Researchers at the 天美传媒 have, for the first time, decoded and predicted the brain activity patterns of word meanings within sentences, and successfully predicted what the brain patterns would be for new sentences.
The study used functional magnetic resonance imaging (fMRI) to measure human brain activation. 鈥淯sing fMRI data, we wanted to know if given a whole sentence, can we filter out what the brain鈥檚 representation of a word is鈥攖hat is to say, can we break the sentence apart into its word components, then take the components and predict what they would look like in a new sentence,鈥� said Andrew Anderson, a research fellow who led the study as a member of the lab of , assistant professor of at Rochester.
鈥淲e found that we can predict brain activity patterns鈥攏ot perfectly [on average 70% correct], but significantly better than chance,鈥� said Anderson, The study is published in the journal .
Anderson and his colleagues say the study makes key advances toward understanding how information is represented throughout the brain. 鈥淔irst, we introduced a method for predicting the neural patterns of words within sentences鈥攚hich is a more complex problem than has been addressed by previous studies,听which have almost all focused on single words,鈥� Anderson said. 鈥淎nd second, we devised a novel approach to map semantic characteristics of words that we then correlated to neural activity patterns.鈥�
Finding a word in a sentence
To predict the patterns of particular words within sentences, the researchers used a broad set of sentences, with many words shared between them. For example: 鈥淭he green car crossed the bridge,鈥� 鈥淭he magazine was in the car,鈥� and 鈥淭he accident damaged the yellow car.鈥� fMRI data was collected from 14 participants as they silently read 240 unique sentences.
鈥淲e estimate the representation of a word 鈥榗ar,鈥� in this case, by taking the neural activity pattern associated with all of the sentences which that word occurred in and we decomposed sentence level brain activity patterns to build an estimate of the representation of the word,鈥� explained Anderson.


What does the meaning of a word look like?
听鈥淐offee has a color, smell, you can drink it鈥攃offee makes you feel good鈥攊t has sensory, emotional, and social aspects,鈥� said senior author Raizada. 鈥淪o we built upon a model created by Jeffrey Binder at the Medical College of Wisconsin, a coauthor on the paper, and surveyed people to tell us about the sensory, emotional, social and other aspects for a set of words. Together, we then took that approach in a new direction, by going beyond individual words to entire sentences.鈥�
听The new semantic model employs 65 attributes鈥攕uch as 鈥渃olor,鈥� 鈥減leasant,鈥� 鈥渓oud,鈥� and 鈥渢ime.鈥� Participants in the survey rated, on a scale of 0-6, the degree to which a given root concept was associated with a particular experience. For example, 鈥淭o what degree do you think of 鈥榗offee鈥� as having a characteristic or defining temperature?鈥� In total, 242 unique words were rated with each of the 65 attributes.
鈥淭he strength of association of each word and its attributes allowed us to estimate how its meanings would be represented across the brain using fMRI,鈥� said Raizada.
The model captures a wider breadth of experience than previous semantic models, said Anderson, 鈥渨hich made it easier to interpret the relationship between the predictive model and brain activity patterns.鈥�
听The team was then able to recombine activity patterns for individual words, in order to predict brain patterns for entire sentences built up out of new combinations of those words. For example, the computer model could predict the brain pattern for a sentence such as, 鈥淭he family played at the beach,鈥� even though it had never seen that specific sentence before. Instead, it had only seen other sentences containing those words in different contexts, such as 鈥淭he beach was empty鈥� and 鈥淭he young girl played 蝉辞肠肠别谤.鈥�
The researchers said the study opens a new set of questions toward understanding how meaning is represented in the brain. 鈥淣ot now, not next year, but this kind of research may eventually help individuals who have problems with producing language, including those who suffer from traumatic brain injuries or stroke,鈥� said Anderson.
The Intelligence Advanced Research Projects Activity and the National Science Foundation supported the research.
