Psycholinguistics

Psycholinguistics examines the mental processes that underlie language acquisition, comprehension and production. It combines insights from linguistics, cognitive psychology, neuroscience and computer science to model how speakers store, retrieve and interpret linguistic information. Key topics include the organisation of the mental lexicon, mechanisms of syntactic parsing, the integration of semantic and phonological cues and the role of working memory and prediction in real-time language processing. Experimental approaches range from behavioural tasks (such as self-paced reading or lexical decision) to neurophysiological measures (EEG/MEG, intracranial recordings) and computational simulations. By elucidating how language is represented and accessed in the brain, psycholinguistics informs theories of language learning, the treatment of language disorders and the development of human-like language technologies.

Research from Nature Portfolio

Intracranial recordings in epilepsy patients reading semantically rich versus non-referential sentences revealed rapid high-frequency activations in inferior frontal sulcus, medial parietal and orbitofrontal cortices. These complementary “mosaics” support parallel pathways for incremental semantic binding, with perturbations in each region correlating with fluctuations in comprehension. In a separate cross-species fMRI study, rule-based auditory sequence processing in macaques and humans engaged homologous ventral frontal and opercular regions, indicating that evolutionarily conserved circuits for sequence analysis were co-opted for human syntax. Complementing these findings, dense electrocorticography in patients listening to natural narrative showed that both human cortex and autoregressive deep-learning models share three core computations: continuous next-word prediction, post-onset surprise signalling and contextual embedding of word meaning. Together, these studies delineate a hierarchical, predictive framework in which rapid, domain-specific language circuits interface with conserved sequence-processing mechanisms.

Psycholinguistics publication trend

The graph below shows the total number of articles in psycholinguistics across all publications each year (not limited to Nature Index journals).

Technical terms

Mental lexicon: The mental store of word representations, including phonological form, syntactic category and semantic features.

Semantic integration: The process of combining individual word meanings into coherent sentence- or discourse-level interpretations.

Phonological encoding: The conversion of a selected lexical item into a sequence of phonemes ready for articulation.

Predictive coding: A hierarchical framework in which higher-order brain areas generate expectations about incoming input and update these hypotheses based on prediction errors.

High-frequency broadband gamma: Neural oscillations in the 70–150 Hz range linked to local cortical processing of semantic or syntactic information.

Contextual embedding: A continuous vector representation of word meaning derived from surrounding linguistic context, analogous to those used by modern language models.

References

1. The spatiotemporal dynamics of semantic integration in the human brain. Nature Communications (2023).
2. Auditory sequence processing reveals evolutionarily conserved regions of frontal cortex in macaques and humans. Nature Communications (2015).
3. Shared computational principles for language processing in humans and deep language models. Nature Neuroscience (2022).
4. Large-scale evidence for logarithmic effects of word predictability on reading time. Proceedings of the National Academy of Sciences of the United States of America (2024).
5. Neurophysiological signatures of prediction in language: A critical review of anticipatory negativities. Neuroscience & Biobehavioral Reviews (2024).
6. A deep hierarchy of predictions enables online meaning extraction in a computational model of human speech comprehension. PLOS Biology (2023).