Publications

Abstract

When perceiving the world around us, we are constantly integrating pieces of information. The integrated experience consists of more than just the sum of its parts. For example, visual scenes are defined by a collection of objects as well as the spatial relations amongst them and sentence meaning is computed based on individual word semantic but also syntactic configuration. Having quantitative models of such integrated representations can help evaluate cognitive models of both language and scene perception. Here, we focus on language, and use a behavioral measure of perceived similarity as an approximation of integrated meaning representations. We collected similarity judgments of 200 subjects rating nouns or transitive sentences through an online multiple arrangement task. We find that perceived similarity between sentences is most strongly modulated by the semantic action category of the main verb. In addition, we show how non-negative matrix factorization of similarity judgment data can reveal multiple underlying dimensions reflecting both semantic as well as relational role information. Finally, we provide an example of how similarity judgments on sentence stimuli can serve as a point of comparison for artificial neural networks models (ANNs) by comparing our behavioral data against sentence similarity extracted from three state-of-the-art ANNs. Overall, our method combining the multiple arrangement task on sentence stimuli with matrix factorization can capture relational information emerging from integration of multiple words in a sentence even in the presence of strong focus on the verb.

Abstract

Recent artificial neural networks that process natural language achieve unprecedented performance in tasks requiring sentence-level understanding. As such, they could be interesting models of the integration of linguistic information in the human brain. We review works that compare these artificial language models with human brain activity and we assess the extent to which this approach has improved our understanding of the neural processes involved in natural language comprehension. Two main results emerge. First, the neural representation of word meaning aligns with the context-dependent, dense word vectors used by the artificial neural networks. Second, the processing hierarchy that emerges within artificial neural networks broadly matches the brain, but is surprisingly inconsistent across studies. We discuss current challenges in establishing artificial neural networks as process models of natural language comprehension. We suggest exploiting the highly structured representational geometry of artificial neural networks when mapping representations to brain data.

Abstract

Typical adults read remarkably quickly. Such fast reading is facilitated by brain processes that are sensitive to both word frequency and contextual constraints. It is debated as to whether these attributes have additive or interactive effects on language processing in the brain. We investigated this issue by analysing existing magnetoencephalography data from 99 participants reading intact and scrambled sentences. Using a cross-validated model comparison scheme, we found that lexical frequency predicted the word-by-word elicited MEG signal in a widespread cortical network, irrespective of sentential context. In contrast, index (ordinal word position) was more strongly encoded in sentence words, in left front-temporal areas. This confirms that frequency influences word processing independently of predictability, and that contextual constraints affect word-by-word brain responses. With a conservative multiple comparisons correction, only the interaction between lexical frequency and surprisal survived, in anterior temporal and frontal cortex, and not between lexical frequency and entropy, nor between lexical frequency and index. However, interestingly, the uncorrected index*frequency interaction revealed an effect in left frontal and temporal cortex that reversed in time and space for intact compared to scrambled sentences. Finally, we provide evidence to suggest that, in sentences, lexical frequency and predictability may independently influence early (<150ms) and late stages of word processing, but interact during later stages of word processing (>150-250ms), thus helping to converge previous contradictory eye-tracking and electrophysiological literature. Current neuro-cognitive models of reading would benefit from accounting for these differing effects of lexical frequency and predictability on different stages of word processing.

Abstract

The meaning of a sentence can be understood, whether presented in written or spoken form. Therefore it is highly probable that brain processes supporting language comprehension are at least partly independent of sensory modality. To identify where and when in the brain language processing is independent of sensory modality, we directly compared neuromagnetic brain signals of 200 human subjects (102 males) either reading or listening to sentences. We used multiset canonical correlation analysis to align individual subject data in a way that boosts those aspects of the signal that are common to all, allowing us to capture word-by-word signal variations, consistent across subjects and at a fine temporal scale. Quantifying this consistency in activation across both reading and listening tasks revealed a mostly left hemispheric cortical network. Areas showing consistent activity patterns include not only areas previously implicated in higher-level language processing, such as left prefrontal, superior & middle temporal areas and anterior temporal lobe, but also parts of the control-network as well as subcentral and more posterior temporal-parietal areas. Activity in this supramodal sentence processing network starts in temporal areas and rapidly spreads to the other regions involved. The findings do not only indicate the involvement of a large network of brain areas in supramodal language processing, but also indicate that the linguistic information contained in the unfolding sentences modulates brain activity in a word-specific manner across subjects.