Highlight
Brain Training Research Reveals Plasticity of Language Processing and Comprehension
Eye fixation patterns are used to measure language comprehension difficulty
Achievement/Results
A recent wave of research has revealed that a variety of cognitive skills, including memory and attention, can be improved following extensive practice with “brain training” tasks. New research conducted by graduate students Erika Hussey and Susan Teubner-Rhodes, and their colleagues at the Center for Advanced Study of Language at the University of Maryland, demonstrates that the benefits of brain training are not restricted to the cognitive tasks that people routinely practice. Interestingly, the benefits transfer to novel measures of language performance that are not part of the training regimen. Both students are trainees in Maryland’s “Biological and Computational Foundations of Language Diversity” program, which is supported by NSF’s Integrative Graduate Education and Research Traineeship (IGERT) program.
Hussey and Teubner-Rhodes’ brain training research targets cognitive control, the ability to regulate thoughts and actions and override dominant biases when necessary. Cognitive control is important in language comprehension due to the sequential nature of language input: in reading this article, your eyes are make successive fixations across the page, and gradually build up sentence meanings as each new word arrives. However, efficient, incremental language processing can be costly when ambiguity arises. An initial commitment to one interpretation may eventually turn out to be wrong, when subsequent words suggest an entirely different meaning. This can lead to temporary confusion and the need to reinterpret what was just read. An example of this can be seen in the New York Times headline: “Google’s computer might betters translation tool.” Hussey and Teubner-Rhodes propose that cognitive control is critical for this reinterpretation process, because the change of interpretation requires comprehenders to override an initial, dominant interpretation bias. By integrating theories and techniques that span the fields of psychology, neuroscience, and linguistics, Hussey and Teubner-Rhodes are the first to investigate whether training of cognitive control abilities generalizes to improvements in language reinterpretation, and hence reduces the number of errors in language understanding.
The research team recruited healthy native-English speaking adults and assigned them to a training or no-training control group. All participants completed pre/post assessments involving reading sentences that are susceptible to misinterpretation, such as “While the thief hid the jewelry sparkled brightly.” Eye movements were recorded to measure moment-by-moment comprehension difficulty. Each sentence was paired with a comprehension question that tested whether successful reinterpretation had occurred. For example: “Did the thief hide himself?”, where an incorrect ‘no’ responses would reflect failure to reinterpret. In between assessments, trainees received an intensive 20-hour training regimen, which included a battery of tasks geared toward improving attention regulation abilities, but did not include any kind of reading practice. All training tasks adjusted difficulty to individual performance levels, keeping participants continually engaged. One training task required participants to practice overriding a dominant strategy by being forced to rely on task-relevant, non-dominant features. In the sentence comprehension task, trainees’ accuracy to comprehension questions improved significantly following training, suggesting more successful reinterpretation. Eye movement patterns also revealed training-related improvements in reanalysis: trainees spent less time re-reading confusing sentence regions post-training. Untrained participants showed no such improvements. Importantly, individuals who improved the most during the training regimen showed the greatest gains in accuracy performance at posttest. These patterns suggest that trainees? gains were mediated by cognitive training.
Hussey and Teubner-Rhodes argue that these performance improvements show that recovery from misinterpretation may be a plastic cognitive skill, adaptable by consistent cognitive control training. Together, these findings have important implications for patient populations with cognitive control impairments that affect language skills, including an inability to revise incorrect interpretations. The present study may also inform ways of determining how to offset the depletion of cognitive control in cases of cognitive-fatigue, stress, and performance-pressure.
Address Goals
This activity addresses NSF’s strategic goal of Discovery by uncovering previously unknown aspects of the plasticity of adult language comprehension abilities.
This activity addresses NSF’s strategic goal of Learning by engaging two PhD students in cross-disciplinary research training that spans multiple fields, involves multiple supervisors, and multiple research paradigms.
