For obvious reasons, studying neurogenesis in primates is useful. Primates are phylogenetically more related to us than rodents, and so understanding their nervous system can better help us to understand our own. For over a decade we have known that neurogenesis continues in adulthood in primates and in many ways, the process is similar to what has been observed in rodents. For example, neurogenesis is reduced with age in primates, is decreased by stress, increased in pathological conditions such as epilepsy, and increased by antidepressant treatment.
My goal in compiling this list was to assess the magnitude of adult neurogenesis in primates. It’s definitely more challenging than assessing the magnitude of neurogenesis in rodents, which we know much more about, and so I had put it off. At this point I haven’t reached a clear conclusion but, in quickly skimming these papers, the number of proliferating cells and/or new neurons averages thousand(s) of cells in the young adult primate hippocampus. The range is much much larger, and many studies cannot be easily compared due to variability in the methods, which is partly understandable since primates are scarce and are often used in multiple studies, thereby limiting the analyses that can be performed.
Thanks for the hat tip earlier. I have two more papers for you (brand new). Both are reviews/opinion pieces/perspectives, but your readership (ha, and you) might enjoy them. And both are in Neuron.
Aimone, Deng and Gage. Resolving new memories: a critical look at the dentate gyrus, adult neurogenesis, and pattern separation.
AND
Sahay, Wilson and Hen. Pattern separation: a common function for new neurons in hippocampus and olfactory bulb.
PS: I think a lot of scientists would also benefit from an audit DURING experiments, not just after 🙂 Food for thought.
Yeah I saw those just came out. A lot of excitement about pattern separation out there! I will have to read them…
I’ve always been a bit confused about the role of pattern separation – is it for learning (acquisition, encoding), storage, and retrieval, all of the above… for any sensory info?
From what Kesner originally said, the DG is required for this when similar overlap occurs and separates the environment into discrete computational parts… and DG plays a role in match-mismatch detection (Kesner’s group and Mizumori’s). Also, this process is thought to help retain “separateness” or uniqueness among memories being recalled. So you have encoding and retrieval.
Right. When you test pattern separation //alone// or in isolation, either with radial maze, cheese board, or the new touch screen system developed (Bussey & van Praag), you can see a deficit or improvement correlated with levels of neurogen. A concern is that you should be careful to distinguish between using pattern separation from the point of view of perception, such as being able to discriminate similar features (although you could argue this requires working or “very, very short, short-term” memory no matter on what time scale you talk about), with pattern separation in the memory sense, such as being able to match a location closely related in space with another location (DNMP paradigm). Is this the same basic computational component no matter what cognitive function of pattern separation you ascribe it to (one relates to perception and other to memory)?
When you are freely testing a rat exploring an environment or learning a water maze task or associative memories – fear conditioning, trace eyeblink – the rat is using many cognitive-like functions and engaging different brain areas (e.g. different neural networks or cell assemblies). You see that there are contradicting results in the literature, where sometimes a decrease in neurogen relates to impaired performance on these tasks but not all tasks (Gould & Shors’ review “Neurogenesis may relate to some but not all types of hippocampal-dependent learning.”). However, should pattern separation not play a role in ALL of this, separating out stimuli from the environment? Is it the case that other regions can ‘compensate’ for pattern separation of the DG (if this is the case) when neurogen. is depleted, since this is a behavioral task that tests many cognitive and motor functions instead of ONLY pattern separation?
Also, how much neurogenesis is necessary for pattern separation (if it is the case)? Is neurogen in dorsal and ventral required, is a mini slab sufficient, such as what Moser said about spatial memory in the dorsal?
I am inclined to think that neurogenesis relates to a more rudimentary, fundamental computational aspect that /lends/ itself to pattern separation, associative memories, spatial memories, etc. but is not directly involved. It just so happens to be involved. If that makes sense? I think figuring out what this fundamental aspect is will help clear some of the literature of conflicting results.
Thoughts, Jason? I’d be interested in your perspective on the whole issue.
*I wrote this kind of fast, so it may be messy!
You raise a lot of questions! I agree there’s a difference between perceiving/processing similar info as distinct and being able to distinguish similar experiences in memory. I think they both involve pattern separation and do point to the fact that this is a process that is subserved by many areas of the brain. There’s also the perspective that pattern separation is a circuit phenomenon that has to do with taking similar inputs and making their representation more distinct. In this sense, the behavior experiments out there do not necessarily reflect pattern separation ability (though they may) – an animal may fail to behaviorally discriminate without having any problem being able to distinguish two things in memory – they may simply choose to treat them as the same…
PS- good article here if you haven’t read it: http://www.ncbi.nlm.nih.gov/pubmed/21460835
Increasing adult hippocampal neurogenesis is sufficient to improve pattern separation.
Nature, 2011
Another goodie:
Changes in Task Demands Alter the Pattern of zif268 Expression in the Dentate Gyrus.
Satvat E, Schmidt B, Argraves M, Marrone DF, Markus EJ.
J Neurosci. 2011 May 11;31(19):7163-7.
http://www.ncbi.nlm.nih.gov/pubmed/21562279
You’ve probably read this one already. I mean, I mentioned the last two papers the day they were released, and you’d already heard of them! It’s hard to keep pace with you.
Anyway, this paper is interesting. It provides some evidence for responses of select granule cell populations depending on the demands of the task in the same environment. It still doesn’t explain the failure (so far, IMO) to demonstrate selective granule cell activation in different environments, but it’s definitely worth a look. I like this paper but still feel the Alme et al 2010 paper should not be so overlooked.
Reading your blog is giving me an upgrade on my knowledge. I wonder where the asymptote will be.
As long as you stay on this site there will be no asymptote
Pingback: Восстанавливаются ли всё же нервные клетки? Часть 2 — Нейроновости