Chris DeVita

Age-Related Changes in Working Memory during Sentence Comprehension: An fMRI Study

Sentence comprehension declines with age, but the neural basis for this change is unclear. We monitored regional brain activity in 13 younger subjects and 11 healthy seniors matched for sentence comprehension accuracy while they answered a simple probe about written sentences. The sentences varied in their grammatical features (subject-relative vs object-relative subordinate clause) and their verbal working memory (WM) demands (short vs long antecedent noun-gap linkage). We found that young and senior subjects both recruit a core written sentence processing network, including left posterolateral temporal and bilateral occipital cortex for all sentences, and ventral portions of left inferior frontal cortex for object-relative sentences with a long noun-gap linkage. Differences in activation patterns for seniors compared to younger subjects were due largely to changes in brain regions associated with a verbal WM network. While seniors had less left parietal recruitment than younger subjects, left premotor cortex, and dorsal portions of left inferior frontal cortex showed greater activation in seniors compared to younger subjects. Younger subjects recruited right posterolateral temporal cortex for sentences with a long noun-gap linkage. Seniors additionally recruited right parietal cortex for this sentence-specific form of WM. Our findings are consistent with the hypothesis that the neural basis for sentence comprehension includes dissociable but interactive large-scale neural networks supporting core written sentence processes and related cognitive resources involved in WM. Seniors with good comprehension appear to up-regulate portions of the neural substrate for WM during sentence processing to achieve comprehension accuracy that equals young subjects.

The Neural Basis for Categorization in Semantic Memory

We asked young adults to categorize written object descriptions into one of two categories, based on a rule or on overall similarity, while we monitored regional brain activity with functional magnetic resonance imaging (fMRI). We found significantly greater recruitment of left dorsolateral prefrontal cortex for rule-based categorization in direct comparison with similarity-based categorization. Recruitment of right ventral frontal cortex and thalamus was uniquely associated with rule-based categorization as well. These observations lend support to the claim that executive functions such as working memory, inhibitory control, and selective attention contribute to rule-based categorization. Right inferior parietal activation was uniquely associated with similarity-based categorization. This region may play an important role in overall feature configuration that is important for this form of categorization. We found other brain regions recruited for both rule-based and similarity-based categorization: Anterior cingulate cortex may support the implementation of executive functions during situations with competing response alternatives; and left inferior parietal cortex may be related to the integration of feature knowledge about objects represented in modality-specific association cortices. We also administered a degraded-similarity condition where the task of categorizing a written object description was made more difficult by perceptually degrading the stimulus materials. The degraded condition and the rule-based condition, but not the similarity-based condition, were associated with caudate activation. The caudate may support resource demands that are not specific for a particular categorization process. These findings associate partially distinct large-scale neural networks with different forms of categorization in semantic memory.

Neural representation of verb meaning: An fMRI study

The neural basis for verb comprehension has proven elusive, in part because of the limited range of verb categories that have been assessed. In the present study, 16 healthy young adults were probed for the meaning associated with verbs of MOTION and verbs of COGNITION. We observed distinct patterns of activation for each verb subcategory: MOTION verbs are associated with recruitment of left ventral temporal‐occipital cortex, bilateral prefrontal cortex and caudate, whereas COGNITION verbs are associated with left posterolateral temporal activation. These findings are consistent with the claim that the neural representations of verb subcategories are distinct. Although the “sensory‐motor” hypothesis may play a role in explaining activation associated with MOTION verbs, the left posterolateral temporal distribution of cortical activation associated with COGNITION verbs cannot be easily explained by the “sensory‐motor” hypothesis. We suggest that left posterolateral temporal activation supports aspects of lexical semantic processing concerned with the neural representation of propositional knowledge contributing to COGNITION verbs. Hum. Brain Mapping 15:124–134, 2002.

The neural basis for novel semantic categorization.

We monitored regional cerebral activity with BOLD fMRI during acquisition of a novel semantic category and subsequent categorization of test stimuli by a rule-based strategy or a similarity-based strategy. We observed different patterns of activation in direct comparisons of rule- and similarity-based categorization. During rule-based category acquisition, subjects recruited anterior cingulate, thalamic, and parietal regions to support selective attention to perceptual features, and left inferior frontal cortex to helps maintain rules in working memory. Subsequent rule-based categorization revealed anterior cingulate and parietal activation while judging stimuli whose conformity with the rules was readily apparent, and left inferior frontal recruitment during judgments of stimuli whose conformity was less apparent. By comparison, similarity-based category acquisition showed recruitment of anterior prefrontal and posterior cingulate regions, presumably to support successful retrieval of previously encountered exemplars from long-term memory, and bilateral temporal-parietal activation for perceptual feature integration. Subsequent similarity-based categorization revealed temporal-parietal, posterior cingulate, and anterior prefrontal activation. These findings suggest that large-scale networks support relatively distinct categorization processes during the acquisition and judgment of semantic category knowledge.

Sentence processing strategies in healthy seniors with poor comprehension: An fMRI study

We used fMRI to examine patterns of brain recruitment in 22 healthy seniors, half of whom had selective comprehension difficulty for grammatically complex sentences. We found significantly reduced recruitment of left posterolateral temporal [Brodmann area (BA) 22/21] and left inferior frontal (BA 44/6) cortex in poor comprehenders compared to the healthy seniors with good sentence comprehension, cortical regions previously associated with language comprehension and verbal working memory, respectively. The poor comprehenders demonstrated increased activation of left prefrontal (BA 9/46), right dorsal inferior frontal (BA 44/6), and left posterior cingulate (BA 31/23) cortices for the grammatically simpler sentences that they understood. We hypothesize that these brain regions support an alternate, nongrammatical strategy for processing complex configurations of symbolic information. Moreover, these observations emphasize the crucial role of the left perisylvian network for grammatically guided sentence processing in subjects with good comprehension.

Neural Basis for Verb Processing in Alzheimer's Disease: An fMRI Study

Patients with probable Alzheimers disease (AD) have difficulty understanding verbs. To investigate the neural basis for this deficit, the authors used functional magnetic resonance imaging to examine patterns of neural activation during verb processing in 11 AD patients compared with 16 healthy seniors. Subjects judged the pleasantness of verbs, including MOTION verbs and COGNITION verbs. Healthy seniors and AD patients both activated posterolateral temporal and inferior frontal regions during judgments of verbs. These activations were relatively reduced and somewhat changed in their anatomic distribution in AD patients compared with healthy seniors, particularly for the subcategory of MOTION verbs, but AD patients showed minimal activation in association with COGNITION verbs. These findings imply that poor performance with verbs in AD is due in part to altered activation of the large-scale neural network that supports verb processing.

Confrontation Naming and Morphometric Analyses of Structural MRI in Frontotemporal Dementia

We studied the neural basis for confrontation naming difficulty in 29 patients with frontotemporal dementia (FTD) by correlating naming with voxel-based morphometric analyses of gray matter volume in structural MRI. We found that naming is significantly impaired in FTD, including patients with semantic dementia (SD), progressive nonfluent aphasia (PNFA), and nonaphasic patients (NON-APH) with a disorder of social and executive functioning. Significant cortical atrophy was found in the left anterior temporal cortex in all three FTD subgroups relative to healthy seniors. We also found significant cortical atrophy in unique anatomic distributions in each FTD subgroup. This included: lateral, ventral, and parahippocampal regions of the left temporal lobe in SD; inferior, orbital, dorsolateral, and premotor regions of the left frontal lobe in PNFA, and bilateral frontal regions in NON-APH. Direct correlations between confrontation naming and gray matter volume revealed distinct patterns in each FTD subgroup. SD patients showed a significant correlation in the left lateral temporal cortex, PNFA patients in several left frontal regions, and NON-APH patients in the right dorsolateral prefrontal cortex. These findings suggest that confrontation naming is supported by a large-scale neural network, and that naming is compromised in FTD due to interruption of the network in several different ways.

Category-specific medial temporal lobe activation and the consolidation of semantic memory: evidence from fMRI

Semantic memory consolidation was studied by comparing medial temporal lobe (MTL) fMRI activation to ANIMAL, IMPLEMENT and ABSTRACT nouns in healthy seniors to that of young adults. Relative to healthy seniors, young adults were predicted to show greater MTL activation for IMPLEMENTS, but not ANIMALS, because the ANIMALS category consists of highly intercorrelated and overlapping features that should require less MTL-mediated binding than IMPLEMENTS over a shorter period of time during concept consolidation. ABSTRACT meanings are context-dependent and do not consist of fixed feature sets. Thus it was predicted that ABSTRACT words would not involve age-related feature binding mediated by the MTL. These predictions were confirmed by the results. Our observations are consistent with the hypothesis that the structure of a category influences the consolidation of knowledge in semantic memory.