* Scott Adams, of Dilbert fame, has some thoughts about torture.
* Al Jazeera wants a copy of that Bush memo. So does conservative MP Boris Johnson, who says he’ll print rhe memo and bollocks to the Official Secrets Act.
* Generate singing text automatically.
* BoingBoing also has a comprehensive roundup of the Sony rootkit fiasco: here, here and here. If you have Sony music CDs and don’t know about this problem, you seriously need to read these and then go here to find out how to mitigate your risk. Read our previous coverage here.
* New test can diagnose schizophrenia before symptoms start. On the downside, there’s still no cure. How early would you want to know?
***Update***
I wasn’t aware that the schizophrenia link is subscribers-only. Here’s the key info:
But Ruben Gur and his colleagues are convinced it works. Earlier this year, they claimed that they could use the technique to detect whether individuals are lying or telling the truth (see Nature 437, 457; 2005). Now they have turned their attention to mental disorders.
They used magnetic resonance imaging (MRI) to scan the brains of 69 schizophrenia patients and 79 healthy controls. The images were analysed by computer to produce an algorithm that could tell the two groups apart. Rather than focusing on specific areas of the brain thought to be affected by the disorder, as has been tried in the past, they looked for subtle changes across the whole brain.
This type of approach has proved successful before — but only for images used to derive the algorithm. As soon as fresh images were introduced, the success rate plummeted. But this time the researchers say that they have overcome this problem and that they were able to classify new individuals as schizophrenic or healthy with 81% accuracy (C. Davatzikos et al. Arch. Gen. Psychiatry 62, 1218–1227; 2005).
If I invented an HIV test that was 81% accurate, I would throw it out the window. No scratch that, there’s a chance that somebody might pick it up and use it. I’d throw it in an incinerating compactor. The accuracy number indicates false negatives rather than false positives (I assume), but assuming that the test has a false-positive rate of more than one percent (which seems safe) and a real schizophrenia prevalence of one in ten thousand (pdf), less than one in a hundred positive tests will actually have the disease. As much as I appreciate the potential for functional MRI (fMRI), it looks to me like they have a ways to go yet.
ppGaz
John, the Nature (schizophrenia) article appears to be subscriber-only material. Is there an extract or executive overview?
Thx.
ppGaz
Sorry …. Tim. Not John.
rilkefan
ppGaz, the front-page poster here is apparently to be referred to as Timjohn or Johntim.
Kinda funny kinda vicious take on PZ Myers’s dissection of Scott Adams.
Mac Buckets
Finally, someone has discovered the killer ap for the intertronics!
RSA
I like the singing site, as clunky as the results are. One of my former students (in computer science) is working in his spare time on a system that should eventually be able to identify a song that you whistle or hum to it. That would be cool, I’ve always thought.
Anderson
Well, you know, I’m of two minds on that …
Louise
Tsk, tsk, Anderson. You’ve used the most common misunderstanding of schizophrenia. Please report to wikipedia immediately for re-education.
demimondian
There are actually a slough of tests for untreatable diseases. One of the most painful is for Huntington’s Chorea, which is a progressive and inevitably fatal disease which typically starts exhibiting symptoms in late middle age — after one’s children are born. It’s also an autosomal dominant; if you’ve got it, half of your biological children will inevitably have it, too.
Imagine, as an adult, being able to have yourself tested if you know yourself to be at risk. Schizophrenia may not be curable, but it’s largely treatable, particularly if caught early enough. Huntington’s chorea…not so much.
rilkefan
Huntington’s Chorea disproves g*d’s existence.
rilkefan
Check out nuchal translucency screening. O(1%) Down’s etc. rate, 85% accuracy rate, 5% false positive rate.
blogReeder
I saw that sedition thing and it looked pretty sad to me. Really sad. I think Scott nailed them for being so un-credible. Look how shrill PZ Meyer got. I don’t understand how anyone can take PZ’s side on this.
Louise
As much as I appreciate the potential for functional MRI (fMRI), it looks to me like they have a ways to go yet.
Potential is the word. One of the problems with the fMRI literature is that neuroscientists and other investigators have (understandably) jumped into using it for their specific interests, even though the baseline maps created/used in the programs for data analysis may not be as reliable as they could be.
It’s a worthy endeavor, though — trying to find a way to ID schizophrenia earlier, and before the crash-and-burn.
demimondian
I’m with Louise, here. Schizophrenia is such a devastating disease that anything which prevents, or even mitigates, the first psychotic episode would prevent immeasurable suffering.
Normals tend to view the mentally ill as odd people whose behavior just needs to be fixed. That’s true, so far as it goes, but it’s not the reason we treat mental illness. Mentally ill people suffer terribly, and not mostly from the stigma associated with being crazy — the diseases themselves create paranoia, isolation, and hallucination, all by themselves. Anything which can mitigate that is a blessing.
demimondian
BTW, rilkefan: the Down’s numbers you quote are for preliminary screens using fetal protein alpha and/or ultrasound tests for certain characteristic deformities. Testing for Downs with chorionic villi sampling or via amniocentesis is much more reliable.
Demi “FDDD was forty-one when our youngest was born” Mondian
Anderson
Right, Louise, but however unfunny my misunderstanding, the proper understanding was even less promising.
Mary
“Inevitably” is overstating it a little. There’s a 50% chance that children will inherit the disease, but it’s an independent coin flip each time. From the HD info page at the NIS:
For example, Woody Guthrie, probably the most famous HD sufferer, had eight children across three marriages. Two children out of eight developed HD and died of it, but three children died so early (ages 4, 19, and 21) that we can’t say what would have happened if they lived through their forties. But his three surviving children (55, 57 and 58) seem to have escaped HD entirely.
scs
Tim I think your source is wrong about the prevalence of schizophrenia. I have always heard it is 1%, not .01%, and a quick google search showed me to be correct. See this as one example.
scs
By the way, I just wanted to put my latest crazy theory out into the blogosphere so one of you scientists can see it. Recently obsessive compulsive disorder has been theorized to have a link to reactions to an infection with the strep bacteria. I’m also wondering if schizophrenia may also have a link with the strep bacteria. First of we all, we know shizophrenia is not entirely genetic, so if it is not, there is probably other causes, such as maybe an infectious cause. Most incidents of schizophrenia take place in late adolescent, roughly around the age that rheumatic fever (caused by the strep bacteria) is supposed to strike young adults. Rheumatic fever affects, among other things, the dopamine system in the brain. Rheumatic fever after-effects are waxing and waning like schizophrenia. Perhaps schizophrenia is an autoimmune response to a strep infection like OCD. I’ll be interested to see if any research ever comes out on my guess.
rilkefan
I thought the blood-brain barrier kept most infections out of the brain – it surprises me that there is a link to OCD. Certainly I’d expect much lower rates of infection today and a fall in the rates of any associated mental illness compared to 50 or 100 years ago.
RSA
But a doctor wouldn’t just say, “Okay, you’ve tested positive,” he’d say, “This test has such and such a false positive rate, and the base rate is such and such, which means the chances for you are such and such.” Knowing that your chances are 1 in 100 rather than 1 in 100,000, to pull some random numbers out the air, could be useful, couldn’t it?
Unless you’re saying that people are terrible in basing decisions on small probabilities (which they generally are), and that there’s no help for it (which is an open question).
Tim F.
Actually, scs is on the right side of progress on this one. Infectious agents are increasingly connected to unexpected illnesses (e.g., HPV and cancer) and I have no doubt that we’ll discover much more as time goes on. The connections can be indirect and you won’t find them unless you’re looking for them.
Tim F.
RSA, the problem is that we don’t have any other test, so if you scan positive at eleven then you just have to wait until college to find out whether you’re in the 1% of positive testees who’s doomed to a lifetime of dementia, or at least aggressive treatment to hold it back. I don’t know if I would want 99% of the test-positives living under that kind of cloud unnecessarily.
RSA
I suppose it’s an issue of proportionality; you make a very good point about the burden of treatment.
demimondian
Tim — you’re ignoring selection bias. For a broadly prescribed test, d-prime needs to be extremely high. For a more narrowly prescribed test, however, where the purpose is to scan people who, say, have a family history, a test this sensitive is a significant improvement.
scs
Well, from what I can gather about rheumatic fever (linked to the strep bacteria), it may not be the bacteria per se that is causing the long lingering problems, but the autoimmune effect that exists after the infection. In other words the infection activates an immune response that at some point crosses over into an automimmune response, long after the active infection is cleared. This exaggerated immune response may be genetic. I think this may be the same problem with OCD and strep. Because the body is constantly exposed to these common bacteria, it is difficult to prevent these problems with antibiotics. However, one new treatment for rheumatic fever is to give antibiotics for up to 10 years after an incidence. Doctors aren’t sure exactly why this works to prevent future problems, but apparently it helps. Maybe it will work for OCD as well.