I have recently purchased a headset from Emotiv (the EEG) and intend to use it to change the world.
One of the most interesting points regarding the Emotiv EEG and EPOC lines are that they are fairly wide spread (meaning a large number of people have them). Because there are a large number of individuals with an EPOC and EEG I have the ability to create an application that can be used by a fairly large number of people. In turn, enabling me to obtain a large sample size for any possible research I intend to do.
In this case, my application is related to what I am going to call, the good story problem. Put simply what makes a good story or book? It has been literally thousands of years humans have been telling stories to one another, it is one of the only reasons to have language and yet we have only a rudimentary understanding of what makes a story a good one. Sure, we all would describe a good story as entertaining or perhaps funny, maybe something that makes you happy, all very vague descriptions. Where’s the quantitative support for a good story!?
What would be much more interesting is to have a clearly defined pattern for a good story, more than just a rough outline. To do this, I need data. Lots of data. Lots lots lots of data. What better way to get this data than through a distributed network of neuroscience buffs like myself? What I intend to do is simple…
I intend to create an application using raw EEG data as well as basic EPOC plugins that allow users to read pdf’s on the screen and correlate books, pages, paragraphs, and perhaps even sentences and words with EEG data/emotions. The application would have to be downloadable and easy to use, both of which I could achieve within the next couple of months. After that I simply have to compress the data and send it to a server somewhere (likely I will simply send it to my home computer). ** I will also have users rate how much they enjoyed a particular story and what not.
Once I have the data I can link specific structure of sentences, paragraphs, etc. to EEG data and emotions. Clearly, there are patterns to “good stories” yet computers have issues detecting patterns which require both analysis of forest and trees. Further, this analysis would be further compounded by reader bias (dislike a particular genre, can’t read well, etc.). I hope to solve these issues by enabling classification of readers thorough a personality test(s), as well as linking emotional data to words, phrases, and sentences.
In all likelihood this will fail miserably, but I do intend to give it my all. I believe it to be clear that good stories follow some basic guidelines, in researching this subject (in this manner) I hope to at least glimpse what I believe to be one of the most interesting questions of humanity.
I will probably use: synaptitude.me to post updates, downloads, and possibly my data.