This really gets me freaked out! Martin says I’m just a grumpy old (?) man. So let me lie along theNeurocriticapproach just for a minute, and just air my frustration:
HOW CAN YOU GET A SCIENCE PUBLICATION WITH A HUMONGOUS ERROR?
Take a look at this image. It’s from a 2007 article in Science byDepue et al.
It’s supposed to show activation in the hippocampus and amygdala. Looks innocent, right? Let’s take a closer look. Slice number 1:
Yeah, that’s the hippocampus, right? Or wait…isn’t that the entorhinal cortex? Or is it the fMRI that is just very (!!!) badly coregistered to the structural MRI?
Hm, let’s take the next one. Slice number 2:
Is that better?
NO! Just the same error as before
How on earth can you make that mistake? this is the entorhinal cortex, too
Image number 3 is really the best possible error:
So let’s start with blobs from left to right:
1) since when did the hippocampus extend all the way into the lateral temporal lobe?
2+3) amygdala is not the entorhinal cortex, nor does it always extend this far anterior
4) SINCE WHEN DID THE HIPPOCAMPUS BECOME WHITE MATTER?
Guys, let’s face it: this is probably one of the bigger, non-spotted errors one can find in visualization of fMRI data. Does it help validate fMRI as a method? NO. How can this error be allowed? I have no idea. But would I trust ANY of the other spatial localizations in this article? NO WAY!
Get a grip, guys! Check your images, and get yourmedial temporal lobe your anatomyright!
-Thomas
This really gets me freaked out! Martin says I’m just a grumpy old (?) man. So let me lie along theNeurocriticapproach just for a minute, and just air my frustration:
HOW CAN YOU GET A SCIENCE PUBLICATION WITH A HUMONGOUS ERROR?
Take a look at this image. It’s from a 2007 article in Science byDepue et al.
It’s supposed to show activation in the hippocampus and amygdala. Looks innocent, right? Let’s take a closer look. Slice number 1:
Yeah, that’s the hippocampus, right? Or wait…isn’t that the entorhinal cortex? Or is it the fMRI that is just very (!!!) badly coregistered to the structural MRI?
Hm, let’s take the next one. Slice number 2:
NO! Just the same error as before
How on earth can you make that mistake? this is the entorhinal cortex, too
Image number 3 is really the best possible error:
So let’s start with blobs from left to right:
1) since when did the hippocampus extend all the way into the lateral temporal lobe?
2+3) amygdala is not the entorhinal cortex, nor does it always extend this far anterior
4) SINCE WHEN DID THE HIPPOCAMPUS BECOME WHITE MATTER?
Guys, let’s face it: this is probably one of the bigger, non-spotted errors one can find in visualization of fMRI data. Does it help validate fMRI as a method? NO. How can this error be allowed? I have no idea. But would I trust ANY of the other spatial localizations in this article? NO WAY!
Get a grip, guys! Check your images, and get yourmedial temporal lobe your anatomyright!
-Thomas