Morphosyntactic Information Research Articles

Shared representations in language processing and verbal short-term memory: The case of grammatical gender

The general idea of language-based accounts of short-term memory is that retention of linguistic materials is based on representations within the language processing system. In the present sentence recall study, we address the question whether the assumption of shared representations holds for morphosyntactic information (here: grammatical gender in German). Three experiments using a modified version of Potter and Lombardi’s (1990) intrusion paradigm tested whether the presentation of nouns similar in meaning to a noun included in a sentence led to more intrusions in sentence recall if the related nouns had the same grammatical gender as compared to a different gender. Indeed, morphosyntactic constraints resulted in weaker (or even no) intrusion effects for gender incongruent as compared to gender congruent intruders. This effect was observed independently of the degree to which phonological information was available during recall and independently of a determiner mismatch between the noun in the sentence and the incongruent intruder.

Corrélats neurophysiologiques du traitement morphosyntaxique

The study investigated cortical responses of young native French speaking schoolchildren during reading of short sentences including morphosyntactic violations in the verb inflection. In French, verbs of the first conjugation have homophonic inflection for the infinitive and past participle. This peculiarity is one of the main source of morphosyntactic errors in writing learners. To look for possible dissociation in the cortical responses to morphosyntactic violations, we compared in 15 schoolchildren (mean age: 10;1 years) the electrophysiological responses to morphological violations of verb inflection in verb with homophonic or heterophonic infinitive/past participle inflections. Our main findings concern the early left anterior negativity (ELAN), which is assumed to reflect the first-pass parsing and the left anterior negativity (LAN) corresponding to the integration of semantic and morphosyntactic information. These ERP’s components appeared both sensitive to the violation in the inflection of verbs with heterophonic but not homophonic infinitive and past participle forms.

Morphological processing in early bilinguals: An ERP study of regular and irregular verb processing

Although the age of acquisition of a language has an effect when learning a second language, the similarity between languages may also have a crucial role. The aim of the present study is to understand the influence of this latter factor in the acquisition of morphosyntactic information. With this purpose, two groups of highly proficient early Catalan–Spanish bilinguals were presented with a repetition-priming paradigm with regular and irregular verbs of Spanish. Catalan and Spanish have a similar suffix ( -o) for regular verbs and completely different alternations for irregular verbs. Two types of irregular verbs were studied (semi-regular verbs with a systematic diphthong alternation, sentir–siento, and verbs with idiosyncratic changes, venir–vengo). Regular verbs showed the same centro-parietal N400 priming effect in the second-language speakers (L2) as in primary-language (L1) speakers. However, differences between groups, in the ERP pattern and the topography of the N400 effect, were observed for irregular morphology. In L1 speakers, the N400 effect was attenuated only for semi-regular verbs. In contrast, L2 speakers showed a reduced N400 priming effect in both irregular contrasts. This pattern of results suggests that the similarity between languages may help for similar structures but may interfere for dissimilar structures, at least when the two languages have very similar morphological systems.

When case meets agreement: event-related potential effects for morphology-based conflict resolution in human language comprehension

We present event-related potential evidence from language comprehension that processing conflicts arising from the same linguistic domain and appearing within the same time range do not interact when they draw upon distinct underlying neural populations. Thus, a combined violation of two morphosyntactic information types, number-agreement and case, engendered a LAN/N400-P600 pattern, while the corresponding single violations are associated with LAN-P600 and N400-P600 responses, respectively. The absence of an interaction between the two negativities indicates that neuronal resource sharing does not result from a similarity of function, but rather requires an overlap of the underlying neuronal populations.

Towards a semantic lexicon for biological language processing

This paper explores the use of the resources in the National Library of Medicine's Unified Medical Language System (UMLS) for the construction of a lexicon useful for processing texts in the field of molecular biology. A lexicon is constructed from overlapping terms in the UMLS SPECIALIST lexicon and the UMLS Metathesaurus to obtain both morphosyntactic and semantic information for terms, and the coverage of a domain corpus is assessed. Over 77% of tokens in the domain corpus are found in the constructed lexicon, validating the lexicon's coverage of the most frequent terms in the domain and indicating that the constructed lexicon is potentially an important resource for biological text processing.

Morphosyntactic and syntactic priming: an investigation of underlying processing mechanisms and the effects of Parkinson's disease

There is now considerable evidence to suggest that non-demented people with Parkinson's disease (PD) experience difficulties using the morphosyntactic aspects of language. It remains unclear, however, at precisely which point in the processing of morphosyntax, these difficulties emerge. The major objective of the present study was to examine the impact of PD on the processes involved in accessing morphosyntactic information in the lexicon. Nineteen people with PD and 19 matched control subjects participated in the study which employed on-line word recognition tasks to examine morphosyntactic priming for local grammatical dependencies that occur both within (e.g. is going) and across (e.g. she gives) phrasal boundaries (Experiments 1 and 2, respectively). The control group evidenced robust morphosyntactic priming effects that were consistent with the involvement of both pre- (Experiment 1) and post-lexical (Experiment 2) processing routines. Whilst the participants with PD also recorded priming for dependencies within phrasal boundaries (Experiment 1), priming effects were observed over an abnormally brief time course. Further, in contrast to the controls, the PD group failed to record morphosyntactic priming for constructions that crossed phrasal boundaries (Experiment 2). The results demonstrate that attentionally mediated mechanisms operating at both the pre- and post-lexical stages of processing are able to contribute to morphosyntactic priming effects. In addition, the findings support the notion that, whilst people with PD are able to access morphosyntactic information in a normal manner, the time frame in which this information remains available for processing is altered. Deficits may also be experienced at the post-lexical integrational stage of processing.

Automatic acquisition of lexical and morpho-syntactic information from raw corpora: application to Serbo-Croat and Russian

En aquest article presentem una metodologia per a l'adquisicio automatica d'informacio lexica i morfosintactica a partir de corpus sense anotar. El sistema fa servir informacio sobre la morfologia flexiva de la llengua que es vol tractar, i tambe informacio lexica i morfosintactica de les paraules pertanyents a classes no flexives i de les paraules que tenen una flexio que no respon a paradigmes regulars. Es un sistema en fase de desenvolupament, per la qual cosa les avaluacions que hi incloem son preliminars.

How Do French Children Use Morphosyntactic Information When They Spell Adverbsand Present Participles?

This study examined how French third (36) and fifth (36) graders used the morphosyntactic context when they spell morphologically complex words with homophonous suffixes (/ã/). Participants had to spell adverbs (/ã/ transcribed ent) and present participles (/ã/ transcribed ant), contrasted on the basis of their frequency, in isolation or embedded within a sentence. Spelling performances were better when the words were dictated in the context of a sentence than in a condition of isolated word dictation. This was the case even in Grade 3, despite the fact that explicit teaching about the transcription of the suffixes /ã/ had been provided only to the fifth graders. The frequency of target words was an important determinant of spelling success even when a sentence context was provided. Thus, even though children benefited from the morphosyntactic information provided by the sentence, they did not rely (systematically) on a morphological rule such as "words ending with /ã/ are always given the ant spelling after the preposition en (e.g., en regardant 'in looking')."

Functional MR imaging exposes differential brain responses to syntax and prosody during auditory sentence comprehension

In two experiments using event-related functional magnetic resonance imaging we studied healthy adults who listened to sentences that either focused on lexical, syntactic, or prosodic information. In the first experiment two sentence conditions were employed: normal speech which contained function and content words, and pseudo speech which contained function and pseudo words. Sentence processing generally activated the superior temporal region (STR) bilaterally. Relative to normal sentences hearing pseudo sentences corresponded to stronger brain responses in the anterior STR (planum polare) and in the fronto-opercular region bilaterally. A second experiment was designed to test whether right hemisphere activation can be explained by processing prosodic aspects of speech, i.e. sentence intonation. In addition to normal and pseudo speech the second study examined degraded speech that neither contained morphosyntactic nor lexical information but only prosodic information, i.e. intonation, amplitude, duration, and spectral tilt. Statistical analyses based on regions of interest found differential activation patterns for frontal and temporal areas in the brain. Relative to sentences, degraded speech produced generally stronger activation in frontal regions. Furthermore, the data pointed to a particular involvement of right fronto-lateral regions in processing sentence melody. For the STR an inverse pattern was found: relative to degraded speech sentence conditions produced stronger activation in anterior, mid, and posterior parts of the left supratemporal plane, particular in the planum polare and planum temporale. In sum, the results show that the left planum polare and the left planum temporale mediate syntactic and semantic processing, whereas right fronto-lateral areas seem to be more sensitive to prosodic cues available in spoken language.

Distinct patterns of neural modulation during the processing of conceptual and syntactic anomalies.

The aim of this study was to gain further insights into how the brain distinguishes between meaning and syntax during language comprehension. Participants read and made plausibility judgments on sentences that were plausible, morphosyntactically anomalous, or pragmatically anomalous. In an event-related potential (ERP) experiment, morphosyntactic and pragmatic violations elicited significant P600 and N400 effects, respectively, replicating previous ERP studies that have established qualitative differences in processing conceptually and syntactic anomalies. Our main focus was a functional magnetic resonance imaging (fMRI) study in which the same subjects read the same sentences presented in the same pseudorandomized sequence while performing the same task as in the ERP experiment. Rapid-presentation event-related fMRI methods allowed us to estimate the hemodynamic response at successive temporal windows as the sentences unfolded word by word, without assumptions about the shape of the underlying response function. Relative to nonviolated sentences, the pragmatic anomalies were associated with an increased hemodynamic response in left temporal and inferior frontal regions and a decreased response in the right medial parietal cortex. Relative to nonviolated sentences, the morphosyntactic anomalies were associated with an increased response in bilateral medial and lateral parietal regions and a decreased response in left temporal and inferior frontal regions. Thus, overlapping neural networks were modulated in opposite directions to the two types of anomaly. These fMRI findings document both qualitative and quantitative differences in how the brain distinguishes between these two types of anomalies. This suggests that morphosyntactic and pragmatic information can be processed in different ways but by the same neural systems.

Semantic and morpho-syntactic priming in auditory word recognition in congenitally blind adults

While several studies have reported a deviation from the normal time course of language acquisition in blind children others have provided evidence for a more efficient processing of the language input in blind than sighted adults. The present study used a semantic and morpho-syntactic priming paradigm to address the question at which processing stage the advantage of blind adults may arise. Congenitally blind adults and sighted controls matched for age, gender and education, first heard an adjective followed by a noun or a pseudo-word. The adjective was or was not semantically associated with the target and it was either correctly or incorrectly inflected for gender with respect to the following noun. Participants decided whether or not the target noun was a legal German word. Nouns primed semantically and morpho-syntactically had shorter lexical decision times than those primed only semantically or only morpho-syntactically and decision times for the latter two conditions were shorter than in a condition without a semantically or morpho-syntactically congruent context. This response pattern did not differ between groups. However, blind participants had shorter reaction times than sighted for pseudo-words, and overall decision times for words tended to be shorter in the blind as well. It is concluded that the faster speech comprehension skills of blind adults most likely originate from a more efficient perceptual analysis rather than from a more extended use of semantic or morpho-syntactic context information.

A Rule-Based Analysis From Raw Korean Text to Morphologically Annotated Corpora

Morphologically annotated corpora are the basis for many tasks of computational linguistics. Most current approaches use statistically driven methods of morphological analysis, that provide just POS-tags. While this is sufficient for some applications, a rule-based full morphological analysis also yielding lemmatization and segmentation is needed for many others. This work thus aims at [1] introducing a rule-based Korean morphological analyzer called Kormoran based on the principle of linearity that prohibits any combination of left-to-right or right-to-left analysis or backtracking and then at [2] showing how it on be used as a POS-tagger by adopting an ordinary technique of preprocessing and also by filtering out irrelevant morpho-syntactic information in analyzed feature structures. It is shown that, besides providing a basis for subsequent syntactic or semantic processing, full morphological analyzers like Kormoran have the greater power of resolving ambiguities than simple POS-taggers. The focus of our present analysis is on Korean text. (Korea University and Universitat Erlangen-Nurnberg)

Fast mapping of verbs by children with specific language impairment

Two studies examined preschoolers' ability to assign verb interpretations to nonsense words encountered in conjunction with novel actions. Experiment 1 examined the ability of children with specific language impairment (SLI) and younger, normally developing peers to glean a verb interpretation when the name of the figure performing the novel action was already known. The two groups of children performed in a similar, accurate fashion. Experiment 2 required pre-schoolers to rely exclusively on morphosyntactic information to determine whether the novel word represented an object or action. When provided with redundant morphosyntactic cues, children with SLI and language- and age-matched peers succeeded in identifying the novel words that referred to objects but not those that referred to actions. Only the age-matched normal peers were above chance levels when a noun interpretation depended on a single grammatical morpheme (e.g., 'We want the koob ' versus 'We want to koob '). The findings suggest that preschoolers, whether or not they have language impairment, have difficulty using morphosyntactic information to bootstrap verbs. Furthermore, redundant but not single morphosyntactic cues facilitate the bootstrapping of nouns.

The Klagenfurt database for sign language lexicons

The Klagenfurt database was originally developed as a base for both monolingual and bilingual dictionaries of Austrian Sign Language. The part dealing with the formational features of signs as well as the morphosyntactic and semantic information have already been finished. We decided on an approach using modified versions of Liddell & Johnson’s (1989) Movement-and-Hold model, the HamNoSys (Prillwitz, S. et al. 1989) and the SignPhon category system (Blees, M. et al. 1996).

The Klagenfurt database for sign language lexicons

The Klagenfurt database was originally developed as a base for both monolingual and bilingual dictionaries of Austrian Sign Language. The part dealing with the formational features of signs as well as the morphosyntactic and semantic information have already been finished. We decided on an approach using modified versions of Liddell & Johnson’s (1989) Movement-and-Hold model, the HamNoSys (Prillwitz, S. et al. 1989) and the SignPhon category system (Blees, M. et al. 1996).

Influence of prosodic boundaries on comprehension of spoken English sentences.

Three experiments investigated the role of prosody in the comprehension of auditory sentences. In Exp. 1 an analysis of three novice talkers and one expert talker verified the production parameters of one type of syntactic ambiguity and showed that pitch cues were more prominent than duration cues. In Exp. 2, 16 listeners used prosodic information to make consistent decisions reliably about phrase boundaries. In Exp. 3, 40 participants listened to sentences in which prosody was inconsistent with later morphosyntactic information, indicated their understanding, and then judged whether a visual target was related to the meaning of the sentence. Inconsistent prosody slowed comprehension and contributed to slower, less accurate judgments of sentence meaning. This suggests that prosodic information contributes to the perception of spoken language and can affect comprehension even when the syntactic structure indicated by prosody is contradicted by subsequent morphosyntactic information.

Comparison of pitch reset characteristics between children’s filler syllable and starter syllable productions

Children’s filler syllables may assist in maintaining rhythmic envelopes and intonational phrasing, and may be markers of potential or actual morphosyntactic information. Therefore, they should exhibit pitch characteristics indicating coherence with other constituents in the same intonational phrase. They contrast with starter syllables, which serve more pragmatic functions such as discourse place holders. An acoustical analysis of pitch reset is presented which distinguishes between filler and starter productions. Pitch reset was evaluated through two measures: the difference in intensity (decibels) between syllable initiation and syllable termination, and boundary duration (milliseconds) from the offset of the targeted syllable to the onset of the next syllable or constituent. Results demonstrated that fillers and starters could be differentiated using pitch reset measures. Typically, fillers did not exhibit pitch reset, described here as a boundary duration less than 200 ms and intensity difference scores ranging from −15 to +17 dB. Starters, however, did typically exhibit pitch reset, described here as a boundary duration greater than 200 ms and intensity difference scores ranging from 0 to +17 dB. This pilot investigation suggests that fillers exhibit similar pitch reset acoustical characteristics to other syntactic constituents in the intonational phrase in which they are located. [Work supported by NIH.]

Automatic rule induction for unknown-word guessing

Words unknown to the lexicon present a substantial problem to NLP modules that rely on morphosyntactic information, such as part-of-speech taggers or syntactic parsers. In this paper we present a t...

La compréhension asyntaxique dans tous ses états. Des représentations linguistiques aux ressources cognitives

Summary: What to do with the various aspects of asyntactic comprehension ? From linguistic representations to cognitive resources. Asyntactic comprehension is a sentence comprehension disorder presented by some brain damaged patients. This trouble is characterized by a selective processing deficit of grammatical information. This review paper discusses the relations between three types of theories explaining asyntactic comprehension. Structural theories define the elements of linguistic representations that are not established in asyntactic comprehenders : these theories suggest a very specific deficit involving some syntactic elements (traces). Procedural theories define the psycholinguistic processes that are deficient in the patients : asyntactic comprehenders should have a deficit of the automatic processing of morphosyntactic information. Resource theories explain the comprehension deficit by a limitation of working memory resources that could result in a slow-down of activation and/or a speed-up of the decline of linguistic information : these theories are appropriate to explain the rapid decline of syntactic information in asyntactic comprehenders. These three theories are not incompatible : structural theories define the linguistic elements that require a rapid processing, and that are especially costly in working memory resources if this rapid processing is deficient. However, the results in favor of these theories should be taken with caution : for experiments, asyntactic comprehenders are often selected according to their expressive disorder (agrammatism). Some case studies suggest a dissociation between agrammatism and asyntactic comprehension. Key words : asyntactic comprehension, syntactic traces, cognitive resources.

The nondeviant nature of deviant phonological systems

Few studies of children with defective speech or deviant phonologies have gone beyond the level of phonetic representation in limited environments. The Lorentz (1976) study of Joe, a 4.5-yr-old child with a phonological disorder, is exceptional inasmuch as underlying phonemic representations and phonological rules are posited in addition to the surface phonetic representations. However, his analysis crucially depends on morphosyntactic information that cannot be independently justified. A reanalysis of Joe's data within Kahn's (1976) syllable-based model of generative phonology reveals that the child's phonological rules are syllabically conditioned exclusively. In particular, the application of certain rules is found to depend on whether or not an internal syllable boundary is present within a polysyllabic word. In the course of the reanalysis, issues concerning naturalness of individual phonological rules, ordering relations between pairs of rules, feature systems, the phonological representation of double-articulated consonants, and the role of the syllable as an element in phonological descriptions are discussed. Paradoxically, Joe's phonological system is found to be “normal,” as measured in terms of the above-mentioned linguistic parameters, yet his speech output is “abnormal,” as measured against the adult norm.