The "Fusiform Face Area (FFA)" and "Parahippocampal Place Area (PPA)" are two of the most popular of these 'clue is in the name' areas. The FFA lights up in response to seeing faces, while the PPA is more into places... so textbooks will tell you.
But how selective are these areas really? We know that the FFA activates more to faces than to other things on average, but is there overlap? Are there non-faces that activate the FFA more than many faces? Or perhaps, are there faces that don't activate the FFA? If there were, it would undermine the whole concept of the FFA as a "face area".
NIH researchers Marieke Mur et al have just examined this question and the results are out now in the Journal of Neuroscience. They show that yes, the FFA is selective for faces and likewise, the PPA is really pretty place-selective.
Which is a nice surprise. The last few weeks have been tough ones for brain scanning, with a crop of studies showing potentially serious flaws in popular neuroimaging methods. I was bracing myself for more bad news when I picked up this paper, but the results are actually quite reassuring.
They fMRI scanned 4 volunteers twice each, and showed them a series of 96 pics - some faces, some places, plus bodies and objects. They found that there was little overlap in the degree of activation across categories: few non-faces activated the FFA as strongly as faces - the closest were bodies, which are living things too so that makes sense. The PPA was even more selective.
See the picture at the top which shows stimuli ranked by how much they activated each area.
Fascinatingly - well, amusingly - the least face-like picture was some garlic, but the most unplaceish image was a cucumber.
Maybe these volunteers just didn't like vegetables.
There was some overlap between faces and non-faces for the FFA activations but it wasn't consistent. The second scan session - with the same 96 pics - found that those non-faces that activated the FFA the first time, didn't do so the second time. So it was just random noise.
Overall this is a lovely study, although it does underline the fact that although these areas are consistently selective, the difference is pretty subtle; there's no "break point". The FFA does prefer faces... but not enormously so.
Mur M, Ruff DA, Bodurka J, De Weerd P, Bandettini PA, and Kriegeskorte N (2012). Categorical, yet graded - single-image activation profiles of human category-selective cortical regions. The Journal of neuroscience : the official journal of the Society for Neuroscience, 32 (25), 8649-62 PMID: 22723705
7 comments:
I'm not a researcher (and nor have I read this paper, only your post and the abstract) but anyway, please explain:
You say (not the abstract) that paper has four subjects, and yet it appears to draw conclusions about all brains how can this be valid? There is a great deal of heterogeneity in human anatomy and brains in particular are very plastic. What is the proof that what can be said of the brain of a volunteer for the paper (probably a 20 something student) can also be said of 80 year old Greenland Eskimo? With such a small group how can you be sure that it's representative even of 20 somethings?
Psychology papers also seems very keen to draw general conclusions from very skewed experimental groups.
I can't imagine that I'm the first person to think this, so I suspect there's a good answer.
I'd be curious to see the fMRI of a reaction to Mt.Rushmore. Would that elicit responses in both areas, or neither?
Max West: That's an absolutely brilliant idea! Someone should do that. Probably it would activate both, maybe it would depend on which part of it you were attending to...:)
It might also depend on whether the participant had _been_ to Mt. Rushmore-- an experience that would reinforce its existence as a place, rather than an image of faces.
So all brain function isn't located in one area nor is it uniformly distributed. Fascinating!
Not.
That's really cool. Just one question, what if the person is blind? Does it still light up?
Blind salmon maybe.
Post a Comment