Multiple Posts About Multiple Comparisons

So the Anonymous winner of the Neuropublishing Joke Contest has revealed himself now as Todd, brilliant MIT neuroscience graduate student and my officemate at lab.

He wrote an explanation of multiple comparisons in the comments section of the last post, and I thought I'd repost it here because I know y'all want more statistics on the blog. Actually, I should have just made him write the explanation for me rather than trying to do it myself. It's the least he could do for a free book*, right?



Anyways, here's multiple comparions take two.  If you want to see the joke that started it all, click back to the last post.

* Apparently Todd already owns a version of The Graveyard Book, but requested it as a prize to throw me off his trail.  However, I didn't know he owned it, so that strategy failed.  After everything was sorted out, I ended up giving him Pillars of the Earth instead.

------------------------

Todd's explanation:

I feel compelled to write my own short explanation of multiple comparisons for a lay audience, because I think I'm going to need it again some day...

Imagine you have a quarter and you want to know if it always comes up heads. You flip it 5 times, and it comes up heads every time.

Because you're an expert in stats, you know that that will only happen 1 out of every 32 times with a normal quarter. In other words, the probability of getting that result with a normal quarter is around 3%. In other words, as Livia pointed out very well, we're going to say that we think this is a trick quarter, but we acknowledge there's a 3% chance that we just got a strange set of coin flips.

In scientific terms, the "null hypothesis" is that the quarter is normal. We tentatively "reject the null hypothesis", because there's only a 3% chance of a normal quarter. This is a key point about science -- EVERYTHING is tentative. We're never, ever sure about anything. We can never directly prove our hypotheses are correct, we can *only* disprove other hypotheses. And we always do this while acknowledging there's a certain chance that we're wrong. Hopefully, that chance is vanishingly small, but not always...

Now, on with the story. Say you go to the bank teller and tell them to open up the vault, because you heard a rumor they might have some counterfeit coins in there. You insist that they flip each of their 20,000 coins five times each, and if any of them come up with heads all five times, you're going to call the cops.

See the problem with this? While there's only a 1/32 chance that any one quarter will come up all-heads, when you do this 20,000 times, you expect several hundred quarters to have all-head results, through pure chance alone.

You need to be much more careful with your threshold for a counterfeit coin because you're testing so many, and so you "correct for multiple comparisons". The simplest way of doing this is just to change your mind about when you're sure a coin is counterfeit. If you're satisfied suspecting a single fake coin after 5 throws, you'd require, say, 18 throws to satisfy yourself that the bank really had a bad quarter.

Livia's explanation was great, but if you didn't get it the first time around, maybe that helped?


The Winning Neuropublishing Jokes and a Statistics Lesson

Okay, I know I'm a day late to announce the winners of the Neuropublishing Jokes Contest.  Unlike other bloggers with convenient excuses for delayed posts, (cough, cough),  I have no excuse except, well, I'll explain later. So sorry about the delay, and without further ado...



*drumroll*

The Grand Prize goes to Anonymous, for the following joke:

A brain scientist, an agent, and an editor walk into a bar...

The brain scientist stubs his toe on the bar and yells, "Ouch! I really felt that in my free nerve endings! My somatosensory cortex is going nuts!" The agent says, "Your screaming has far too much jargon. I can't sell it." And the ScienceDaily editor rewrites it to, "Leading scientists prove toes cause pain; suggest removal."


This one got the most votes from random people I pulled to my blog to help me judge.   In an unexpected twist, though, this joke is actually not eligible for a prize because Anonymous chose his/her other two jokes as the official entry.  But this just received so many compliments that I wanted to award the prize anyways, if just for bragging rights.

The book prize will go to the runners up (yes, there was a tie). 

From Liana Brooks:

Q:How many brain scientists does it make to write a bestseller? 
A: None. They taught the lab rat to do it.

Again from Anonymous:

Q: How many brain scientists does it take to write a best-seller?
A: Thousands! Of course, after you correct for multiple comparisons, only a handful are doing any real work.

Congratulations!  Both runners up requested The Graveyard Book as their prize.  So Liana, I'll be contacting you about your mailing address, and Anonymous, I have a good guess about who you are, but please contact me as well.

Okay, and here's the reason I've been procrastinating on the results.  I guess, *sigh*,  I'm going to have to explain Anonymous's second joke.  I know I'm going to explain it slightly wrong, and some statistician will come out and tell me I'm dumb, and it'll be embarrassing for all involved (where by "all involved" I mean me).  But I'll give it a try...

*rolls up sleeves*

In an ideal world, we wouldn't have to do statistics on experimental data.  If we were doing an experiment on whether morning or evening testing would result in better scores, one ideal data set would be if all morning tests were better than all evening tests:

Morning:  99, 97, 92, 95, 98
Evening:  85, 90, 82, 70, 88

  However, that's never true in the real world.  In reality, our data is noisy because of factors like individual variation, testing conditions, phase of the moon, etc.  Therefore, rather than a clean difference between the datasets, we usually end up with two overlapping datasets:

Morning: 99, 97, 92, 82, 55
Evening: 85, 92, 70, 95, 88

So see how Morning tests are mostly better, but there's alot of overlap?  With datasets like this then, there's two possible interpretations.

1.    Morning testing is better on average than Evening testing (ie, the experimental conditions are Actually different)
or
2.  The two testing conditions are the same, and the difference you get is just a fluke of the specific samples you took.  (ie, the experimental conditions are Actually the same, aka the Null hypothesis)

To get an answer, we perform a statistical test that calculates the probability of getting our data set if the conditions are Actually the Same.  This is called the p value.   In other words, if the p value is less than 5%, there is a less than 5% chance that the conditions are Actually the Same.

It's standard in the sciences now that if the p value is less than 5%, we conclude that our experimental conditions are probably different. 

With me so far? 

Okay, so the whole p value and statistics thing works fine if you just do one experiment with one statistical test at a time.  However, when you're analyzing brain imaging data, you're interested in a whole bunch of different areas.  Usually, we divide the brain into tiny cubes  a few millimeters wide, and perform a statistical test on every single one.  Now we have a problem, because even if every single one the cubes are Actually the same for the two experimental conditions,  5% of them are going to pass our test, just because of random chance.  Say we're testing 100,000 voxels -- that's 5000 voxels that will light up in our brain image due to random chance! 

Therefore, for neuroimaging, we have to do a more stringent statistical test, and this is called Correcting for Multiple Comparisons (cuz, we're testing multiple cubes, see?).  So if you're doing an expeirment, you might get activations in a whole bunch of voxels, but once you correct for multiple comparions, only a handful are actually activated.

Get it?  Funny huh?

Um, get it?

Eh, well, it's really funny to neuroscientists.  Just take my word for it.

Thanks to all the good folks who entered the contest.  Do go over to the contest and check out all the entries.  It was fun  :-)

Transition Between Storylines Without Losing Your Reader (Sky Village and Lost Mission)

Note:  Thanks for all who entered the neuropublishing jokes contest.  Winners will be announced on Saturday.

Presenting multiple storylines with different point of view (POV) characters is tricky. It’s hard to ask the reader, who’s already invested in one narrative, to start again with a new one. I find that if I’m looking for a reason to stop reading, I often do so at a storyline break.

Many novels use a clean break (new chapter or section) between storylines. While that works, I’ve seen some books attempt smoother transitions. These transitions tend to be plot specific, so they won’t work with any book, but I’ll describe a few here as brainstorming fodder. I find that a good transition keeps me interested for the following reasons:



1) Keeps the momentum: There’s less of a psychological break between sections.  Sometimes a good transition will also introduce a hook that motivates me to read the new storyline.
2) Plot refresher: If we’re returning to a previous storyline, the transition makes it easier to reorient myself. I don’t have do the work of remembering what happened and where the plot is.
3)Emotional refresher: A good transition will remind me why I’m interested in going back to that other storyline -- why I care.

The first example is from Sky Village by Monk and Nigel Ashland, which Peta Andersen lent me to help with POV changes in my own manuscript.

Sky Village begins with the story of Mei, also called Dragonfly, who goes to live with relatives when her mother is captured. She has a book, called the Tree Book, that her mother used to read from. When Mei opens the book, she finds that the stories her mother read to her, about a boy Breaker and his sister Riley, were real. What’s more, she realizes she can speak with Breaker through the book and learns that Riley has been captured by demons.


“I can’t believe this,” she said, her finger still tingling. Somehow, all this time, the Tree Book had been sharing stories about real kids. But why? And would she be able to talk to the others?
You’re really her, Breaker said. You’re dragonfly.
Mei was speechless. Finally, she asked, “Where’s Riley? Is Riley okay?”
After a moment of silence, Breaker responded. She will be soon.

The next chapter launches from Breaker’s point of view, starting from a point in time before Riley’s capture. Because of this transition, I'm invested in Breaker before his section even begins.

The storylines switch back and forth in a similar way throughout the book. At the end of a character’s section, Mei and Breaker talk, and we transition into the other character's story. As a reader, I found it helpful because it reminded me where we were in the other narrative.

For example, here’s another transition.


[Mei speaking] “Morning Man says my mother was able to communicate with the birds. But why is this happening to us? What if we can’t control it, Breaker?”
[Breaker speaking] In a few hours, I have to conjure a demon for the first time. So I guess I’ll find out.

[Start of Breaker's chapter]

As a reader, I remember, “Oh yeah, last time, Breaker was trying to conjure demons.” It’s much easier to rejoin his story that way.

What if characters from two storylines don't communicate? Athol Dickson uses a different approach in Lost Mission. He has several related narratives, some of which take place in different centuries. He transitions between them by means of an omniscient narrator. Here is one such transition between the story of an 18th century Franciscan friar to a present day woman in Mexico.


But let us be more patient than the friar, for this is just the first of many journeys we shall follow as our story leads us back and forth through space and time. Indeed, the events Fray Alejandro has set in motion have their culmination far into the future. Therefore, leaving the Franciscan and his solitary ship, we cross many miles to reach a village known as Ricon de Dolores, high among the Sierra Madres mountains of Jalisco, Mexico. And we fly further still, centuries ahead of Alejandro, to find ourselves in these, our modern times.

I have mixed feelings about this approach. It was fine for a few switches, but after 345 pages of this , I got a little annoyed. However, these transitions did manage to keep the momentum going better. It’s almost like, since there wasn’t any white space between two sections, the eye just keeps going.

Would anyone like to share good ways of transitioning between storylines?



Note:  I received Lost Mission as a free review copy.

A 100% Absolutely and Completely Realisitic Neuroscience Love Scene (Love At First Sight Blogfest)

Brain scientists wonder about many things.  Besides obvious mysteries like  whether or not we'll ever solve the problem of consciousness, there's also the perennial question of why we're so underrepresented in the romance genre.

Think about it.  When was the last time you went through a checkout aisle and saw a  Nora Roberts with a dreamy fMRI technician on the cover?  The sad truth is, our research generally has more sex appeal than we do.

So when I heard about the Love at First Sight Blogfest, I knew it was my responsibility to up the percentage of steamy brain scientist love stories available.  So without further ado, I present:

A 100% Absolutely and Completely Realisitic Neuroscience Love Scene



He rushed in five minutes before scantime, hauling a laptop bag over one shoulder and flashing an apologetic grin.

“Sorry I’m late. I went to the wrong scanner.”

She forced a polite smile but didn't try too hard to hide her annoyance. “The patient’s name is Alicia. I've explained the procedure, and she’s in the bathroom changing. Can you help me set up?”

“Sure.” He strode past her only to stop outside the control room door. “I don't think I know the combination.”

“Four eight three.” The panel flashed green, and the lock clicked.

“Four eight three,” he muttered as he followed her in.

“You’ve scanned before?”

“Yes, but we used a GE system at Stanford, not the Siemens.” He reached into his pockets, fishing out cell phone, keys, and wallet and dumping them onto the back table. She glanced at his shirt pocket and scanned his waist for a belt before demetaling herself as well. Both of them paused at next doorway, automatically patting their hip and back pockets before entering the air conditioned scanner room.

“And this is the magnet,” she said over the soft thrum of the helium pump. The scanner took up a good portion of the room, a giant horizontal cylinder with a man-sized bore.

She pointed at the bottom cabinet as she walked by. “The linens are in
there.” He opened it, bending his lanky frame in half to peer inside. By the time she finished plugging in the headcoil, he had covered the scanner bed with new sheets, readied a pillow, and placed a packet of earplugs for the patient on top. She felt slightly guilty about her rudeness earlier. The imaging center was notoriously hard to navigate.

A knock sounded from the control room. “That's probably the patient.”

He stayed out of the way as she performed the final safety checks, gave instructions to Alicia and rolled her into the scanner. “Squeeze the emergency ball if you need us,” she told her as they returned to the control room.

She settled in front of the scanner controls and dove straight into the preliminary scans. A low mechanical buzz came through the intercom, and a grainy brain image loaded onto the screen. Halfway through setting up the next scan, she paused.

“Sorry I'm not explaining more, but we're running late and this is a long paradigm. I'll try to walk you through next time.”

“No problem. I'll just look over your shoulder.”

“Thanks.” For a few minutes, there was no sound except for keyboard, mouse, and the low pitched scanner noise.

“How many subjects have you run on this paradigm?” he asked when her typing slowed.

“Seven.” She scanned the screen, double checking the parameters. “The pilots were promising, but now the group analysis doesn't show any activation. I'll try a few more before I give up.”

“Well, we'll keep our fingers crossed then.”

The scanner made a high pitched, repetitive trill as it began its functional runs. She monitored the display for a few more moments before speaking again. “To be honest, I think I'm shooting myself in the foot by using SPM's volume based normalization. The VWFA is variable enough as it is. I really should look at it again with--”

“Surface based normalization.”

She cocked her head and glanced in his direction. He looked down at his hands. When he spoke again, his voice was soft. “I feel the same. Exactly the same, in fact, about volume based normalization. It doesn't make sense why we still stick with it--”

“When surface based normalization is so obviously superior.” Impulsively, she swiveled her chair to face him. Their eyes met.

In the next room, the scanner continued to sing.

* * * * *

Whew (*fanning self*).  That's about all I can handle for now.  And now you have some insight into the love lives of neuroscientists.

Thanks to Simon Larter for alerting me to the blogfest, Courtney Reese for hosting,  Lady Glamis, whose recent bit of flash fiction inspired the ending line, and Good Omens for inspiring the second to last paragraph.

Be sure to check out the other entries in the Blogfest for some less nerdy fun.

How the Brain Responds to a Loved One's Pain

In Tender Morsels, Liga's daughter Urdda grew up not knowing the circumstances of her birth. When Urdda was fifteen, her mother finally told her of the brutal gang rape that led to her conception. Urdda ran out of the house weeping, unable to deal with the new knowledge. The following passage describes her feelings.

Why couldn’t there have been … some small tale of betrayal or bad luck for which Urdda could have consoled Mam. This was too great a pain, too monstrous a series of injuries. It lumped in the past like…. Like a bear on a hearthrug, impossible to ignore.

When we hear about a loved one suffering, we often suffer along with them. Like Urdda, we feel their pain as if it were our own. Today, I'll talk about the neural basis of this phenomenon.

Oh yeah, I have a new blog now...

It occurs to me that I never announced my new science blog "Reading and word recognition research." 
The main purpose of the new blog is to force myself to read up on my dissertation topic (nothing like the threat of public humiliation to force you to get your facts right). The articles are therefore  pretty academic and have a much narrower focus.  If you're curious (or just a glutton for punishment), feel free to hop on over and poke around.

Invoking Strong Emotions in the Reader (Tender Morsels)

I recently read Tender Morsels by Margo Lanagan. It’s not for the faint of heart, but I admire the way Lanagan invokes mood and strong emotion in the reader.


Tender Morsels tell the story of Liga, a girl who lives with her sexually abusive father. Every few months, Liga gets pregnant and her father forces her to abort the pregnancies. When Liga's father dies in a freak accident, she finally carries a pregnancy to term and gives birth to her daughter Branza.

Newly freed from her father, Liga thinks about destroying the bed they had slept in but leaves it because it is her only reminder of her mother. However, Liga's good fortune doesn't last. Not long after Branza's birth, a gang of village boys invade her house and rape her.

(Oh, did I mention this book is marketed as YA? I’m not going to go into why Knopf was smoking crack with this designation, but Peta Andersen has a nice discussion on her blog on the limits of YA.)

But back to the discussion at hand. The following passage broke my heart.  This passage appears right after the town boys leave Liga.

Liga only walked, only walked away. Slowly, because to walk was to hurt, she put the distance, step by step, between herself and her father’s house, where all her troubles had happened. No matter now that Mam had died in that bed. At least Da had called on Mam’s memory as he misused it. But that strangers should come, and with no awareness of its sacredness, one by one, have of Liga there, and think that that was the place to do such things – well, Mam must be truly dead and gone, and not watching from anywhere; clearly she was of no help to Liga now.

What was it about this passage that made me feel so deeply for Liga? I think it’s several factors.

1. It taps into the universal mother-child bond and the pain of losing a mother.
2.It invokes a well known symbol (mother as protector), and declares it useless. “Clearly she was of no help to Liga now.” Remember the first time you realized your parents couldn’t protect you from everything?
3.It takes away the one sliver of hope that remained to Liga. Her mother's memory was the only thing that kept Liga going during the years of abuse, and it was really effective to declare that memory useless now. It sent a clear message that Liga had no emotional reserves left.

I imagine these techniques and would work for emotions other than sadness as well. For example, you could have one thread of sadness throughout the story that gets resolved in the end. Actually, there's a good example of that in Princess Academy. Perhaps fodder for a future post

Do you remember a passage that made you really sad?  How did it do that?



Hope you enjoyed the post!  To receive regular updates from this blog, use one of the subscription options in the left sidebar.

Fight Scene Blogfest, in which Kyra gets beat up

Edit: Scene taken down pending revisions. 8/1/2010

I may regret this later, but I just found out that there's a fight scene blogfest going on right now.  If there's one thing my manuscript doesn't lack, it's fight scenes -- I'm always wracking my brain trying to come up with more original angles.  So even though I'm super late, joining in the fun.

Most of my fight scenes are based on my real life experience in jujitsu.  My most convincing fight scenes also happen to be the ones where the heroine gets beat up.  I'll let you connect the dots there...

Since I'm doing a last minute copy and paste from my office, the writing's still a bit rough.  But I guess that's why they call it a work-in-progress. 

And here we go!



* * * * *
[Yeah, I took it down.  Reading it was making me cringe.  Perhaps I'll put up a later version.]

* * * * *

Comments and suggestions welcome!  Once I get home, I'm looking forward to reading everybody else's entries.