A predictive coding model of the N400

Cognition, Journal Year: 2024, Volume and Issue: 246, P. 105755 - 105755

Published: Feb. 29, 2024

Language: Английский

---

Error-based Implicit Learning in Language: The Effect of Sentence Context and Constraint in a Repetition Paradigm

Journal of Cognitive Neuroscience, Journal Year: 2024, Volume and Issue: 36(6), P. 1048 - 1070

Published: Jan. 1, 2024

Abstract Prediction errors drive implicit learning in language, but the specific mechanisms underlying these effects remain debated. This issue was addressed an EEG study manipulating context of a repeated unpredictable word (repetition complete sentence or repetition new context) and constraint. For manipulation constraint, unexpected words were presented either high-constraint (eliciting precise prediction) low-constraint sentences (not eliciting any prediction). Repetition-induced reduction N400 amplitudes power alpha/beta frequency band larger for with their as compared context, suggesting that happens not only at level individual items additionally improves sentence-based predictions. These processing benefits did differ between constraint conditions, prediction update might be proportional to amount unpredicted semantic information, rather than precision violated. In addition, consequences violations, reflected frontal positivity increased theta power, reduced repetition. Overall, our findings suggest powerful adaptation mechanism allows language system quickly adapt its predictions when information is processed, irrespective reduce potential costs strong

Language: Английский

---

Predictive language processing: integrating comprehension and production, and what atypical populations can tell us

Frontiers in Psychology, Journal Year: 2024, Volume and Issue: 15

Published: May 21, 2024

Predictive processing, a crucial aspect of human cognition, is also relevant for language comprehension. In everyday situations, we exploit various sources information to anticipate and therefore facilitate processing upcoming linguistic input. the literature, there are variety models that aim at accounting such ability. One group propose strict relationship between prediction production mechanisms. this review, first introduce very briefly concept predictive during Secondly, focus on attribute prominent role sensorimotor in ("prediction-by-production" models). Contextually, provide summary studies investigated speech auditory perception comprehension/prediction tasks healthy, typical participants. Then, an overview limited existing literature specific atypical/clinical populations may represent suitable testing ground models-i.e., with impaired Ultimately, suggest more widely in-depth prediction-by-production accounts, involvement atypical both model as targets possible novel speech/language treatment approaches.

Language: Английский

---

Anticipatory prediction in older readers

Memory & Cognition, Journal Year: 2025, Volume and Issue: unknown

Published: May 6, 2025

Abstract It is well-established that skilled, young-adult readers rely on predictive processing during online language comprehension; however, fewer studies have investigated whether this extends to healthy, older adults (60 + years). The aim of the present research was assess make use lexical prediction by investigating they demonstrate costs for incorrect predictions in a controlled experimental design. eye movements sample (60–86 years) were recorded as read strongly and weakly constraining sentences containing predictable word or an unpredictable alternative either semantically related unrelated. To determine depends stimuli presentation format, second experiment presented same materials self-paced reading task which each sentence appears one at time readers’ own pace. Older showed benefits expected input eye-movement measures reading. They also unexpected across both methodologies, but only when unrelated best completion. Taken together, results suggest processes remains relatively preserved with age. implications these findings understanding fundamental component comprehension are discussed.

Language: Английский

---

Uncovering patterns of semantic predictability in sentence processing

Published: June 14, 2024

Psycholinguistic researchers have used the cloze task to measure predictability of upcoming words, but largely discarded variability in structure responses people provide. This semantic may be important for understanding selection during language production; however, it has proven difficult model participants' responses, and thus uncertainty. Recent advances large models (LLMs) permit us approximate degree most methods are restricted symbolic or hand-crafted meaning representations. We show two studies that Bayesian Gaussian mixture can cluster LLM representations produce coherent, taxonomically similar clusters. apply these clustering algorithms response time data a serial influences how quickly able provide response. clear effects competition on naming speed. In addition providing novel operationalizations what might look like task, we explain this method is extensible other datasets applications interest processing psycholinguistics.

Language: Английский

---

Intimate relationships regulate female brain activity in a competitive context: evidence from EEG and functional connectivity analysis

Cerebral Cortex, Journal Year: 2024, Volume and Issue: 34(7)

Published: July 1, 2024

Competition is common in life, and intimate relationships are essential. Understanding how impact an individual's competitive process crucial. This study explored the of competitor gender on female competition using electroencephalography analysis. The results revealed that females exhibited a smaller median absolute value reaction time difference (DRT) between their partners competitors when were absent compared to present. Additionally, showed greater average amplitudes N2 posterior contralateral component (N2pc) Late Positive Potential (LPP), increased activation alpha frequency band, enhanced theta band functional connectivity central parietal lobe occipital lobe. Furthermore, competing with individuals same as opposed opposite gender, percentage wins N2pc. A significant negative correlation was noted DRT wave N2pc LPP. These findings suggest more engaged tasks not present have improved decision-making same-gender individuals. provides evidence for influence lovers competition, helping adapt social promoting healthy relationships.

Language: Английский

---

Is there a cost when predictions are not met? A VWP study investigating L1 and L2 speakers

Quarterly Journal of Experimental Psychology, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 13, 2024

Research has found that both first language (L1) and second (L2) speakers make predictions about upcoming linguistic information, with predictive behaviour being impacted by individual differences methodological factors. However, it is not clear whether a cost incurred when prediction made, but met. L2 have less experience their parsing can be cognitively demanding, which together may lead to incur costs differently relative L1 speakers. In this study using the visual world paradigm, we test predict in same way, within time frame, if are We also explore role of proficiency speech rate. groups similar way frame. addition, neither group target was most likely alternative, though take longer shift attention object argue reflects slowing lexical access rather than specific prediction. only rate included analysis, highlighting importance attending studies paradigm. Overall, supports research showing multiple partial information predicting one candidate while inhibiting candidates.

Language: Английский

---

The neural dynamics of word recognition and integration

Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Journal Year: 2023, Volume and Issue: unknown, P. 980 - 995

Published: Jan. 1, 2023

Listeners recognize and integrate words in rapid noisy everyday speech by combining expectations about upcoming content with incremental sensory evidence. We present a computational model of word recognition which formalizes this perceptual process Bayesian decision theory. fit to explain scalp EEG signals recorded as subjects passively listened fictional story, revealing both the dynamics online auditory neural correlates integration words. The reveals distinct processing depending on whether or not they can be quickly recognized. While all trigger response characteristic probabilistic — voltage modulations predicted word’s surprisal context these are amplified for require more than roughly 150 ms input observe no difference latency responses according words’ times. Our results support two-part comprehension, an eager temporally independent integration. However, we also developed alternative models signal incorporating showed similar performance improvements. discuss potential future modeling steps may help separate hypotheses.

Language: Английский

---

Error-based implicit learning in language: the effect of sentence context and constraint in a repetition paradigm

bioRxiv (Cold Spring Harbor Laboratory), Journal Year: 2023, Volume and Issue: unknown

Published: Dec. 14, 2023

Abstract Prediction errors drive implicit learning in language, but the specific mechanisms underlying these effects remain debated. This issue was addressed an electroencephalogram (EEG) study manipulating context of a repeated unpredictable word (repetition complete sentence or repetition new context) and constraint. For manipulation constraint, unexpected words were presented either high constraint (eliciting precise prediction) low sentences (not eliciting any prediction). Repetition induced reduction N400 amplitudes power alpha/beta frequency band larger for with their as compared to context, suggesting that happens not only at level individual items additionally improves sentence-based predictions. These processing benefits did differ between conditions, prediction update might be proportional amount unpredicted semantic information, rather than precision violated. Additionally, consequences violations, reflected frontal positivity increased theta power, reduced repetition. Overall, our findings suggest powerful adaptation mechanism allows language system quickly adapt its predictions when information is processed, irrespective reduce potential costs strong

Language: Английский

---