The angular gyrus serves as an interface between the non-lexical reading network and the semantic system: evidence from dynamic causal modeling.

Understanding encoded language, such as written words, requires multiple cognitive processes that act in a parallel and interactive fashion. These processes and their interactions, however, are not fully understood. Various conceptual and methodical approaches including computational modeling and neuroimaging have been applied to better understand the neural underpinnings of these complex processes in the human brain. In this study, we tested different predictions of cortical interactions that derived from computational models for reading using dynamic causal modeling. Morse code was used as a model for non-lexical decoding followed by a lexical-decision during a functional magnetic resonance examination. Our results suggest that individual letters are first converted into phonemes within the left supramarginal gyrus, followed by a phoneme assembly to reconstruct word phonology, involving the left inferior frontal cortex. To allow the identification and comprehension of known words, the inferior frontal cortex then interacts with the semantic system via the left angular gyrus. As such, the left angular gyrus is likely to host phonological and semantic representations and serves as a bidirectional interface between the networks involved in language perception and word comprehension.

On the evolution of polysensory superior temporal sulcus and middle temporal gyrus: A key component of the semantic system in the human brain.

The primary auditory cortex and other early auditory cortical areas lie on Heschl's gyrus within the Sylvian fissure. On the adjacent lateral surface of the superior temporal gyrus, the cortex processes higher order auditory information leading to auditory perception. On the ventral surface of the temporal lobe in the primate brain, there are areas that process higher order visual information leading to visual perception. These sensory-specific auditory and visual processing regions are separated by areas that integrate multisensory information within the deep superior temporal sulcus in both the macaque monkey and human brains. In the human brain, the multisensory integration cortex expands and forms the adjacent middle temporal gyrus. The expansion of this multisensory region in the language-dominant hemisphere of the human brain is critical for the emergence of semantic processing, namely, the processing of conceptual information that is not sensory specific but rather relies on multisensory integration.

Stimulus frequency alone can account for the size effect in number comparison.

In a number comparison task, the size effect (i.e, smaller values are easier to compare than larger values) is usually attributed to a psychophysics-based representation. However, alternative models assume that the size effect is a frequency effect: Smaller numbers are easier to process because they are observed more frequently. Previous studies have demonstrated that the frequency of the digits fundamentally influences the comparison size effect: In new number symbols, the frequency entirely determines the size effect. In contrast, in Arabic notation, the size effect aggregates the frequency in the actual session and the previous regular size effect. Here, we investigate whether the previously acquired regular size effect can depend on the frequency of the stimuli as well or on a psychophysics-based representation that is not yet active in new symbols. Participants in the study compared numbers that were denoted with new symbols, with the frequency of the symbols being changed throughout the session. We found that the frequency of the stimuli in both halves of the session was aggregated in the size effect. In addition, no psychophysics-based size effect was found throughout the session. These results confirm that the size effect can be created and shaped purely by the frequency of the symbols, while a psychophysics-based representation is not necessary to account for these size effect-related phenomena.

Symbolic number comparison and number priming do not rely on the same mechanism.

In elementary symbolic number processing, the comparison distance effect (in a comparison task, the task is more difficult with smaller numerical distance between the values) and the priming distance effect (in a number processing task, actual number is easier to process with a numerically close previous number) are two essential phenomena. While a dominant model, the approximate number system model, assumes that the two effects rely on the same mechanism, some other models, such as the discrete semantic system model, assume that the two effects are rooted in different generators. In a correlational study, here we investigate the relation of the two effects. Critically, the reliability of the effects is considered; therefore, a possible null result cannot be attributed to the attenuation of low reliability. The results showed no strong correlation between the two effects, even though appropriate reliabilities were provided. These results confirm the models of elementary number processing that assume distinct mechanisms behind number comparison and number priming.

Dopamine-Related Reduction of Semantic Spreading Activation in Patients With Parkinson's Disease.

Impaired performance in verbal fluency (VF) tasks is a frequent observation in Parkinson's disease (PD). As to the nature of the underlying cognitive deficit, it is commonly attributed to a frontal-type dysexecutive syndrome due to nigrostriatal dopamine depletion. Whereas dopaminergic medication typically improves VF performance in PD, e.g., by ameliorating impaired lexical switching, its effect on semantic network activation is unclear. Data from priming studies suggest that dopamine causes a faster decay of semantic activation spread. The aim of the current study was to examine the impact of dopaminergic medication on the dynamic change of word frequency during VF performance as a measure of semantic spreading activation. To this end, we performed a median split analysis of word frequency during phonemic and semantic VF task performance in a PD group tested while receiving dopaminergic medication (ON) as well as after drug withdrawal (i.e., OFF), and in a sample of age-matched healthy volunteers (both groups n = 26). Dopaminergic medication in the PD group significantly affected phonemic VF with improved word production as well as increased error-rates. The expected decrease of word frequency during VF task performance was significantly smaller in the PD group ON medication than in healthy volunteers across semantic and phonemic VF. No significant group-difference emerged between controls and the PD group in the OFF condition. The comparison between both treatment conditions within the PD group did not reach statistical significance. The observed pattern of results indicates a faster decay of semantic network activation during lexical access in PD patients on dopaminergic medication. In view of improved word generation, this finding is consistent with a concept of more focused neural activity by an increased signal-to-noise ratio due to dopaminergic neuromodulation. However, the effect of dopaminergic stimulation on VF output suggests a trade-off between these beneficial effects and increased error-rates.

Deep neural networks reveal topic-level representations of sentences in medial prefrontal cortex, lateral anterior temporal lobe, precuneus, and angular gyrus.

When reading a sentence, individual words can be combined to create more complex meaning. In this study, we sought to uncover brain regions that reflect the representation of the meaning of sentences at the topic level, as opposed to the meaning of their individual constituent words when considered irrespective of their context. Using fMRI, we recorded the neural activity of participants while reading sentences. We constructed a topic-level sentence representations using the final layer of a convolutional neural network (CNN) trained to classify Wikipedia sentences into broad semantic categories. This model was contrasted with word-level sentence representations constructed using the average of the word embeddings constituting the sentence. Using representational similarity analysis, we found that the medial prefrontal cortex, lateral anterior temporal lobe, precuneus, and angular gyrus more strongly represent sentence topic-level, compared to word-level, meaning, uncovering the important role of these semantic system regions in the representation of topic-level meaning. Results were comparable when sentence meaning was modelled with a multilayer perceptron that was not sensitive to word order within a sentence, suggesting that the learning objective, in the terms of the topic being modelled, is the critical factor in capturing these neural representational spaces.

The autocracy of meaning: Intact visuo-imitative processes may not compensate for meaningful gestures.

The dual-route models of action distinguish between a semantic and a non-semantic visuo-motor route to execute different types of gestures. Despite the large amount of evidence in support to the model, some aspects are debated. One issue concerns the recruitment of the visuo-motor route to correctly execute meaningful gestures when the semantic route is damaged. Debated predictions of the dual-route model were investigated in a sample of 32 patients with left hemisphere stroke lesions compared to 27 healthy controls. Group analysis showed that patients were equally impaired on meaningful and meaningless gestures. Single-case analysis demonstrated that most cases were more impaired on meaningful than on meaningless gestures both when they are given in separate lists and when they are intermingled. Impaired performance on the imitation of meaningful gestures in both the separate and mixed list but spared performance on meaningless gestures in the separate list is against the hypothesis that the intact visuo-motor route compensates for damage to the semantic route. These results suggest that the damaged semantic route interferes with the visuo-motor route and prevents the processing of meaningful gestures along the visuo-motor route. Furthermore, an explorative analysis was conducted to investigate the relationship between gestures imitation and pantomime of object use on verbal command and between gestures imitation and performance on linguistic tasks. Although no significant correlation emerged, patients with moderate/severe impairment on the AAT performed significantly worse on meaningful, but not on meaningless gestures than patients with mild/minimal language impairment, suggesting that praxis and language systems are independent but dynamically interact.

The update of semantic memories in amnestic mild cognitive impairment.

It is still controversial to what extent neocortical consolidated memories are susceptible of change by processes of reconsolidation and transformation throughout experience, and whether the medial temporal lobes are necessary for this update of semantic consolidated memories, as they are for episodic remembering. We hypothesize that patients with amnestic mild cognitive impairment (aMCI) who have deficits in episodic memory may also have difficulties in updating information on added new features of objects. Sixteen participants with aMCI and 20 healthy control participants performed a semantic word-to-picture task, in which they were asked to identify as belonging to a given semantic category NEW objects, that have incorporated novel features, as well as OLD items, semantically and visually SIMILAR items and UNRELATED items. Patients with aMCI made a greater percentage of errors than healthy controls. Participants globally made greater percentages of errors in difficult types of items, namely NEW and SIMILAR, as compared to easier ones, OLD and UNRELATED. Importantly, an item by diagnostic group interaction effect was observed, and post hoc analysis showed that patients with aMCI made a higher percentage of errors than controls in NEW items only. In conclusion, patients with aMCI had a particular difficulty in identifying the NEW items of the word-to-picture task as compared to the control participants, supporting the concept of a flexible and dynamic conceptual knowledge system, involving the update of semantic memories and the integration of new attributes in a constant transformation process, which is impaired in these patients.

Transition From Sublexical to Lexico-Semantic Stimulus Processing.

Resembling letter-by-letter translation, Morse code can be used to investigate various linguistic components by slowing down the cognitive process of language decoding. Using fMRI and Morse code, we investigated patterns of brain activation associated with decoding three-letter words or non-words and making a lexical decision. Our data suggest that early sublexical processing is associated with activation in brain regions that are involved in sound-patterns to phoneme conversion (inferior parietal lobule), phonological output buffer (inferior frontal cortex: pars opercularis) as well as phonological and semantic top-down predictions (inferior frontal cortex: pars triangularis). In addition, later lexico-semantic processing of meaningful stimuli is associated with activation of the phonological lexicon (angular gyrus) and the semantic system (default mode network). Overall, our data indicate that sublexical and lexico-semantic analyses comprise two cognitive processes that rely on neighboring networks in the left frontal cortex and parietal lobule.

Questionnaire Sémantique de Québec (QueSQ). Développement, validation et normalisation.

The main objective of this study was to present the development, validation and normalization of the Quebec Semantic Questionnaire (QueSQ). The QueSQ is a 12-item questionnaire designed for the rapid screening of semantic disorders. Psycholinguistic parameters that can influence performance, such as the nature of semantic features and superordinate semantic categories, were taken into account during its development. Psychometric qualities of QueSQ were demonstrated during the validation process. QueSQ normative data were established on the basis of age and level of education, from a sample of 100 Franco-Quebecers aged 50 years and over.

L'objectif principal de cet article est de présenter le développement, la validation et la normalisation du Questionnaire Sémantique de Québec (QueSQ). Le QueSQ est un questionnaire comportant 12 items, permettant le dépistage rapide des troubles sémantiques. Il a été conçu en tenant compte des paramètres psycholinguistiques pouvant influencer la performance, soit la nature des traits sémantiques et la catégorie sémantique superordonnée. L'étude de validation a permis de mettre en évidence certaines des qualités psychométriques du QueSQ. Les données normatives du QueSQ, établies en fonction de l'âge et du niveau de scolarité, ont été obtenues auprès de 100 personnes franco-québécoises âgées de 50 ans et plus.

Co-activation of Taxonomic and Thematic Relations in Spoken Word Comprehension: Evidence From Eye Movements.

Evidence from behavior, computational linguistics, and neuroscience studies supported that semantic knowledge is represented in (at least) two semantic systems (i.e., taxonomic and thematic systems). It remains unclear whether, and to what extent taxonomic and thematic relations are co-activated. The present study investigated the co-activation of the two types of semantic representations when both types of semantic relations are simultaneously presented. In a visual-world task, participants listened to a spoken target word and looked at a visual display consisted of a taxonomic competitor, a thematic competitor and two distractors. Growth curve analyses revealed that both taxonomic and thematic competitors attracted visual attention during the processing of the target word but taxonomic competitor received more looks than thematic competitor. Moreover, although fixations on taxonomic competitor rose faster than thematic competitor, these two types of competitors started to attract more fixations than distractor in a similar time window. These findings indicate that taxonomic and thematic relations are co-activated by the spoken word, the activation of taxonomic relation is stronger and rise faster than thematic relation.

The Semantic System Supports the Processing of Mathematical Principles.

Although numerous studies have shown that brain regions around the intraparietal sulcus play an important role in general mathematical or numerical processing, little is known about the specific neural correlates for processing mathematical principles. In the present study, we compared the activation intensity, multi-voxel activation patterns, and functional connectivity (FC) related to processing mathematical principles (including arithmetic and logic) with those related to arithmetic. Twenty right-handed undergraduates (10 male; aged 18-25 years) participated in the study. Results of whole-brain univariate analysis showed that brain activity in the left angular gyrus (AG) was consistently stronger for mathematical principles than for computation. Multiple-voxel activation patterns at the left middle temporal gyrus (MTG) differed between mathematical principles and arithmetical computation. Additionally, psychophysiological interaction analysis showed that the functional connectivities between (1) the left middle temporal gyrus and the intraparietal sulcus, (2) left middle temporal gyrus and left inferior frontal cortex (IFG), and (3) the intraparietal sulcus (IPS) and left angular gyrus were consistently stronger for mathematical principles than for computation. As the AG, MTG and orbital part of IFG were key regions of the semantic system, these results provided direct evidence that the semantic system plays an important role in the processing of mathematical principles. Although numerous studies have shown that brain regions around the intraparietal sulcus play an important role in numerical processing, little is known about the specific neural correlates for processing mathematical principles. This study determined how processing mathematical principles differs from mathematical computation in the brain in terms of activity levels and functional connections. Results from the univariate, multi-voxel, and functional connectivity analyses consistently revealed that the left angular gyrus, left middle temporal gyrus, and left inferior frontal gyrus were more involved in the processing of mathematical principles than in computation. These regions are connected with the intraparietal sulcus, the core region involved in mathematical processing. As the AG, MTG and orbital part of IFG were key regions of the semantic system, these results provide direct evidence for a crucial role of the semantic system in the processing of mathematical principles.

Dissociating motor-speech from lexico-semantic systems in the left frontal lobe: insight from a series of 17 awake intraoperative mappings in glioma patients.

Functional brain mapping during awake surgery procedures is the gold standard technique in the management of left frontal lobe tumors. Nevertheless, a unified picture of the language subsystems encountered during left frontal lobe mapping is still lacking. We retrospectively analyzed the 49 cortical and the 33 axonal sites of functional language mapping performed in 17 patients operated for a left frontal lobe glioma under awake conditions. Sites were tagged on the postoperative MRI, based on anatomical landmarks and intraoperative photography. All MRIs and tags were then registered in the MNI template. Speech disturbances related to motor functions (speech arrest-with or without superior limb arrest-, stuttering, and vocalization) were grouped together as "motor-speech" responses. Anomias, semantic paraphasia, perseverations, and PPTT errors were classified as "lexico-semantic" responses. MNI-registered axonal sites were used as seed for computing disconnectome maps from a tractogram atlas of ten healthy individuals, as implemented in the BCB toolkit. The cortical distribution of lexico-semantic responses appeared to be located anteriorly (pars triangularis of the inferior frontal gyrus and posterior end of the middle and superior frontal gyrus) compared to motor-speech responses (lower end of the precentral gyrus and pars opercularis). Within the white matter, motor-speech responses and lexico-semantic responses overlapped on the trajectory of the aslant and fronto-striatal tracts, but the lexico-semantic sites were located more anteriorly (mean Y coordinate on the MNI system was 21.2 mm for lexico-semantic sites and 14.3 mm for the motor-speech sites; Wilcoxon test: W = 60.5, p = 0.03). Moreover, disconnectome maps evidenced a clear distinction between the two subsystems: posterior fronto-striatal and frontal aslant tracts, corpus callosum and cortico-spinal tract were related to the motor-speech sites, whereas anterior frontal aslant tract, inferior-fronto-occipital fasciculus (IFOF) and anterior thalamic radiations were related to the lexico-semantic sites. Hence, we evidenced distinct anatomical substrates for the motor-speech and lexico-semantic systems. Regarding the aslant/fronto-striatal system, an anterior to posterior gradient was found, with a lexico-semantic role for the anterior part and a motor-speech involvement for the posterior part. For tumors abutting the precentral sulcus, posterior boundaries of the resection are made of motor-speech sites, meaning that the anteriorly located lexico-semantic system is no more functional, as a result of network reorganization by plasticity.

Two components of the Indo-Arabic numerical size effect.

In the symbolic number comparison task, the size effect (better performance for small than for large numbers) is usually interpreted as the result of the more general ratio effect, in line with Weber's law. In alternative models, the size effect might be a result of stimulus frequency: smaller numbers are more frequent, and more frequent stimuli are easier to process. It has been demonstrated earlier, that in artificial new number digits, the size effect reflects the frequencies of those digits. In the present work we investigate whether frequency also directs the size effect in Indo-Arabic numbers, in which notation, unlike in new symbols, the frequencies are already firmly established for the participants. We found that frequency has an effect on the size effect in Indo-Arabic notation, but this influence is limited. However, this limited size effect change is acquired fast at the beginning of the session. We argue that these results are more in line with the frequency-based accounts of the size effect.

The relationship between semantic access and introspective awareness.

There have long been speculations about the relationship between consciousness and language. This study aimed to determine whether an individual's level of introspective awareness, based on self-report, relates to accessibility of their semantic system as evaluated by the N400. Thirty-five university students completed the study. All were right-handed, with normal or corrected-to-normal vision, without known neurological or psychological health issues. They first performed on a lexical decision task while their brain electrophysiological responses were recorded. Then, they provided assessment ratings about their levels of introspective awareness. Analysis revealed moderate to strong correlations (Pearson's rs = 0.49-0.62) between awareness self-ratings and ease of semantic access as indexed by the N400. Correlation between the self-report measure and the objective neurophysiological measure suggests that subjective assessment of awareness may deserve more credibility, which in addition to reflecting subjective perception and evaluation about one's own higher order mental functioning, may also interact with the neurophysiological processes contributive and subject to such awareness. Implications for future research on the role of semantic network in the mechanism of introspective awareness are discussed.

The semantic system is involved in mathematical problem solving.

Numerous studies have shown that the brain regions around bilateral intraparietal cortex are critical for number processing and arithmetical computation. However, the neural circuits for more advanced mathematics such as mathematical problem solving (with little routine arithmetical computation) remain unclear. Using functional magnetic resonance imaging (fMRI), this study (N = 24 undergraduate students) compared neural bases of mathematical problem solving (i.e., number series completion, mathematical word problem solving, and geometric problem solving) and arithmetical computation. Direct subject- and item-wise comparisons revealed that mathematical problem solving typically had greater activation than arithmetical computation in all 7 regions of the semantic system (which was based on a meta-analysis of 120 functional neuroimaging studies on semantic processing). Arithmetical computation typically had greater activation in the supplementary motor area and left precentral gyrus. The results suggest that the semantic system in the brain supports mathematical problem solving.

Symbolic Numerical Distance Effect Does Not Reflect the Difference between Numbers.

In a comparison task, the larger the distance between the two numbers to be compared, the better the performance-a phenomenon termed as the numerical distance effect. According to the dominant explanation, the distance effect is rooted in a noisy representation, and performance is proportional to the size of the overlap between the noisy representations of the two values. According to alternative explanations, the distance effect may be rooted in the association between the numbers and the small-large categories, and performance is better when the numbers show relatively high differences in their strength of association with the small-large properties. In everyday number use, the value of the numbers and the association between the numbers and the small-large categories strongly correlate; thus, the two explanations have the same predictions for the distance effect. To dissociate the two potential sources of the distance effect, in the present study, participants learned new artificial number digits only for the values between 1 and 3, and between 7 and 9, thus, leaving out the numbers between 4 and 6. It was found that the omitted number range (the distance between 3 and 7) was considered in the distance effect as 1, and not as 4, suggesting that the distance effect does not follow the values of the numbers predicted by the dominant explanation, but it follows the small-large property association predicted by the alternative explanations.

Exploring classical conditioning for strengthening the links between semantic and lexical representations in pure anomia: Preliminary findings from a single case study.

The same language symptom might arise at different functional loci in people with aphasia. Therefore, it is plausible that different therapeutic interventions should be adopted to approach the same difficulties in different patients. Although this point of view is still widely accepted, recently the focus has shifted from the functional locus of a rehabilitative intervention to the mechanisms of action underlying the relearning process. We maintain that both aspects should be taken into account when programming a rehabilitative intervention; furthermore, investigating relearning mechanisms might shed new light on the functional architecture of the disrupted processes. Here, we investigated, in a single case study, whether classical conditioning was a suitable relearning paradigm for targeting word-finding difficulties in pure anomia, that is in a patient with an impairment in accessing intact output lexical representations from a spared semantic system. Using a word-repetition task on picture presentation, we contrasted a condition in which the stimulus onset asynchrony between word and picture stimuli was well suited to produce classical conditioning with a condition in which repetition training could not benefit from this learning mechanism. Only classical conditioning training exerted a significant, long-lasting effect on our patient's naming skill. Tentative implications of our results for the functional architecture of single-word processing are discussed.

Effects of categorical representation on visuospatial working memory in autism spectrum disorder.

We tested whether individuals with autism spectrum disorder (ASD) are impaired in visuospatial working memory or in the use of the semantic system, in particular in categorization processes at the service of working memory. The performance of high-functioning individuals with ASD (N = 21) in a visual same-different task adapted from Lachmann and van Leeuwen [e.g., Lachmann, T., & van Leeuwen, C. (2010). Representational economy, not processing speed, determines preferred processing strategy of visual patterns. Acta Psychologica, 134(3), 290-298] was compared to those of typically developed controls (N = 25). In a categorical identity task, two successive patterns had to be judged as the same if they belonged to the same equivalence set (cf. Garner, W. R., & Clement, D. E. (1963). Goodness of pattern and pattern uncertainty. Journal of Verbal Learning and Verbal Behavior, 2, 446-452), including all possible rotation and reflection transformations (R&R category), and as different otherwise. In a physical identity task, only patterns that matched in both shape and orientation had to be responded to as the same; all others, including category matches, had to be classified as different. Equivalence sets had different sizes (ESS). Earlier studies showed an increase in reaction time (RT) with increasing ESS and, for the physical identity task, a response conflict for category matching. Both of these effects were interpreted as evidence for a categorical code by which individual patterns are mentally represented. Assuming that categorization processes are deficient in individuals with ASD, we expected no ESS effects and, in the physical identity task, absence of a response conflict for these individuals. In contrast, we found individuals with ASD to be generally as sensitive to ESS as controls, and they showed a response conflict in the physical identity task. Thus, categorical processing seems to be intact in ASD. However, a strong overall group effect was found in RTs: Individuals with ASD are considerably slower than controls.

Numerical distance and size effects dissociate in Indo-Arabic number comparison.

Numerical distance and size effects (easier number comparisons with large distance or small size) are mostly supposed to reflect a single effect, the ratio effect, which is a consequence of activation of the analog number system (ANS), working according to Weber's law. In an alternative model, symbolic numbers can be processed by a discrete semantic system (DSS), in which the distance and size effects could originate in two independent factors: the distance effect depending on the semantic distance of the units, and the size effect depending on the frequency of the symbols. Whereas in the classic view both symbolic and nonsymbolic numbers are processed by the ANS, in the alternative view only nonsymbolic numbers are processed by the ANS, but symbolic numbers are handled by the DSS. The present work contrasts the two views, investigating whether the sizes of the distance and size effects correlate in nonsymbolic dot comparison and in symbolic Indo-Arabic comparison tasks. If a comparison is backed by the ANS, the distance and size effects should correlate, because the two effects are merely two ways to measure the same ratio effect. However, if a comparison is supported by another system-for example, the DSS-the two effects might dissociate. In the present measurements, the distance and size effects correlated very strongly in the dot comparison task, but they did not correlate in the Indo-Arabic comparison task. Additionally, the effects did not correlate between the Indo-Arabic and dot comparison tasks. These results suggest that symbolic number comparison is not handled by the ANS, but by an alternative representation, such as the DSS.