Text-Based Natural Language Processing in the Human Brain

Natural Language Processing (NLP) is a subfield of artificial intelligence focused on processing and interpreting human language. With the recent success of NLP approaches across a diverse range of application areas, such as machine translation and sentiment analysis, the question arises as to how far the seemingly human-like performances indeed resemble the processing of language in the human brain. Neurological studies of connected speech are central to providing deeper insights into the possible connection between language representations in the human brain and language models used in NLP. The collection of connected speech data in multiple cohorts of older adults will provide the basis for this work. By employing state-of-the-art NLP methods such as the Transformer, numerical feature representations will be generated for sentences and multi-sentence units extracted from textual transcriptions of the collected speech data. These NLP features will be compared to functional brain connectivity patterns to study their link to human language processing. Another aim of this work is to investigate whether NLP features based on connected speech from older adults can be used to differentiate between healthy aging, preclinical, and the early dementia stage of Alzheimer’s disease. As a result, this work will potentially provide a deeper understanding of the connection between the derived NLP features and the aging brain and the possible existence of human brain counterparts of given NLP representations.