The emergentist-connectionist approach assumes that language processing reflects interaction between primary neural systems (Primary Systems Hypothesis). This idea offers an overarching framework that generalizes to various kinds of (English) language and nonverbal cognitive activities. The current study advances this approach with respect to language in two new and important ways. The first is the provision of a neuroanatomically constrained implementation of the theory. The second is a test of its ability to generalize to a language other than English (in this case Japanese) and, in particular, to a feature of that language (pitch accent) for which there is no English equivalent. A corpus analysis revealed the presence and distribution of typical and atypical accent forms in Japanese vocabulary, forming a quasiregular domain. Consequently, according to the Primary Systems Hypothesis, there should be a greater semantic impact on the processing of words with an atypical pitch accent. In turn, when word meaning is intrinsically less rich (e.g., abstract words), speakers should be prone to regularization errors of pitch accent. We explored these semantic-phonological interactions, first, in a neuroanatomically constrained, parallel-distributed processing model of spoken language processing. This model captured the accent typicality effect observed in nonword repetition in Japanese adults and children and exhibited the predicted semantic impact on repetition of words with atypical accent patterns. Second, also as predicted, in word repetition and immediate serial recall of spoken words, human participants exhibited reduced pitch–accent accuracy and/or slower RT for low imageability words with atypical accent patterns, and they generated accent errors reflecting the more typical accent patterns found in Japanese.

Prior lesion and functional imaging studies have highlighted the importance of the left ventral occipito-temporal (LvOT) cortex for visual word recognition. Within this area, there is a posterior–anterior hierarchy of subregions that are specialized for different stages of orthographic processing. The aim of the present fMRI study was to dissociate the effects of subword orthographic typicality (e.g., cider [high] vs. cynic [low]) from the effect of lexicality (e.g., pollen [word] vs. pillen [pseudoword]). We therefore orthogonally manipulated the orthographic typicality of written words and pseudowords (nonwords and pseudohomophones) in a visual lexical decision task. Consistent with previous studies, we identified greater activation for pseudowords than words (i.e., an effect of lexicality) in posterior LvOT cortex. In addition, we revealed higher activation for atypical than typical strings, irrespective of lexicality, in a left inferior occipital region that is posterior to LvOT cortex. When lexical decisions were made more difficult in the context of pseudohomophone foils, left anterior temporal activation also increased for atypical relative to typical strings. The latter finding agrees with the behavior of patients with progressive anterior temporal lobe degeneration, who have particular difficulty recognizing words with atypical orthography. The most novel outcome of this study is that, within a distributed network of regions supporting orthographic processing, we have identified a left inferior occipital region that is particularly sensitive to the typicality of subword orthographic patterns.

It has been claimed that semantic dementia (SD), the temporal variant of fronto-temporal dementia, is characterized by an across-the-board deficit affecting all types of conceptual knowledge. We here confirm this generalized deficit but also report differential degrees of impairment in processing specific semantic word categories in a case series of SD patients ( N = 11). Within the domain of words with strong visually grounded meaning, the patients' lexical decision accuracy was more impaired for color-related than for form-related words. Likewise, within the domain of action verbs, the patients' performance was worse for words referring to face movements and speech acts than for words semantically linked to actions performed with the hand and arm. Psycholinguistic properties were matched between the stimulus groups entering these contrasts; an explanation for the differential degrees of impairment must therefore involve semantic features of the words in the different conditions. Furthermore, this specific pattern of deficits cannot be captured by classic category distinctions such as nouns versus verbs or living versus nonliving things. Evidence from previous neuroimaging research indicates that color- and face/speech-related words, respectively, draw most heavily on anterior-temporal and inferior-frontal areas, the structures most affected in SD. Our account combines (a) the notion of an anterior-temporal amodal semantic “hub” to explain the profound across-the-board deficit in SD word processing, with (b) a semantic topography model of category-specific circuits whose cortical distributions reflect semantic features of the words and concepts represented.

On the basis of a theory about the role of semantic knowledge in the recognition and production of familiar words and objects, we predicted that patients with semantic dementia would reveal a specific pattern of impairment on six different tasks typically considered “pre-” or “non-” semantic: reading aloud, writing to dictation, inflecting verbs, lexical decision, object decision, and delayed copy drawing. The prediction was that all tasks would reveal a frequency-by-typicality interaction, with patients performing especially poorly on lower-frequency items with atypical structure (e.g., words with an atypical spelling-to-sound relationship; objects with an atypical feature for their class, such as the hump on a camel, etc). Of 84 critical observations (14 patients performing 6 tasks), this prediction was correct in 84/84 cases; and a single component in a factor analysis accounted for 87% of the variance across seven measures: each patient's degree of impairment on atypical items in the six experimental tasks and a separate composite score reflecting his or her degree of semantic impairment. Errors also consistently conformed to the predicted pattern for both expressive and receptive tasks, with responses reflecting residual knowledge about the typical surface structure of each domain. We argue that these results cannot be explained as associated but unrelated deficits but instead are a principled consequence of a primary semantic impairment.

Previous studies of patients with phonological and surface alexia have demonstrated a double dissociation between the reading of pseudo words and words with atypical spelling-to-sound relationships. A corresponding double dissociation in the neuronal activation patterns for pseudo words and exception words has not, however, been consistently demonstrated in normal subjects. Motivated by the literature on acquired alexia, the present study contrasted pseudo words to exception words and explored how neuronal interactions within the reading system are influenced by word type. Functional magnetic resonance imaging was used to measure neuronal responses during reading in 22 healthy volunteers. The direct comparison of reading pseudo words and exception words revealed a double dissociation within the left frontal cortex. Pseudo words preferentially increased left dorsal premotor activation, whereas exception words preferentially increased left pars triangularis activation. Critically, these areas correspond to those previously associated with phonological and semantic processing, respectively. Word-type dependent interactions between brain areas were then investigated using dynamic causal modeling. This revealed that increased activation in the dorsal premotor cortex for pseudo words was associated with a selective increase in effective connectivity from the posterior fusiform gyrus. In contrast, increased activation in the pars triangularis for exception words was associated with a selective increase in effective connectivity from the anterior fusiform gyrus. The present investigation is the first to identify distinct neuronal mechanisms for semantic and phonological contributions to reading.

Previous studies have found that the lateral posterior fusiform gyri respond more robustly to pictures of animals than pictures of manmade objects and suggested that these regions encode the visual properties characteristic of animals. We suggest that such effects actually reflect processing demands arising when items with similar representations must be finely discriminated. In a positron emission tomography (PET) study of category verification with colored photographs of animals and vehicles, there was robust animal-specific activation in the lateral posterior fusiform gyri when stimuli were categorized at an intermediate level of specificity (e.g., dog or car). However, when the same photographs were categorized at a more specific level (e.g., Labrador or BMW), these regions responded equally strongly to animals and vehicles. We conclude that the lateral posterior fusiform does not encode domain-specific representations of animals or visual properties characteristic of animals. Instead, these regions are strongly activated whenever an item must be discriminated from many close visual or semantic competitors. Apparent category effects arise because, at an intermediate level of specificity, animals have more visual and semantic competitors than do artifacts.

PET was used to investigate the neural correlates of action knowledge in object representations, particularly the left lateralized network of activations previously implicated in the processing of tools and their associated actions: ventral premotor cortex (VPMCx), posterior middle temporal gyrus (PMTG), and intraparietal sulcus (IPS). Judgments were made about the actions and functions associated with manipulable man-made objects (e.g., hammer); this enabled us to measure activations in response to both explicit and implicit retrieval of knowledge about actions associated with manipulable tools. Function judgments were also made about nonmanipulable artifacts (e.g., traffic light) providing a direct comparison for manipulable objects. Although neither the left VPMCx nor the left PMTG were selective for tool stimuli (nonmanipulable objects also activated these areas relative to a visual control condition), both regions responded more strongly to manipulable objects, suggesting a role for these cortical areas in the processing of knowledge associated with tools. Furthermore, these activations were insensitive to retrieval task, suggesting that visually presented tools automatically recruit both left VPMCx and left PMTG in response to action features that are inherent in tool representations. In contrast, the IPS showed clear selectivity for explicit retrieval of action information about manipulable objects. No regions of cortex were more activated by function relative to action judgments about artifacts. These results are consistent with the brain's preferential responsiveness to how we interact with objects, rather than what they are used for.

Two types of theoretical account have been proposed to explain the phenomenon of category-specific impairment in tests of semantic memory: One stresses the importance of different cortical regions to the representation of living and nonliving categories, while the other emphasize the importance of statistical relationships among features of concepts belonging to these two broad semantic domains. Theories of the latter kind predict that the direction of a domain advantage will be determined in large part by the overall damage to the semantic system, and that the profiles of patients with progressive impairments of semantic memory are likely to include a point at which an advantage for one domain changes to an advantage for the other. The present series of three studies employed semantic test data from two separate cohorts of patients with probable dementia of Alzheimer's type (DAT) to look for evidence of such a crossover. In the first study, longitudinal test scores from a cohort of 58 patients were examined to confirm the presence of progressive semantic deterioration in this group. In the second study, Kaplan-Meier survival curves based on serial naming responses and plotted separately for items belonging to living and nonliving domains indicated that the representations of living concepts (as measured by naming) deteriorated at a consistently and significantly faster rate than those of nonliving concepts. A third study, carried out to look in detail at the performance of mildly affected patients, employed an additional cross-sectional cohort of 20 patients with mild DAT and utilized a graded naming assessment. This study also revealed no evidence for a crossover in the advantage of one domain over the other as a function of disease severity. Taken together with the model of anatomical progression in DAT based on the work of Braak and Braak (1991), these findings are interpreted as evidence for the importance of regional cerebral anatomy to the genesis of semantic domain effects in DAT.

Previous studies have suggested differences in the neural substrates of recognition memory when the contributions of perceptual and semantic information are manipulated. In a within-subjects design PET study, we investigated the neural correlates of the following factors: material type (objects or faces), semantic knowledge (familiar or unfamiliar items), and perceptual similarity at study and test (identical or different pictures). There was consistent material-specific lateralization in frontal and temporal lobe regions when the retrieval of different types of nonverbal stimuli was compared, with objects activating bilateral areas and faces preferentially activating the right hemisphere. Retrieval of memories for nameable, familiar items was associated with increased activation in the left ventrolateral prefrontal cortex, while memory for unfamiliar items involved occipital regions. Recognition memory for different pictures of the same item at study and test produced blood flow increase in left inferior temporal cortex. These results have implications for our understanding of the neural correlates of perceptual and semantic contributions to recognition memory.

Three patients with semantic dementia, involving progressive deterioration of semantic memory, performed immediate serial recall of short sequences of familiar words. On the basis of their performance in other tasks of word comprehension and production, the stimuli were selected individually for each patient as either known or unknown words. All patients showed a marked advantage in recall of known as compared to familiar but now unknown words. Errors consisted primarily of incorrect combinations of correct phoneme sequences in the stimulus string, with a large number of errors preserving onsethime syllable structure (e.g., mint, rug reproduced as “rint, mug”). Discussion focuses on the implication of these errors for the structure of phonological representations, and in particular on a hypothesis that meaning plays a crucial role in binding the elements of phonological word forms.