WOW! A brilliant and difficult study

I saw over this holiday a few brief media reports of a study from Carnegie-Mellon that is quite different from what is usual. The research team placed subjects in an fMRI where they read on a screen one word at a time (every .5 seconds) from Chapter 9 of Harry Potter and the Sorcerer’s Stone (where he learns to fly a broom and has an early confrontation with Malfoy. As they did so the researchers monitored brain activity. Now this took 45 minutes and produced an enormous amount of data so these good folks developed a sophisticated computer program for its analysis. This is one of the first, if not the first, attempt to study the brain’s functioning during reading as it occurs in real life, which, as we know, is a very complex action. Oh wait, there is more.



They were able to correlate brain activity to the segment read because of this design. The technical details here are really something and the computational aspect way over my head, but they had analyzed 195 features in the text, parsing out words, syntax, lexical features, presence of discourse, character identity, action, and emotion, word length, place in sentence string, and on and on, so that they would be able to see the brain’s response to these. The program learned to differentiate these features among the brain activity, so that afterwards, the program was able to differentiate which of two novel passages the subjects read with 75% accuracy by looking at their brain activity.


Now some results were expected. Word length was processed in an occipital region, a visual area so no surprise there. Character actions were processed by areas such as the posterior temporal lobe and angular gyrus (close to Wernicke’s area) previously associated with detection of biological motion. (I bet mirror neurons were involved). Characters’ perspectives comprised of their intentions and feelings were processed by right parietal areas and these are known to be important for the theory of mind (what I would call the empathic understanding of another’s interiority). Working memory areas in bilateral temporal lobes were used to build up comprehension of the text as it was read word by word. They also found signs of bilateral syntactic processing normally associated with just the left side. And much, much more.   I may post on this again after digesting it longer but the methodology is spectacular and I am so impressed at their ability to carry it out. It will be very helpful in extending functional brain scans to more natural, real-lie activities. Their method for parsing language and resolving the temporal parameters of linguistic processing with brain activity is truly brilliant and a lot of work, I am sure. Most importantly it shows, even as rudimentary as it might be at this time, how diverse areas of the brain, front and back, right and left, participate in processing symbolic information of some intellectual complexity. (Think plot, action, character, feelings, discourse, and the language used to communicate it all. Why, pretty soon, they might find signs of why some fiction is better art). Again, I say, WOW!

To read the study please go to:

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s