[Previous] Old Anti-Marriage Essay | Home | [Next] Static Memes and Irrationality

curi reads a correlation study

This is my real-time unedited (just formatting cleanup) comments on an "original research article" in Frontiers in Systems Neuroscience, "Valproate reopens critical-period learning of absolute pitch".

anonymous-1 wrote:
study says drug can help learn perfect pitch:

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3848041/pdf/fnsys-07-00102.pdf
they claim it's double blind, but doesn't someone being able to do something he otherwise couldn't (they claim) tell him which group he's in and therefore unblind it?

> On days 8–14 of each treatment, we instructed participants to undergo an on-line training program for approximately 10 min per day. During each online training session, they observed a video, which trained associations between piano tones and proper names.

so it could actually be a drug for better boring video watching focus?

oh god they put ppl thru a bunch of junk tests to try to control for mood, depression, mania, being smart

> We counted a training session as complete if the subject both watched the full length of the video (up to within 15 s of the end) and answered the subsequent test question correctly.

did they decide those rules before they started?

> There was no significant correlation between the number of completed training sessions and performance

hahaha

> The experiment was double-blind, as neither participants, nor experimenters knew the randomization for treatment conditions. However, we did ask participants to intuit in which arm they received VPA treatment, and why they thought so. We also instructed them to write down any side effects they experienced during the experiment. Out of the 18 participants who completed the second treatment arm, 17 guessed correctly.

hahaha i told you it wasn't blind

fucking liars

they found out during the study it was not blind

then publish it as a blind study

what scumbags


i think the prior study asserting the critical period exists at all might be more interesting. at least if it's any good. b/c i find a critical period a bit intuitively surprising. like i wouldn't rly expect it

> Second, the analysis of the crossover, i.e., of the 17 participants for whom we have data from both arms, revealed an order-dependent effect of treatment. For participants who took VPA first, AP performance was significantly higher after VPA treatment than after placebo. In contrast, for participants who initially took placebo, there was no such difference. It may be that carry-over effects impeded performance on the AP task in the second treatment arm.

that's odd

> Relatedly, it needs to be noted that we did not test how long the effect of the improvement in AP perception lasted.

so they did not study learning perfect pitch. they studied doing better on certain tests while actively on drugs, but not any kind of longer term skill improvement. so the study title:

> Valproate reopens critical-period learning of absolute pitch

that's bullshit. they didn't study that.

> In sum, our study is the first to show a change in AP with any kind of drug treatment. The finding that VPA can restore plasticity in a fundamental perceptual system in adulthood provides compelling evidence that one of the modes of action for VPA in psychiatric treatment may be to facilitate reorganization and rewiring of otherwise firmly established pathways in the brain and its epigenome (Shen et al., 2008).

wow such bullshit

like it's bad enough they are claiming it creates plasticity for pitch stuff – maybe it just makes u better at pitch without plasticity? among other things – but then to start saying they found out about psychiatry... ugh
the big picture tho is this is explanationless "science". they don't know what VPA does or how it works, and they are focusing on correlations (btwn taking VPA and high scores on pitch tests) not explanations

> If confirmed by future replications, our study will provide a behavioral paradigm for the assessment of the potential of psychiatric drugs to induce plasticity. In particular, the AP task may be useful as a behavioral correlate. If further studies continue to reveal specificity of VPA to the AP task (or to tasks on which training or intervention is provided), critical information will have been garnered concerning when systemic drug treatments may safely be used to reopen neural plasticity in a specific, targeted way.

i think they are saying here that they have no idea if VPA (their drug) has anything to do with pitch, or just helps learning more generally

the intended use for approving psychiatry drugs is disturbing

Refuting the study like this took under 20 minutes. Then people discussed the point about whether the study was blind:

anonymous-1:
    How does that make it not double-blind?
curi:
    if you know what group you're in, that isn't blind.
    do you know what blind means? *confused*
anonymous-1:
    but you don't know, you guess
curi:
    they could tell which they were in
anonymous-1:
    they guessed which they were in
curi:
    so you think 17 out of 18 got it right by coincidence, and there was no unblinding information?
anonymous-1:
    still blind?
    they weren't told until after the study
    this is a standard thing in psych studies to find out whether the person can guess about placebo?
curi:
    if you can guess better than chance, then you have information about which group you're in (or ESP). that information means it's not fully blind. in this case they appear to have quite a lot of such info.
    "standard thing in psych studies" is not reassuring!!!
Justin Mallone:
    ya lol
    psych studies typically trash
anonymous-1:
    not by coincidence, by stuff like feeling the drug
curi:
    right, that makes it not blind.
anonymous-1:
    so yes no unblinding info
curi:
    feeling it is unblinding info
anonymous-1:
    you don't know for sure though
    hmmm
    why should that be considered unblinding?
curi:
    but if you know (from feeling it) better than chance, you know something about whether you have the placebo or not. you have information about it (just not PERFECT information if you don't know for SURE). so it's not fully blind. (and, again, they seem to in this case have had LOTS of info, so not close to blind)
Justin Mallone:
    doing double blind can be hard
curi:
    yeah in medical studies they sometimes use complex active placebos to try to make stuff blind
Justin Mallone:
    the fact that lots of stuff is done incompetently doesn’t lower the bar tho
curi:
    like try to find stuff that'll have the same side effects and other feelable consequences
anonymous-1:
    oh cool @ complex active placebos
    didn't know about that. makes sense

Elliot Temple on January 14, 2016

Messages

Want to discuss this? Join my forum.

(Due to multi-year, sustained harassment from David Deutsch and his fans, commenting here requires an account. Accounts are not publicly available. Discussion info.)