Saturday, September 05, 2009

RULE 34: WHAT NETPORN TELLS US ABOUT THE BRAIN

That is (or was) the title of a book by Ogi Ogas and Sai Gaddam,1 to be published by Dutton in 2010. How on earth do I know this? Back in July, The Neurocritic noticed a number of visitors to the post Voodoo Correlations in Social Neuroscience (about the infamous paper by Vul et al.) who came from a link on the LiveJournal of shaggirl. Specifically, the traffic was from her entry A response from our friendly scientists... I took note of it, thought I might write about it but other topics took precedence. But this week, a larger surge of visitors arrived from various LiveJournal sites.2 The reason? Let's start at the beginning.



Shaggirl was "contacted by a couple of guys researching and writing about online fanfic."
Hi, Shaggirl:

I'm a cognitive neuroscientist at Boston University writing a book for Dutton (an imprint of Penguin) about how the Internet reveals new insights into some of the oldest circuits in our brain which control romantic attraction and sexual behavior. I was very much hoping you might be willing to chat about Crack Van on LJ.

. . .

For our research, we're quite interested in learning about how people creatively use text and fiction to express and explore sexuality. If you're willing, we'd like to ask questions about Crack Van and about adult fanfic in general. If you'd like, we'd be happy to include a positive mention of you and/or Crack Van in the book (or respect your privacy, if you'd prefer).

If you have any questions about our research or book, please don't hesitate to ask! I look forward to hearing from you! :)

Dr. Ogi Ogas
Department of Cognitive and Neural Systems
Boston University
I do not want to begin a lengthy discourse on fan fiction, other than to quote Wikipedia:
Fan fiction (alternately referred to as fanfiction, fanfic, FF, or fic) is a broadly-defined term for fan labor regarding stories about characters or settings written by fans of the original work, rather than by the original creator. Works of fan fiction are rarely commissioned or authorized by the original work's owner, creator, or publisher; also, they are almost never professionally published. Fan fiction, therefore, is defined by being both related to its subject's canonical fictional universe and simultaneously existing outside the canon of that universe.
So who is Ogi Ogas? While a grad student at Boston University, he entered a well-known game show, won $500,000 and wrote about it in SEED magazine:
Who Wants to Be a Cognitive Neuroscientist Millionaire?

A researcher uses his understanding of the human brain to advance on a popular quiz show.

Even with all 4 choices available, it's obvious the answer is D ("Heroin: the sedative for coughs"). Prescription cough syrup contains codeine, which is a cough suppressant. [However, it is easy to say that when the stakes are nil. ANYWAY.]

In response to some questions about their book project, Gaddam replied:

Before putting any questions, let me give a brief overview of our scientific perspective and how it will inform the book. As cognitive neuroscientists, we are respectful of the fascinating diversity of the neural landscape. And this diversity, we believe, is reflected in the terrain of erotic fantasy.

The internet and e-publishing now allow for a revolutionary and unprecedented disclosure of all our fantasies, not just those decided as marketable and mainstream for print. Digital publishing seems to have lead to an explosion in the array of fantasies we can now experience and learn from; the loop of imagination, desire, and actuality is now tighter. We want to 'neuro-scientifically' explore what this blossoming of fantasy means for us as individuals, and as a society. How does this access to all manner of fantasies imaginable change our brains?

Given this broad overview, we have some specific questions, and more general ones about adult fanfic, that we are hoping you can help us with. I do apologize if some questions seem naive and/or I inadvertently mis-characterize something in my ignorance!
He then goes on to ask a number of questions about adult fanfic. Dana (shaggirl) replied in a lengthy, thoughtful post in which she explains fandom ("composed primarily of well educated women, most of whom self-identify as geeks"), objects to the word "netporn" in their book title, and answers all 7 questions. Then A response from our friendly scientists... where yours truly gets a mention:
. . .

One quick note about the subtitle of the book: the current title is largely driven by the pre-publication marketing needs of the publication industry, and was determined by our editor. The eventual subtitle will likely change; it may still contain the term netporn, though this is increasingly unlikely [NOTE: Unlike the screenshot above, the current book project list from Gail Ross shows the NETPORN subtitle has indeed been removed]. ...

In terms of turning this into hard science, our primary goal is to make a strong case for the study of fantasy as being accessible in this era of copious datastreams from humans. To give you one example of the kind of current study in neurosexual behavior that we feel is quite limited, there was an article in the Time magazine (http://www.time.com/time/health/article/0,8599,1911103,00.html) on how girls focus more on best friends while boys are more interested in group dynamics and 'packs'. The authors of the study tied this observation to the differential activation patterns in the brain based on how the subject group of boys and girls rated pictures of strangers. The article state that the "nucleus accumbens (which is associated with reward and motivation), hypothalamus (associated with hormone secretion), hippocampus (associated with social learning) and insula (associated with subjective feelings) all become more active."

The suspect and cavalier methodology of this study evoke some of the general problems that plague brain imaging experiments that purportedly examine social behavior [1]. We hope to make the case that such methodological contortions are not required when a vast pool of people provide behavioral data on the internet through quantifiable activity.

Polling feedback from groups would be immensely useful. We would definitely love your help in reaching other fans...

[1] Here's a paper criticizing such casual treatment of the brain-behavior linkage, which has received a lot of attention in the neuroscience community (http://cns.bu.edu/~gsc/Articles/Vul_09_Voodoo.pdf). The content requires an understanding of imaging methodology, but the abstract is definitely accessible. Here's a posting from a popular neuroscience blog describing the article (http://neurocritic.blogspot.com/2009/01/voodoo-correlations-in-social.html)
Two things:

(1) They said the content of their book will be "Using new digital sources of data to illuminate brain regions and neural pathways involved in romantic and sexual behavior." However, they did not actually propose to measure brain function at all. They might be skeptical of fMRI, but it's a lot better than no real neural data at all.

(2) A poll. That generated an enormous amount of controversy in the fanfic community. [They're a very prolific bunch.] Hundreds of posts. Thousands of comments.

SurveyFail - an entire wiki devoted to the controversy.


For Science!

The poll was taken down (although you can view the questions elsewhere: part 1, part 2) and the ogi_ogas LiveJournal deleted (but here's a remnant):
So what kind of scientists are you anyway?

We're brain modelers. We're aligned with the relatively new field of cognitive neuroscience...

. . .

Our current research project will result in a model that makes novel behavioral and physiological predictions.
From a poorly written survey? And just when you think their project is a wee bit grandiose, the pomposity reaches stunning proportions:
The greater one's mathematical ability, the greater the opportunity for designing a powerful and sophisticated model. Superior models, such as those of Stephen Grossberg and Gail Carpenter, may include unsolvable systems of differential equations. Still, useful and productive models can be constructed using relatively straightforward mathematics.
Oh, and here's a ridiculous question:
How is straight female interest in slash fiction like straight male interest in "shemale"* models? And why in the world does this matter?

* ...We are aware that the term "shemale" may be used as a pejorative, but we are here referring to the narrowly defined term used in the adult industry for certain models.
One does not need to be an expert in any of these fields to see the problems inherent in the Ogas and Gaddam approach of inferring brain function from data on netporn usage (broadly construed) and unscientific polls. But it gets worse...
The structure and activity of our subcortical circuits are shaped by neurohormones such as testosterone, estrogen, oxytocin, progesterone, and vasopressin; these circuits function differently in men and women. As cognitive neuroscientists, we draw upon a wide variety of empirical data sources to model these circuits, including brain imaging studies, primate research, cognitive science experiments, machine learning algorithms--and behavioral data. The Internet offers large, unprecedented sources of data on human activity: one of these data sets is fan fiction.

We're deeply interested in broad-based behavioral data that involves romantic or erotic cognition and evinces a clear distinction between men and women. Fan fiction matches this criteria perfectly. Let us make clear, however: fan fiction is not the subject of our research. Our subject is the human brain. For us, fan fiction is a wonderfully rich source of data--like single-neuron recordings in rhesus monkeys--albeit a unique and invaluable one.
...because fan fiction is just as precise as single-unit recording at revealing putative sex differences in subcortical circuits. In spite of a wink and a nod to "erotic cognition" Ogas & Gaddam reveal elsewhere in their manifesto that they're not at all interested in cortical function.

Furthermore, were the ethics of conducting research on human subjects taken into consideration? Did the pair have proper approval from Boston University's Institutional Review Board? Here's deadlychameleon's investigation:
I called the Boston University IRB office. The direct approach works.

They've gotten a lot of emails regarding Dr. Ogas. He is no longer in any way affiliated with Boston University, except as a recent graduate. They have asked him to stop using his official Boston University email address in connection with this project, or his website. He is officially on his own, and this project is NOT IRB APPROVED.

That is the official status as stated by the Boston University IRB office.
And in the end, from a neuroscientific viewpoint, the premise of their work was fatally flawed. Dutton, are you listening?

Footnotes

1 You'll notice these links are from the Google cache, because they took down their websites after the resulting furor. Both are recent graduates from the Department of Cognitive and Neural Systems at Boston University. Gaddam had not yet updated his resume to reflect this.

2 Other visitors were from dreamwidth.org, a LiveJournal knockoff that had escaped my awareness until now.


Subscribe to Post Comments [Atom]

15 Comments:

At September 05, 2009 7:21 AM, Blogger Elf said...

This post has been included in a linkspam roundup.

 
At September 05, 2009 12:01 PM, Blogger The Neurocritic said...

Thanks for the mention. The whole incident and its aftermath have been eye opening. They were opportunistic, that much is clear. Were they deliberately exploitative from the outset? Or just a little naïve? Or was the entire survey just a stunt to sell more books in the end?

 
At September 05, 2009 12:22 PM, Anonymous Anonymous said...

Wow, Thanks for this long entry, Neurocritic. There are no words to describe this, but it is in tune with what I hear about many folks in the Department of Cognitive and Neural Systems (yes, BC is not that far from BU). Actual data? Testing humans? That is NOT the kind of things these people are interested in or know anything about. These people are definitely NOT cognitive neuroscientists and they should not be using (abusing) this label. They should just stick with "modeling" low-level phenomena using yet another ART variant. The lack of IRB is just another example of how clueless they are about anything that actually has to do with cognitive neuroscience.

 
At September 05, 2009 12:50 PM, Anonymous neededalj said...

Thanks for providing an outsider perspective on the neuroscience. My own critique is here:

http://neededalj.livejournal.com/940.html

I was so appalled by their 'science' that I stepped in not just to debunk what they were doing, but to at least try to protect cognitive neuroscience's good name. I wouldn't blame people in fandom for writing off the whole discipline after these two researchers ran roughshod over every ethical and scientific research principle in existence.

 
At September 05, 2009 2:48 PM, Blogger The Neurocritic said...

Anonymous in Boston - Thanks for providing a local viewpoint. They do appear to be way out of their element in speculating about complex human behavior and brain function. And promising a powerful and sophisticated mathematical model of the subcortical circuits responsive to erotic fantasy (and the sex differences therein) seems more than a little grandiose.

I wonder how they pitched the idea to their agent? To the publisher? Winning $500,000 in a highly public fashion, using the methods of "cognitive neuroscience", would lend itself to a popular book. But I'm not sure how their NETPORN idea got the green light from Dutton.

neededalj - Thanks for the link to your post on Why Ogi Ogas and Sai Gaddam Are Phrenologists, which concludes:

"Ogi Ogas and Sai Gaddam are attemping to profit off of bad science by cloaking it in complicated terminology and cutting edge technology. We shouldn't let them."

Hopefully all the non-Cog Neuro folks who were affected by the debacle will not be alienated by the good efforts of others in the field.

 
At September 05, 2009 4:03 PM, Anonymous Anonymous said...

Heh, I had to Google it. :) Regarding Ogas, Wiki says:
"...he developed a new suite of cognitive techniques for Grand Slam, including calming techniques as well as mathematical, verbal, and mnemonic heuristics derived from his brain research." Really? I could not find anything about his "brain research" on pubmed. What am I missing?

 
At September 05, 2009 8:20 PM, Anonymous jumpuphigh said...

Thank you for a wonderful summation of the situation. I will be sending people here for a thorough yet consise accounting.

Thanks again!

 
At September 05, 2009 10:35 PM, Blogger nike_victory said...

As someone following this interesting fallout, I must commend you on such a nice summation. However, I feel I should point out that Dreamwidth is not an LJ knock-off, but a code-fork and, although it's still in open-beta, it's already looking rather impressive as a social-networking platform.

 
At September 06, 2009 4:32 AM, Anonymous Manna Francis said...

Oh, you missed my favourite part of his 'logic' (from a comment posted at shaggirl's LJ, now deleted), which is a glorious mix of patronising attitude, bad data, and deep stupidity:

Male rhesus monkeys will pay juice to look at pictures of female perinea (the scientific term for “bright pink monkey butt”). That sounds like human males. But wait a minute--female rhesus monkeys will pay juice to look at pictures of male penises. That doesn't sound like human females at all. The number of women who pay to look at porn is so close to zero that whenever CCBill (the largest billing system for adult sites) sees a female name on a credit card payment, it gets flagged. Usually a female name means a guy has borrowed his mom's, sister's, or girlfriend's credit card, and most of the the women will refuse to pay, so CCBill can't let any female payment go through unexamined. So now we have one clue: something is different in the human female sex circuits than in the monkey female sex circuits. We can now look at the relevant neural structures for clues for this difference.

This is exactly equivalent to saying that because in the 1800s, women didn't publicly buy and wear trousers, while men did, that this must have reflected an underlying difference in brain structure between men and women. Clearly, there's been some speedy evolution at work since.

Or, just maybe, that there are some cultural pressures at work? But, nah. Culture doesn't matter, because that's not what they're looking at!

The most charitable interpretation I can put on all this is that they're very confused. No, I don't really believe that.

 
At September 06, 2009 12:43 PM, Blogger The Neurocritic said...

Manna Francis - Ha, ha! That's amusing. Human males are like monkey males, but human females are not like monkey females. Now we've learned something about sex circuits in the brain. No cultural influences involved in why "women don't pay for porn." Here's one explanation: human females are smart enough to find free porn.

Thanks, jumpuphigh and nike_victory. For the footnoted Dreamwidth "LJ knockoff" comment, I meant the interface looks just like LJ.*

Anonymous of 4:03 PM - Here's his list of publications, which existed on his BU site before it was taken down (N.B.: the list includes his dissertation).

Publications
Ogas, O. (2009) A Neural Architecture for Visual Classification: Fast-Priming and Opponent-Color Computational Systems. Doctoral Dissertation, Boston University.

Carpenter, G., Martens, S., Ogas, O. (2005) Self-organizing information fusion and hierarchical knowledge discovery: a new framework using ARTMAP neural networks. Neural Networks, 18(3), 287-295

Carpenter, G., Martens, S., Mingolla, E., Ogas, O. (2004) Biologically Inspired Approaches to Automated Feature Extraction and Target Recognition. AIPR 2004: 33rd Workshop on Applied Imagery Pattern Recognition.

Carpenter, G. A., Martens, S., Ogas, O. J. (2004) Self-organizing Hierarchical Knowledge Discovery by an ARTMAP Image Fusion System. Proceedings of the 7th International Conference on Information Fusion (Fusion 2004), pp. 235-242.

Srinivas K, Ogas O. (1999) Disorders of somesthetic recognition: a theoretical review. Neurocase, 5, 83–93.

* I wasn't baiting, but I kinda expected this reaction (i.e., if anyone from Dreamwidth or LJ even found my post). No offense intended.

 
At September 06, 2009 2:11 PM, Anonymous Anonymous said...

Yes on the comment about Dreamwidth. It's developed LJ into significantly new capabilities, while remaining completely compatible with LJ and with IJ (which *is* a knock-off or clone), and is attracting a lot of users for that reason. It's a positive improvement and advance.

 
At September 06, 2009 6:21 PM, Anonymous Anonymous said...

I hope DW has new capabilities relative to LJ, since it costs $40/year to create an account there.

 
At September 07, 2009 10:10 AM, Blogger codeman38 said...

Just to set the record straight...

To create a Dreamwidth account, one can just find one of the many forums that's dedicated to sharing Dreamwidth invite codes with interesting people; the invite code thing is mainly to prevent the site from growing beyond its bounds. And if that ends up not being an option, it only costs $3 to create a month's paid account; once you've created an account, it's yours forever, though you lose a few features after the paid period is up.

And for what it's worth, LiveJournal did the same thing with invite codes when *it* started up, too.

 
At September 09, 2009 5:08 PM, Blogger Sean said...

This comment has been removed by the author.

 
At December 16, 2009 6:16 AM, Blogger türkiye ve hayata dair herşey said...

http://machineslikeus.com

 

Post a Comment

Links to this post:

Create a Link

<< Home

eXTReMe Tracker