Brain imaging and language processing--an introduction to special issue papers.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to A Friederici.
Explore the source record for details and available documents.
The magnetic correlate of the P300 wave was registered using a seven-channel detector. The spatial variation of the corresponding field suggests the contribution of outer brain layers to the neural activity of M300.
Several recent studies characterize agrammatic comprehension as a computational rather than a conceptual deficit. A restriction in the computational capacity of the processor may affect sentence comprehension performance unevenly due to the varying demands imposed by different sentence structures and different tasks. It is argued that structures involving longer inferential chains suffer greater degradation than structures involving shorter inferential chains. Many of the observed properties of agrammatic comprehension may be seen to follow from this principle alone.
Broca's aphasics and normal controls were tested to determine relative sparing and impairment of word order, grammatical morphology, and semantic information in a sentence interpretation task. Patients were native speakers of English, German, or Italian, languages that vary drastically in the "cue validity" or information value of these three sources of information. Word order was selectively spared while grammatical morphology was selectively impaired in all three languages. Nevertheless, language-specific patterns of sentence interpretation remained in Broca's aphasics, even within the impaired morphological component, supporting an interpretation in terms of "accessing" rather than a "loss." Testing with Wernicke's aphasics, anomics, and some additional age-matched controls suggested that the selective vulnerability of morphology is not specific to agrammatic patients, at least in this paradigm.
Aspects of grammatical morphology in Broca's and Wernicke's aphasia were elicited under controlled conditions in three language groups: English, Italian and German. Results suggest that the agrammatism/paragrammatism distinction does not work well for richly-inflected languages. Language-specific ratios of closed class morphology were preserved even among non-fluent patients, with significantly more morphology produced by German and Italian patients. German and Italian patients were also much more likely to furnish the article before nouns--despite or perhaps because of the fact that articles are more complex and informative in those languages. Although patients assigned the correct article most of the time, there were a significant number of article errors (i.e. paragrammatic substitution). Error analyses showed that substitutions are not random, reflecting difficulty in access rather than loss. Substitutions were more common in German, where the complex case and gender markings on the article increase the probability of error. Within each language, error patterns were quite similar for Broca's and Wernicke's aphasics. However, at a detailed level patient group differences in error production were detected. German Broca's aphasics tend to avoid difficult case forms by substituting a simpler, less-marked morphosyntactic frame. Wernicke's aphasics try instead to produce the more marked, oblique constructions, resulting in a less conservative error pattern.
Previous research suggests that English Broca's and Wernicke's aphasics retain sensitivity to pragmatic factors governing forms of reference, in particular the ability to choose lexical expressions that convey givenness and newness of information. The present study investigates the generality of this phenomenon across patients and language types. Normal and aphasic speakers of English, German, and Italian described nine picture triplets in which one element varied while the others remained constant. Dependent variables included lexicalization versus ellipsis, pronominalization, and definite and indefinite article use. For a subset of German and Italian patients, data were compared to performance in a biographical interview. Results indicate that (a) the pragmatics of reference are preserved in both Broca's and Wernicke's aphasics, despite syndrome-specific problems in retrieving content words and/or closed-class grammatical elements, and (b) certain language-specific patterns of reference are also preserved, including crosslinguistic differences in subject omission. Differences between picture description and the biographical interview reinforce this conclusion. Evidence for the preservation of pragmatics in aphasis is not surprising in its own right, but evidence for the sparing of language-specific relations among pragmatic, lexical, and morphosyntactic patterns can be used to argue against any strong view of grammatical impairment as a disconnection syndrome and/or a loss of grammatical competence. Instead, these data support theories in which grammatical impairment is viewed as a performance deficit.