UCL

University College London corpus

UCL dataset [Frank et al., 2013].

UCL is a dataset of word-by-word reading times collected through self-paced reading and eye-tracking experiments to evaluate computational psycholinguistic models of English sentence comprehension. 361 sentences from narrative sources, ensuring they were understandable without context, and recorded reading times from participants using both methods.

For more details check out the original paper [Frank et al., 2013].

DatasetDefinition class implementation: UCL