Inhibition, prediction and their interaction with cognitive control and working memory in sentence comprehension

• Applicants: Professor Dr. Frank Rösler; Professor Dr. Shravan Vasishth
• Subject Area: General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages ; General, Cognitive and Mathematical Psychology ; Human Cognitive and Systems Neuroscience
• Term: from 2016 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 325493514

• Final Report Year: 2022

Final Report Abstract

The project resulted in two full-length journal articles that were published in high-quality journals: Neuropsychologia and Neurobiology of Language. An R package eeguana was developed by Bruno Nicenboim as a by-product of the project. The Neuropsychologia article presents the first clear evidence for the preactivation effect using Bayes factors, and shows that the published literature suggests that the original claim by DeLong et al. (2005) is supported, but that the effect size of the pre-activation effect is likely to be very small. This has an important practical implication: the effect is unlikely to be detected accurately with conventional sample sizes. Neither the original DeLong et al. (2005) nor the larger-scale replication attempt of Nieuwland et al. (2018) could have detected the effect accurately with the sample sizes they used. The Neurobiology of Language article investigates the predictability effect as a function of constraint and showed clear evidence for the claimed post-negative positivity (PNP) argued for in previous work, but this PNP was found to have a posterior distribution rather than the claimed anterior distribution. Our conclusions is that the observed PNP probably reflects the cost of resolving a conflict between the predicted and observed word. Thus, although the project ended up not exploring the issue of individual differences as originally planned, it was highly successful in answering two important open questions about predictive processing in sentence comprehension.

Publications

• eeguana: A package for manipulating EEG data in R.
  Nicenboim, Bruno
  
• Are words pre-activated probabilistically during sentence comprehension? Evidence from new data and a Bayesian random-effects meta-analysis using publicly available data. Neuropsychologia, 142, 107427.
  Nicenboim, Bruno; Vasishth, Shravan & Rösler, Frank
  
• Does entropy modulate the prediction of German long-distance verb particles?. PLOS ONE, 17(8), e0267813.
  Stone, Kate; Vasishth, Shravan & von, der Malsburg Titus
  
• Understanding the Effects of Constraint and Predictability in ERP. Neurobiology of Language, 4(2), 221-256.
  Stone, Kate; Nicenboim, Bruno; Vasishth, Shravan & Rösler, Frank
  
• Workflow techniques for the robust use of bayes factors.. Psychological Methods, 28(6), 1404-1426.
  Schad, Daniel J.; Nicenboim, Bruno; Bürkner, Paul-Christian; Betancourt, Michael & Vasishth, Shravan