The information contained in a sensory signal plays a critical role in determining what neural processes are engaged. Here we used interleaved silent steady-state (ISSS) functional magnetic resonance imaging (fMRI) to explore how human listeners cope with different degrees of acoustic richness during auditory sentence comprehension. Twenty-six healthy young adults underwent scanning while hearing sentences that varied in acoustic richness (high vs. low spectral detail) and syntactic complexity (subject-relative vs. object-relative center-embedded clause structures). We manipulated acoustic richness by presenting the stimuli as unprocessed full-spectrum speech, or noise-vocoded with 24 channels. Importantly, although the vocoded sentences were spectrally impoverished, all sentences were highly intelligible. These manipulations allowed us to test how intelligible speech processing was affected by orthogonal linguistic and acoustic demands. Acoustically rich speech showed stronger activation than acoustically less-detailed speech in a bilateral temporoparietal network with more pronounced activity in the right hemisphere. By contrast, listening to sentences with greater syntactic complexity resulted in increased activation of a left-lateralized network including left posterior lateral temporal cortex, left inferior frontal gyrus, and left dorsolateral prefrontal cortex. Significant interactions between acoustic richness and syntactic complexity occurred in left supramarginal gyrus, right superior temporal gyrus, and right inferior frontal gyrus, indicating that the regions recruited for syntactic challenge differed as a function of acoustic properties of the speech. Our findings suggest that the neural systems involved in speech perception are finely tuned to the type of information available, and that reducing the richness of the acoustic signal dramatically alters the brain’s response to spoken language, even when intelligibility is high.

The scope of reference of a word's meaning can be highly variable. We present a novel paradigm to investigate the flexible interpretation of word meaning. We focus on quantifiers such as “many” or “few,” a class of words that depends on number knowledge but can be interpreted in a flexible manner. Healthy young adults performed a truth value judgment task on pictorial arrays of varying amounts of blue and yellow circles, deciding whether the sentence “Many/few of the circles are yellow” was an adequate description of the stimulus. The study consisted of two experiments, one focusing on “many,” one on “few.” Each experiment had three blocks. In a first “baseline” block, each individual's criterion for “many” and “few” was assessed. In a second “adaptation” block, subjects received feedback about their decisions that was different from their initial judgments in an effort to evaluate the flexibility of a subject's interpretation. A third “test” block assessed whether adaptation of quantifier meaning induced in block 2 then was generalized to alter a subject's baseline meaning for “many” and “few.” In Experiment 1, a proportion of yellow circles as small as 40% was reinforced as “many”; in Experiment 2, a proportion of yellow circles as large as 60% was reinforced as “few.” Subjects learned the new criterion for “many” in Experiment 1, which also affected their criterion for “few” although it had never been mentioned. Likewise, in Experiment 2, subjects changed their criterion for “few,” with a comparable effect on the criterion for “many” which was not mentioned. Thus, the meaning of relational quantifiers like “many” and “few” is flexible and can be adapted. Most importantly, adapting the criterion for one quantifier (e.g., “many”) also appeared to affect the reciprocal quantifier (in this case, “few”). Implications of this result for psychological interventions and for investigations of the neurobiology of the language-number interface are discussed.

Quantifiers such as many and some are thought to depend in part on the conceptual representation of number knowledge, while object nouns such as cookie and boy appear to depend in part on visual feature knowledge associated with object concepts. Further, number knowledge is associated with a frontal-parietal network while object knowledge is related in part to anterior and ventral portions of the temporal lobe. We examined the cognitive and anatomic basis for the spontaneous speech production of quantifiers and object nouns in non-aphasic patients with focal neurodegenerative disease associated with corticobasal syndrome (CBS, n=33), behavioral variant frontotemporal degeneration (bvFTD, n=54), and semantic variant primary progressive aphasia (svPPA, n=19). We recorded a semi-structured speech sample elicited from patients and healthy seniors (n=27) during description of the Cookie Theft scene. We observed a dissociation: CBS and bvFTD were significantly impaired in the production of quantifiers but not object nouns, while svPPA were significantly impaired in the production of object nouns but not quantifiers. MRI analysis revealed that quantifier production deficits in CBS and bvFTD were associated with disease in a frontal-parietal network important for number knowledge, while impaired production of object nouns in all patient groups was related to disease in inferior temporal regions important for representations of visual feature knowledge of objects. These findings imply that partially dissociable representations in semantic memory may underlie different segments of the lexicon.

Previous work has shown that the meaning of a quantifier such as “many” or “few” depends in part on quantity. However, the meaning of a quantifier may vary depending on the context, e.g. in the case of common entities such as “many ants” (perhaps several thousands) compared to endangered species such as “many pandas” (perhaps a dozen). In a recent study (Heim et al. 2015 Front. Psychol.) we demonstrated that the relative meaning of “many” and “few” may be changed experimentally. In a truth value judgment task, displays with 40% of circles in a named color initially had a low probability of being labeled “many”. After a training phase, the likelihood of acceptance 40% as “many” increased. Moreover, the semantic learning effect also generalized to the related quantifier “few” which had not been mentioned in the training phase. Thus, fewer 40% arrays were considered “few.” In the present study, we tested the hypothesis that this semantic adaptation effect was supported by cytoarchitectonic Brodmann area (BA) 45 in Broca’s region which may contribute to semantic evaluation in the context of language and quantification. In an event-related fMRI study, 17 healthy volunteers performed the same paradigm as in the previous behavioral study. We found a relative signal increase when comparing the critical, trained proportion to untrained proportions. This specific effect was found in left BA 45 for the trained quantifier “many”, and in left BA 44 for both quantifiers, reflecting the semantic adjustment for the untrained but related quantifier “few.” These findings demonstrate the neural basis for processing the flexible meaning of a quantifier, and illustrate the neuroanatomical structures that contribute to variable meanings that can be associated with a word when used in different contexts.