNL:98, author={S. P. Gennari and M. C. MacDonald}, year={2006}, title={Acquisition of negation and quantification: Insights from adult production and comprehension.}, journal={Language Acquisition}, volume={13}, pages={125-168}, URL={http://lcnl.wisc.edu/publications/archive/98.pdf}, } @inbook{LCNL:99, author={K. Patterson and M. C. MacDonald}, year={2006}, title={Sweet nothings: Narrative speech in semantic dementia.}, publisher={Psychology Press}, location={Hove, UK}, editor={S. Andrews}, booktitle={From inkmark to ideas: Current issues in lexical processing}, pages={299-317}, URL={http://lcnl.wisc.edu/publications/archive/99.pdf}, } @conference{LCNL:100, author={Elizabeth I. Pierpont and Mark S. Seidenberg and Mary Ella Pierpont}, year={2007}, month={November}, title={Development of communication and adaptive skills in individuals with Noonan syndrome}, organization={the International Symposium on Cardiofaciocutaneous syndrome and Noonan syndrome}, location={Washington, D.C.}, URL={http://lcnl.wisc.edu/publications/archive/100.pdf}, } @article{LCNL:101, author={Mark S. Seidenberg and Maryellen C. MacDonald and Todd R. Haskell}, year={2007}, title={Semantics and phonology constrain compound formation}, journal={The Mental Lexicon}, volume={2}, pages={287-312}, URL={http://lcnl.wisc.edu/publications/archive/101.pdf}, } @article{LCNL:102, author={Silvia P. Gennari and Maryellen C. MacDonald}, year={2008}, title={Semantic indeterminacy in object relative clauses}, journal={Journal of Memory and Language}, volume={58}, number={2}, pages={161-187}, URL={http://lcnl.wisc.edu/publications/archive/102.pdf}, } @article{LCNL:103, author={Daniel J. Acheson and Justine B. Wells and Maryellen C. MacDonald}, year={2008}, title={New and updated tests of print exposure and reading abilities in college students}, journal={Behavior Research Methods}, volume={40}, number={1}, pages={278-289}, URL={http://lcnl.wisc.edu/publications/archive/103.pdf}, } @conference{LCNL:104, author={Rachel S. Sussman and Maryellen C. MacDonald}, year={2007}, title={When is a spatula just a spatula?: Investigating the link between observed events and the online interpretation of verb-instrument biases.}, organization={the 4th Hopkins Workshop on Language: Grammar in Cognition}, location={Baltimore, MD}, URL={http://lcnl.wisc.edu/publications/archive/104.pdf}, } @conference{LCNL:106, author={Daniel J. Acheson and Maryellen C. MacDonald}, year={2007}, title={Phonological Similarity Effects in Verbal Working Memory and Language Production Tasks}, publisher={48th Annual Meeting of the Psychonomic Society}, location={Long Beach, CA}, URL={http://lcnl.wisc.edu/publications/archive/106.pdf}, } @article{LCNL:132, author={Maryellen C. MacDonald}, year={1994}, title={Probabilistic constraints and syntactic ambiguity resolution}, journal={Language and Cognitive Processes}, volume={9}, number={2}, pages={157-201}, URL={http://lcnl.wisc.edu/publications/archive/132.pdf}, } @article{LCNL:113, author={Laura Gonnerman and Mark S. Seidenberg and Elaine Andersen}, year={2007}, title={Graded semantic and phonological similarity effects in priming: Evidence for a distributed connectionist approach to morphology.}, journal={Journal of Experimental Psychology:General}, volume={136}, number={2}, pages={323-345}, comments={This article reports research (dating from the mid-90s, actually) addressing whether there is a distinct, structurally defined level of morphological structure (a la Marslen-Wilson, Tyler, Waksler, & Older, Psych Rev, 1994) or whether morphological units arise from correlations among semantics, phonology, and (in literate people) orthography. }, abstract={A considerable body of empirical and theoretical research suggests that morphological structure governs the representation of words in memory and that many words are decomposed into morphological components in processing. We investigated an alternative approach in which morphology arises from the interaction of semantic and phonological codes. A series of cross-modal lexical decision experiments show that the magnitude of priming reflects the degree of semantic and phonological overlap between words. Crucially, moderately similar items produce intermediate facilitation (e.g., lately-late). This pattern is observed for word pairs exhibiting different types of morphological relationship, including suffixed-stem (e.g., teacher-teach), suffixed-suffixed (e.g., saintly-sainthood) and prefixed-stem pairs (preheat-heat). The results can be understood in terms of connectionist models employing distributed representations rather than discrete morphemes.}, keywords={morphology, derivational morphology, connectionist models}, URL={http://lcnl.wisc.edu/publications/archive/113.pdf}, } @inbook{LCNL:118, author={Mark S. Seidenberg}, year={2006}, title={Connectionist Models of Reading}, publisher={Oxford University Press}, editor={Gareth Gaskell}, booktitle={The Oxford Handbook of Psycholinguistics}, chapter={14}, edition={first}, pages={235-25}, note={As advertised: a pretty simple overview of connectionist models of reading. Might dispel some misconceptions about methodology, general approach.}, comments={As advertised: a pretty simple overview of connectionist models of reading. Might dispel some misconceptions about methodology, general approach.}, URL={http://lcnl.wisc.edu/publications/archive/118.pdf}, } @article{LCNL:135, author={Silvia P. Gennari and Maryellen C. MacDonald}, year={2009}, month={April}, title={Linking production and comprehension processes: The case of relative clauses}, journal={Cognition}, volume={111}, number={1}, pages={1-23}, abstract={Six studies investigated the relationship between production and comprehension by examining how relative clause production mechanisms influence the probabilistic information used by comprehenders to understand these structures. Two production experiments show that accessibility-based mechanisms that are influenced by noun animacy and verb type shape relative clause production. Two corpus studies confirm these production mechanisms in naturally occurring productions. Two comprehension studies found that nouns and verb types occurring in structures that speakers do not produce are difficult to comprehend. Specifically, the probability of producing a passive structure for a verb type in a given animacy configuration, as measured in the production and corpus studies, predicts comprehension difficulty in active structures. Results suggest that the way in which the verb roles are typically mapped onto syntactic arguments in production plays a role in comprehension. Implications for the relationship between production, comprehension and language learning are discussed.}, URL={http://lcnl.wisc.edu/publications/archive/135.pdf}, } @inproceedings{LCNL:123, author={Jon A. Willits and Rachel S. Sussman and Michael S. Amato}, year={2008}, title={Event Knowledge vs. Verb Knowledge.}, booktitle={Cognitive Science Society}, URL={http://lcnl.wisc.edu/publications/archive/123.pdf}, } @article{LCNL:134, author={Daniel J. Acheson and Maryellen C. MacDonald}, year={2009}, title={Twisting tongues and memories: Explorations of the relationship between language production and verbal working memory. }, journal={Journal of Memory and Language}, volume={60}, pages={329-350}, URL={http://lcnl.wisc.edu/publications/archive/134.pdf}, } @conference{LCNL:133, author={Daniel J. Acheson and Massihula Hamidi and Jeffrey R. Binder and Bradley R. Postle}, year={2008}, title={Exploring the Relationship between Verbal Working Memory and Language Production using fMRI and TMS}, organization={Society for Neuroscience}, location={Washington, DC}, URL={http://lcnl.wisc.edu/publications/archive/133.pdf}, } @conference{LCNL:131, author={Michael S. Amato and Maryellen C. MacDonald}, year={2008}, title={The Influence of Second-Order, Non-Adjacent Constraints on Reading Times in an Artificial Language Paradigm.}, organization={CUNY Conference on Human Sentence Processing}, URL={http://lcnl.wisc.edu/publications/archive/131.pdf}, } @inproceedings{LCNL:136, author={Michael S. Amato and Jon A. Willits and Maryellen C. MacDonald}, year={2009}, title={Verb aspect and argument activation: World vs. Word Knowledge}, journal={CUNY Conference on Human Sentence Processing 2009}, abstract={Much sentence processing work has investigated the role of structural patterns vs. item-specific information (e.g., verb transitivity). Relatively little work has examined information that is neither item-specific nor structural. We address this gap in four studies examining the role of aspect in verb argument processing. Ferretti, Kutas, and McRae (2007) found shorter naming RTs for target locations (e.g., kitchen) following related imperfect aspect verbs (IA: was cooking) compared to perfect aspect (PA: had cooked). They argued that IA verbs evoke a situation model that selectively activates likely location arguments from an event schema, while PA does not. This suggests an abstract effect of aspect that is independent of the specific verb being used. Study 1 replicated Ferretti et al.’s materials and procedure and found the same effect of aspect. Study 2 was a corpus analysis investigating aspect-verb-noun co-occurrence probabilities, namely whether kitchen is more likely given was cooking than had cooked. We counted target locations within a 9-word window of the IA/PA in two corpra: a 532 million word corpus of Wikipedia articles and the 360 million word Corpus of Contemporary American English. Location targets were more probable in the context of the IA than the PA verbs, suggesting the IA advantage in naming could arise from participant sensitivity to conditional probabilities between specific pairs of verb-forms and nouns. Study 3 used corpus data to manipulate aspect and co-occurrence probability independently. 30 target locations were selected, each with 6 verb primes crossing aspect (IA/PA) and verb-location relationship (related high-probability, related low-probability, unrelated). Chosen verbs had large probability differences in IA and PA to the same target: primes for Ocean and P(ocean|prime) high-prob low-prob unrelated was drifting (1.11%) was sinking (0.37%) was assaulting (0.00%) had sunk (1.73%) had drifted (0.20%) had assaulted (0.00%) In norming, high and low probability primes were judged equally related to the target, and more so than were unrelated primes. The same naming task was used with these new items (n=78). ANOVA revealed a main effect of relatedness; RTs were shorter following related primes, F1(2,72)=6.57, p<.01, F2(2,24)=8.18, p<.01; and a marginal interaction: low-probability verb primes sped target naming only in IA, F1(2,72)=2.89, p=.06, F2(2,24)=4.83, p=.02. There was no main effect of aspect, Fs<1. These results suggest primes activated arguments that were statistically likely for the specific verb regardless of aspect, but that low probability locations got a boost from IA primes. Study 4 used sentence completion to test the effects of aspect and co-occurrence probability in sentence contexts. 95 participants completed sentence fragments containing a subject NP and IA/PA phrase (Mary was drifting...). The original Ferretti et al. verbs elicited more location completions in IA than PA, t(20)=.2.75, p<.05. However the Study 3 verbs, deliberately chosen to eliminate the IA probability advantage for one particular location, elicited an equal number of locations across aspects t<1. These results confirm that aspect influences argument activation during sentence processing, but also demonstrate that the effect is contingent on verb-specific information. Implications for generalizations from item-specific constraints are discussed. }, language={English}, keywords={word vs. world, aspect, item specific}, URL={http://lcnl.wisc.edu/publications/archive/136.pdf}, } @article{LCNL:137, author={Amit Almor and Justin M. Aronoff and Maryellen C. MacDonald and Laura M. Gonnerman and Daniel Kempler and Houri Hintiryan and UnJa L. Hayes and Sudha Arunachalam and Elaine S. Andersen}, year={2009}, title={A common mechanism in verb and noun naming deficits in Alzheimer’s patients}, journal={Brain and Language}, volume={111}, pages={8-19}, URL={http://lcnl.wisc.edu/publications/archive/137.pdf}, } @conference{LCNL:138, author={Jessica L. Montag and Maryellen C. MacDonald}, year={2009}, title={Measuring Production Difficulty in Object Relative Clauses}, organization={the 22nd Annual CUNY Conference on Human Sentence Processing}, location={Davis, CA}, URL={http://lcnl.wisc.edu/publications/archive/138.pdf}, } @inproceedings{LCNL:139, author={Jessica L. Montag and Maryellen C. MacDonald}, year={2009}, title={Word Order Doesn’t Matter: Relative Clause Production in English and Japanese}, editor={N.A. Taatgen & H. van Rijn}, booktitle={Proceedings of the 31st Annual Meeting of the Cognitive Science Society}, keywords={language production; relative clauses; animacy; cross-linguistic studies}, URL={http://lcnl.wisc.edu/publications/archive/139.pdf}, } @conference{LCNL:140, author={Michael S. Amato and Jon A. Willits and Maryellen C. MacDonald and Rachel S. Sussman}, year={2009}, title={Learning Language Statistics when World Statistics are Equal}, organization={Psychonomics Society}, abstract={Knowing verbs entails two types of knowledge: knowledge about the kinds of events in the world they refer to, and their patterns of co-occurrence with other words in language. We report a word learning experiment, which investigates learning from language statistics when world statistics are held constant. Participants learned about novel verbs by watching them acted out in videos, and by reading and hearing the verbs embedded in English sentences. Events in videos were always performed with instruments. For some verbs, sentences always mentioned an instrument; for others only 20% of sentences mentioned an instrument. Despite instruments being equally prevalent in the events the verbs referred to, self-paced reading times showed participants differentially expected instrument arguments based on the unequal language statistics. This suggests an important role for word co-occurrence patterns during comprehension, in addition to the accepted role of world knowledge about the events to which verbs refer.}, language={English}, keywords={language statistics, events}, URL={http://lcnl.wisc.edu/publications/archive/140.pdf}, } @article{LCNL:141, author={Maryellen C. MacDonald and Robert Thornton}, year={2009}, title={When language comprehension reflects production constraints: Resolving ambiguities with the help of past experience}, journal={Memory & Cognition}, volume={37}, pages={1177-1186}, URL={http://lcnl.wisc.edu/publications/archive/141.pdf}, } @article{LCNL:142, author={Morten H. Christiansen and Maryellen C. MacDonald}, year={2009}, title={A usage-based approach to recursion in sentence processing}, journal={Language Learning}, volume={59}, pages={126-161}, URL={http://lcnl.wisc.edu/publications/archive/142.pdf}, }