Activating Brain Systems for Syntax and Semantics

ized sentences and the simpler baseline condition contained subject-relativized sentences; the exact sentence types were slightly different than those used in our studies. Subjects then read a second sentence and indicated whether that sentence described the meaning Human languages consist of multiple types of elements: of the first sentence (e.g., statement: " the general that phonemes, words, syntactic structures, intonational the politician introduced praised the reporter " ; test: " the contours, etc. Are particular regions of the brain special-general introduced the politician "). Just et al. found acti-ized for representing and/or processing these different vation in Wernicke's area (BA 22) as well as Broca's elements? Dapretto and Bookheimer (1999 [this issue area and in the right hemisphere homologs of these of Neuron]) have addressed this question in the domains regions. Similar discrepancies have been found in stud-of word meaning (lexical semantic representations) and ies of lexical semantic processing (Peterson et al., 1988; sentence structure (syntax), using functional magnetic Demonet et al., 1992; Vandenberghe et al., 1996). resonance imaging (fMRI). Why do these differences arise, and what can be done In Dapretto and Bookheimer's study, subjects made to resolve them? judgments about the synonymity of two sentences. Two Different results could be due to technical factors: conditions were presented. In the first condition, the different sensitivities of positron emission tomography sentences had the same form but had one change in (PET) and MR techniques, inaccuracies inherent in nor-vocabulary. Subjects were to say that the sentences malization of individual brains affecting discernible acti-were the " same " if the different words were synonyms vation in a region of interest, different statistical tech-(e.g., " the lawyer questioned the witness, " " the attorney niques in use in different laboratories. In principle, these questioned the witness ") and that they were " different " differences can be resolved by reanalysis of data, efforts if they were not (e.g., " the man was attacked by the to replicate experiments using different technology and doberman, " " the man was attacked by the pitbull "). In analyses, etc. But this seems unlikely to be the whole the second condition, the words in the sentences re-story. The differences between the results of Dapretto mained the same but the syntactic structure of the sen-and Bookheimer and those of Just et al., for instance, tence changed. Subjects were to say that the sentences do not seem likely to …