Auditory semantic processing in dichotic listening: effects of competing speech, ear of presentation, and sentential bias on N400s to spoken words in context.

The effects of ear of presentation and competing speech on N400s to spoken words in context were examined in a dichotic sentence priming paradigm. Auditory sentence contexts with a strong or weak semantic bias were presented in isolation to the right or left ear, or with a competing signal presented in the other ear at a SNR of -12 dB. Target words were congruent or incongruent with the sentence meaning. Competing speech attenuated N400s to both congruent and incongruent targets, suggesting that the demand imposed by a competing signal disrupts the engagement of semantic comprehension processes. Bias strength affected N400 amplitudes differentially depending upon ear of presentation: weak contexts presented to the le/RH produced a more negative N400 response to targets than strong contexts, whereas no significant effect of bias strength was observed for sentences presented to the re/LH. The results are consistent with a model of semantic processing in which the RH relies on integrative processing strategies in the interpretation of sentence-level meaning.

Lexico-semantic processing in children with specific language impairment: the overactivation hypothesis.

UNLABELLED: The hypothesis indicating an overactivation of the lexico-semantic network in children with specific language impairment (SLI) was tested using an auditory pair-primed paradigm (PPP), where participants made a lexical-decision on the second word of a noun pair that could be semantically related, or not, to the first one. Though children with SLI were proven to be as accurate as children matched for receptive vocabulary age, they presented a larger priming effect in the PPP in terms of both reaction time and accuracy. These results preliminarily support the hypothesis of an overactivation of the lexico-semantic network. LEARNING OUTCOMES: As a result of reading this paper the reader will be able to (1) understand how the pair-primed paradigm can contribute to investigate the online spreading of the activation within the lexico-semantic network; (2) be aware of the fact that during development it is unlikely that some cognitive domain are completely typical (residual normality), while others develop atypically; (3) the reader will be aware that children with SLI present some subtle abnormalities in the lexico-semantic network, which appears to be overactive.

Children with specific language impairment: The effect of argument-structure complexity on auditory sentence comprehension.

Children with specific language impairment (SLI) demonstrate consistent comprehension problems. The present study investigated whether these problems are driven primarily by structural complexity or length. A picture-sentence matching task was presented to 30 children: (1) 10 children with SLI, (2) 10 comprehension-matched children with typical language development (TLD) and (3) 10 children with TLD matched for chronological age. Argument-structure complexity was manipulated independently of length, which was also independently varied. Results showed that argument-structure complexity had a greater influence on comprehension in children with SLI than in the comparison groups, with transitive sentences eliciting more errors than intransitive ones. This effect was not dependent on sentence length, which did not appear to affect the comprehension level. The results support the view that comprehension problems in children with SLI are principally related to the structural complexity of the sentence rather than the amount of material to be processed.

Speaking of events: the case of C.M.

C.M. is an agrammatic patient who on assessment tests shows a disproportionate difficulty when producing verbs compared with nouns. In three experiments, we investigated whether C.M. also has difficulties with nouns referring to events and whether event nouns and verbs show similar patterns of disruption. Experiment 1 suggested that she is sensitive to argument structure complexity and has a greater impairment in the production of event nouns and verbs than object nouns. Experiment 2 revealed that C.M. finds derivationally complex words, such as event nouns, difficult to produce. However, morphological complexity does not completely explain C.M.'s problems with event nouns. In Experiment 3, an assessment of C.M.'s ability to use different aspects of semantic and syntactic knowledge relative to event nouns and verbs showed an almost identical performance with the two types of words. The relevance of the findings with respect to models of word production is considered.

The argument-structure complexity effect in children with specific language impairment: evidence from the use of grammatical morphemes in French.

PURPOSE: The hypothesis that the linguistic deficit presented by children with specific language impairment (SLI) is caused by limited cognitive resources (e.g., S. Ellis Weismer & L. Hesketh, 1996) was tested against the hypothesis of a limitation in linguistic knowledge (e.g., M. L. Rice, K. Wexler, & P. Cleave, 1995). METHOD: The study examined the influence of the argument-structure complexity of a target sentence on the production of grammatical morphemes in French children with SLI compared with younger children matched for grammatical level in production (GL) and children of the same chronological age (CA). A sentence production task was used where the target sentences varied in terms of argument complexity and length. RESULTS: The results indicated that children with SLI used articles and auxiliaries in obligatory contexts significantly less often than both the GL and CA control groups: More complex argument structures elicited the highest number of grammatical morpheme omissions; this effect was larger in children with SLI than in the GL group and was independent of the length of the sentences, which failed to show any influence on the production of grammatical morphemes. CONCLUSION: These results support the hypothesis that grammatical-morpheme deficit in SLI depends at least in part on limited processing capacities.