Detailed instructions for replicating some of my paper results

• Marten van Schijndel and Tal Linzen. "Can Entropy Explain Successor Surprisal Effects in Reading?" In Proceedings of SCiL. 2019.

  Explores the extent to which the future surprisal effect in reading times is purely a consequence of reader uncertainty about the future. Also explores whether humans make boundedly rational predictions when they read (Simon, 1982), or whether they compute some composite of all (or a very large number) of future states. We find that future surprisal is not wholly explained by future entropy, and we find that the model that best predicts reading times does not restrict the number of predictions available to the model (contra bounded rationality).

• Marten van Schijndel and Tal Linzen. "A Neural Model of Adaptation in Reading." In Proceedings of EMNLP. 2018.

  It has been argued that humans rapidly adapt their lexical and syntactic expectations to match the statistics of the current linguistic context. We provide further support to this claim by showing that the addition of a simple adaptation mechanism to a neural language model improves our predictions of human reading times compared to a non-adaptive model. We analyze the performance of the model on controlled materials from psycholinguistic experiments and show that it adapts not only to lexical items but also to abstract syntactic structures.

• Marten van Schijndel, William Schuler, and Peter Culicover. "Frequency Effects in the Processing of Unbounded Dependencies." In Proceedings of CogSci. 2014.

  Essentially shows that normed cloze probabilities provide insufficient frequency controls in psycholinguistic studies. Demonstrates how to combine cloze probabilities and corpus statistics to better predict behavioral data. This very simple approach is able to account for a number of previous subcategorization results.