Linking production and comprehension processes: The case of relative clauses

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.     

Persian speaking children's acquisition of relative clauses

The current study examined the acquisition of relative clauses (RCs) in Persian-speaking children. Persian is a relatively unique data point in crosslinguistic research in acquisition because it is a head-final language with post-nominal RCs. Children (N = 51) aged 2 to 7 years completed a picture-selection task that tested their comprehension of subject-, object-, and genitive-RCs. The results showed that the children experienced greater difficulty processing object and genitive RCs when compared to subject RCs, suggesting that the children have particular difficulty processing sentences with non-canonical word order. The results are discussed with reference to a number of theoretical accounts proposed to account for sentence difficulty.     

Experience and sentence processing: statistical learning and relative clause comprehension

Many explanations of the difficulties associated with interpreting object relative clauses appeal to the demands that object relatives make on working memory. MacDonald and Christiansen [MacDonald, M. C., & Christiansen, M. H. (2002). Reassessing working memory: Comment on Just and Carpenter (1992) and Waters and Caplan (1996). Psychological Review, 109, 35-54] pointed to variations in reading experience as a source of differences, arguing that the unique word order of object relatives makes their processing more difficult and more sensitive to the effects of previous experience than the processing of subject relatives. This hypothesis was tested in a large-scale study manipulating reading experiences of adults over several weeks. The group receiving relative clause experience increased reading speeds for object relatives more than for subject relatives, whereas a control experience group did not. The reading time data were compared to performance of a computational model given different amounts of experience. The results support claims for experience-based individual differences and an important role for statistical learning in sentence comprehension processes.     

Processing strategies in the acquisition of relative clauses: universal principles and language-specific realizations

In this study we present a cross-linguistic analysis of the strategies used by Korean, Japanese and English-speaking children in processing sentences with relative clauses. The results of an experiment on the comprehension of relative clauses in Korean are reported and compared with prior research on the acquisition of relative clauses in Japanese and English. In our experiment on Korean, 6-year-old children acted out sentences with left-branching and center-embedded relative clauses in two matrix word orders, SOV and OSV, and two intonation conditions, “clear” (syntactically motivated) and “list” intonation. The findings provide strong evidence for a basic left-to-right processing strategy and significant roles for a canonical sentence strategy and for a parallel function strategy in the comprehension of relative clauses in Korean. Intonation had a significant role as a parsing cue in center-embedded sentences; error analysis revealed the over-extension of strategies appropriate for processing simple and conjoined sentences. A rank ordering of sentence types based on the predictions of the available processing strategies was found to predict the obtained order of difficulty in comprehending sentences with relative clauses in Korean, Japanese, and English quite accurately. We propose that an adequate cross-linguistic account of relative clause comprehension must be based upon an integrated view of multiple universal processing strategies, whose application will depend upon language-specific structural properties of relative clauses and upon the developmental stage of the child.     

Age constraints on first versus second language acquisition: evidence for linguistic plasticity and epigenesis

Does age constrain the outcome of all language acquisition equally regardless of whether the language is a first or second one? To test this hypothesis, the English grammatical abilities of deaf and hearing adults who either did or did not have linguistic experience (spoken or signed) during early childhood were investigated with two tasks, timed grammatical judgement and untimed sentence to picture matching. Findings showed that adults who acquired a language in early life performed at near-native levels on a second language regardless of whether they were hearing or deaf or whether the early language was spoken or signed. By contrast, deaf adults who experienced little or no accessible language in early life performed poorly. These results indicate that the onset of language acquisition in early human development dramatically alters the capacity to learn language throughout life, independent of the sensory-motor form of the early experience.     

Shyness–anxiousness and receptive language skills development in Spanish- and English-speaking preschoolers

The present study utilized a short-term longitudinal research design to model the relationship between shyness–anxiousness and receptive language skills. Hypotheses regarding the direction of the causal relationship, mediation, and moderation were evaluated. Subjects included 340 Head Start attendees from primarily English- and Spanish-speaking homes. Results suggest that a unidirectional relationship between shyness–anxiousness and receptive language skills emerges in preschool. Shyness–anxiousness impacted receptive language skills and this relationship was mediated by communication competence. Neither gender nor native language moderated the indirect relationship from shyness–anxiousness to receptive language through communication competence. These results are discussed with respect to elucidating the nature of the bidirectional relationship between shyness–anxiousness and language skills and identifying intervention targets to improve language outcomes for shy–anxious preschoolers.     

Another word on parsing relative clauses: Eyetracking evidence from Spanish and English

Ambiguity as to what the relative clause modifies in phrases such as Someone shot the maid of the actress who was divorced/Alguien disparó contra la criada de la actriz que estaba divorciada tends to be resolved differently in different languages (and in different forms of complex noun phrases). In English, there is a weak but seldom significant tendency for the relative clause to be taken as modifying the second noun phrase, the actress, but in Spanish, several researchers have found a significant preference for the relative clause's modifying the first noun phrase, la criada. The present experiments compare Spanish and English readers' eye movements while reading exactly comparable sentences in their native languages and find a significant reading time advantage in Spanish when it is forced to modify the first noun phrase, but in English when the relative clause is forced to modify the second noun phrase. Theoretical implications of the findings for previous explanations of the phenomenon are discussed.     

Statistical language learning: mechanisms and constraints

What types of mechanisms underlie the acquisition of human language? Recent evidence suggests that learners, including infants, can use statistical properties of linguistic input to discover structure, including sound patterns, words, and the beginnings of grammar. These abilities appear to be both powerful and constrained, such that some statistical patterns are more readily detected and used than others. Implications for the structure of human languages are discussed.     

Comprehension of reversible relative clauses in specifically language impaired and normally developing Greek children

This paper aims to investigate the syntactic comprehension of reversible relative clauses in a group of eight Greek children with specific language impairment (SLI) and two control groups of normally developing children matched on chronological and language age, respectively. An experiment using an acting out procedure was undertaken. Group analysis revealed that SLI children's performance is qualitatively different than that of both control groups. Interpreting the data, processes involved in syntactic comprehension are taken into consideration. It is claimed that processing demands impede SLI children's performance due to a deficit in their competence grammar.     

Learning Verbs Without Arguments: The Problem of Raising Verbs

This paper addresses the problem of learning the class of raising verbs (e.g. seem). These verbs are potentially problematic for learners in that unlike typical main verbs, these verbs do not stand in a semantic relation with any Noun Phrase (NP) arguments. Moreover, a second class of verbs, known as control verbs, shares certain distributional properties with raising verbs, but the two verb classes differ in important structural properties. The central problem addressed here is that of how a learner would distinguish raising verbs from control verbs, given their partial overlap in distribution. A series of experiments with English-speaking adults using a fill-in-the-blank questionnaire revealed two main types of cues that led participants to distinguish the two verb classes: inanimate NPs and semantically empty subjects (its raining) yielded the highest proportion of raising verb responses from adults, while animate NPs paired with eventive predicates yielded a high rate of control verb responses. On the basis of these results, suggestions are made as to how one should study the learning of these verbs in children.This research was supported by an NSF Postdoctoral Fellowship at the Institute for Research in Cognitive Science. I owe many, many thanks to Lila and Henry Gleitman, John Trueswell and the whole cast of participants in the Cheese seminar at IRCS during 2000–2002. In addition I would like to thank Ash Asudeh, Robin Clark, Jeffrey Lidz, Elliott Moreton, Julien Musolino, Anna Papafragou, Carson Schutze and an anonymous reviewer for comments and suggestions. I also thank audiences at University of North Carolina, University of Delaware, UCLA, Penn Linguistics Colloquium 26, Chicago Linguistic Society 38 and the CUNY Graduate Center for stimulating discussions. All errors and shortcomings are my own.     

Effects of age, speed of processing, and working memory on comprehension of sentences with relative clauses

Two hundred participants, 50 in each of four age ranges (19-29, 30-49, 50-69, 70-90) were tested for working memory, speed of processing, and the processing of sentences with relative clauses. In Experiment 1, participants read four sentence types (cleft subject, cleft object, subject-subject, subject-object) in a word-by-word, non-cumulative, self-paced reading task and made speeded plausibility judgments about them. In Experiment 2, participants read two types of sentences, one of which contained a doubly center embedded relative clause. Older participants' comprehension was less accurate and there was age-related slowing of online processing times in all but the simplest sentences, which increased in syntactically complex sentences in Experiment 1. This pattern suggests an age-related decrease in the efficiency of parsing and interpretation. Slower speed of processing and lower working memory were associated with longer online processing times only in Experiment 2, suggesting that task-related operations are related to general speed of processing and working memory. Lower working memory was not associated with longer reading times in more complex sentences, consistent with the view that general working memory is not critically involved in online syntactic processing. Longer online processing at the most demanding point in the most demanding sentence was associated with better comprehension, indicating that it reflects effective processing under some certain circumstances. However, the poorer comprehension performance of older individuals indicates that their slower online processing reflects inefficient processing even at these points.     

Perspective-taking and comprehension of passive sentences by Japanese-speaking children

This study demonstrates that children's difficulties in the interpretation of passives are attributed to their perspective-taking ability. Thirty-six Japanese preschool children participated in act-out sentence comprehension tasks. They were asked to manipulate two toy animals to demonstrate the meaning of two types of stimulus sentences: Type I had the child's toy, whose reference involved the child's actual name (e.g., Jun-kun no neko Jun's cat) encoded as grammatical subject, while Type II had the child's toy encoded as non-subject. Since passive structures take the perspective of the patient-denoting subject NP, it is assumed that only Type I passives have the perspective that matches that of the child.The results show that children's performance on passives was significantly better in Type I than in Type II sentences. But this difference was not observed for active sentences. For those who showed (nearly) perfect performance on active sentences, only Type I passives were equally well understood. These results strongly suggest that perspective-taking difficulties mask children's true competence on passives and that even 6-year-olds may not yet have attained the full perspective-taking ability required for comprehension of passive sentences.     

Influence of oral language and phonological awareness on children's bilingual reading

The purpose of this study was to assess the roles of oral language and phonological awareness on reading performance in grade 3 bilingual students. Several hierarchical models assessed the best predictors of third grade English and Spanish word attack, word identification and reading comprehension. Predictor variables were measures of phonological awareness, expressive vocabulary, receptive vocabulary, and syntax in both English and Spanish. The results showed that within language contributions of expressive vocabulary and syntax best predicted literacy when compared to phonological awareness measures.     

The effect of noun animacy on the processing of unambiguous sentences: evidence from French relative clauses

Two experiments, one using self-paced reading and one using eye tracking, investigated the influence of noun animacy on the processing of subject relative (SR) clauses, object relative (OR) clauses, and object relative clauses with stylistic inversion (OR-SI) in French. Each sentence type was presented in two versions: either with an animate relative clause (RC) subject and an inanimate object (AS/IO), or with an inanimate RC subject and an animate object (IS/AO). There was an interaction between the RC structure and noun animacy. The advantage of SR sentences over OR and OR-SI sentences disappeared in AS/IO sentences. The interaction between animacy and structure occurred in self-paced reading times and in total fixation times on the RCs, but not in first-pass reading times. The results are consistent with a late interaction between animacy and structural processing during parsing and provide data relevant to several models of parsing.     

Language-specific and universal influences in children's syntactic packaging of Manner and Path: a comparison of English, Japanese, and Turkish

Different languages map semantic elements of spatial relations onto different lexical and syntactic units. These crosslinguistic differences raise important questions for language development in terms of how this variation is learned by children. We investigated how Turkish-, English-, and Japanese-speaking children (mean age 3;8) package the semantic elements of Manner and Path onto syntactic units when both the Manner and the Path of the moving Figure occur simultaneously and are salient in the event depicted. Both universal and language-specific patterns were evident in our data. Children used the semantic-syntactic mappings preferred by adult speakers of their own languages, and even expressed subtle syntactic differences that encode different relations between Manner and Path in the same way as their adult counterparts (i.e., Manner causing vs. incidental to Path). However, not all types of semantics-syntax mappings were easy for children to learn (e.g., expressing Manner and Path elements in two verbal clauses). In such cases, Turkish- and Japanese-speaking children frequently used syntactic patterns that were not typical in the target language but were similar to patterns used by English-speaking children, suggesting some universal influence. Thus, both language-specific and universal tendencies guide the development of complex spatial expressions.     

The acquisition of allophonic rules: statistical learning with linguistic constraints

Phonological rules relate surface phonetic word forms to abstract underlying forms that are stored in the lexicon. Infants must thus acquire these rules in order to infer the abstract representation of words. We implement a statistical learning algorithm for the acquisition of one type of rule, namely allophony, which introduces context-sensitive phonetic variants of phonemes. This algorithm is based on the observation that different realizations of a single phoneme typically do not appear in the same contexts (ideally, they have complementary distributions). In particular, it measures the discrepancies in context probabilities for each pair of phonetic segments. In Experiment 1, we test the algorithm's performances on a pseudo-language and show that it is robust to statistical noise due to sampling and coding errors, and to non-systematic rule application. In Experiment 2, we show that a natural corpus of semiphonetically transcribed child-directed speech in French presents a very large number of near-complementary distributions that do not correspond to existing allophonic rules. These spurious allophonic rules can be eliminated by a linguistically motivated filtering mechanism based on a phonetic representation of segments. We discuss the role of a priori linguistic knowledge in the statistical learning of phonology.     

Meaning through syntax is insufficient to explain comprehension of sentences with reduced relative clauses: comment on McKoon and Ratcliff (2003)

The authors argue that the meaning through syntax (MTS) model proposed by G. McKoon and R. Ratcliff fails to account for the comprehension of sentences with reduced relative clauses. First, the theory's core assumptions regarding verb-based event representations and how they link to constructions are incompatible with well-established analyses from the lexical semantics literature. Second, the MTS theory provides neither a principled nor a consistent account for why some reduced relatives are hard whereas others are easy. Finally, McKoon and Ratcliff's critique of constraint-based models is flawed in that sometimes they tested a nonexistent theory and sometimes they provided evidence for the constraint-based models against which they were arguing.     

Verb movement in acquisition and aphasia: same problem, different solutions-evidence from Dutch

This paper focuses on verb movement in agrammatism and child language. We present data from a sentence completion experiment with 6 Dutch agrammatic aphasics and 21 Dutch-speaking children. The experiment compares completion of matrix clauses (which require verb movement) and embedded clauses (where such movement is not required) in these two populations. The results reveal a clear asymmetry: Both agrammatics and children do very well with embedded clauses but fail in 50% with the matrix clauses. It is concluded that the problem which both populations are facing is one of verb movement rather than verb inflection. An error analysis of the responses reveals that, although both agrammatics and children try to avoid movement, they apply different strategies to achieve this goal.     

On strategies for processing relative clauses: A comparison of children and adults

Adults were tested for the way in which they process four types of subject and object relative clauses. The results support an anti-interruption and anti-rearrangement constraint that has been proposed by Slobin. The reason why interruption and rearrangment of linguistic units is hard for adults is explained in terms of language-processing strategies that they are hypothesized to be using, in particular the Adjacency strategy. Adult behavior is compared to the performance of 4- and 5-year-old children in a previous study. The results of these two studies support the claim that children and adults are following the same strategies in processing these sentences, and that the difference between them is in which strategies they rely most heavily on, and, consequently, which sentences they make the most mistakes on.This research was supported by a Grant-in-Aid from the Graduate School of the University of Minnesota.This paper was presented at the Thirty-sixth Summer Meeting of the Linguistic Society of American, on July 29, 1974, in Amherst, Massachusetts.     

Controversies about CP: a comparison of language acquisition and language impairments in Broca's aphasia

In both language acquisition research and the study of language impairments in Broca's aphasia there is an ongoing debate whether or not phrase-structure representations contain the Complementizer Phrase (CP) layer. To shed some light on this debate, I will provide data on German child language and on German agrammatic Broca's aphasia. Analyses of subordinate clauses, wh-questions, and verb placement indicate that early child grammars do not generate the CP layer yet, whereas the ability to project the CP layer is retained in agrammatism.