3 ms·
Extraordinary claims require extraordinary evidence, not a study with a sample size of 32, including the control group. I remain super skeptical that the statis
by eloff 11y ago
Extraordinary claims require extraordinary evidence, not a study with a sample size of 32, including the control group. I remain super skeptical that the statistically significant result actually implies a real result (and the smaller the study, the lower the chances.)
- neurotech1 11y agoIMHO This article is interesting but slightly misleading. Its known from other research that certain tasks are easier in a certain "brain state" which tDCS could stimulate. A classic example would be that a Blue Angels pilot goes through the entire display routine, in their heads before they strap into the cockpit. [0] If the pilots didn't mentally prepare to fly precisely, even a highly experienced pilot couldn't achieve that level of performance. Its likely that the brain activity between different flights, and between different flight leaders, would be somewhat consistent. [0] https://www.youtube.com/watch?v=UBIXmkpO6H4&feature=youtu.be&t=1357 https://www.youtube.com/watch?v=UBIXmkpO6H4&feature=youtu.be...
- cpncrunch 11y agoIt just says "the reduced variance reached statistical significance in >33% of individual N-back trials comparing DLPFC stim with DLPFC sham" which I think isn't quite the same as having a statistically significant overall result. If you look at the results the sham group is better in quite a few of the graphs. I get the impression that the study results weren't quite as impressive as they had hoped, and they have added some spin to make it look better.
- Laaw 11y agoSmall sample sizes don't mean low confidence, why do people constantly make that mistake? What you should be asking about are confidence intervals and reproducibility.