Sunday, August 4, 2013

Weekly Brain Update: Deciding and Disseminating

Written early in the week of July 29…

In the July 24 issue of the Journal of Neuroscience I saw several articles that piqued my interest.  The first article, in fact, was a manifesto urging neuroscientists to be more active in communicating with the general public about work in this field.  While the primary goal in doing these posts has been to motivate myself to read the Journal every week, one byproduct of this is a public dissemination of information.  Hopefully it is acceptable to define “public” as the 5 or 6 of you who read this blog!
There were a number of articles that were focused on the generation of new neurons in the brain.  Many studies have demonstrated that learning is associated with an increase in the production of new neurons in the dentate gyrus of the hippocampus, one of two places in the adult brain where new neurons are made (see below for the other!).  This week’s study by Haditsch and colleagues showed that the increased production of new neurons caused by learning a new task was due to more precursor cells being generated, as opposed to the greater survival of already-formed new cells, which has been the predominant view of how this works.  Furthermore, they showed that the signal to increase the production of precursor cells was dependent on signals from the forebrain.  Another article about new cells was by Tailor and colleagues, who obtained hindbrain neuroepithelial stem cells from 5-7 week human embryos (a controversial source, for certain) and showed that the cells could be stimulated to multiply into functional neurons that would work as part of a circuit in the cerebellum. With so many neural diseases in which specific types of cells degenerate, cell replacement would be a valuable therapeutic tool.  It’s still science fiction now, but this type of research is making some progress.  Speaking of degenerative diseases, the third new-neuron article was about an animal model of Alzheimer’s Disease in which the authors, led by Cheng, examined the function of olfactory sensory neurons that had been temporarily infected with humanized mutated amyloid precursor protein, which is thought to be the main culprit in the formation of neural plaques.  Olfactory sensory neurons are the second type of neuron that gets regenerated throughout adulthood.  Every three weeks your olfactory sensory neurons are getting replaced by new ones, and so they have to constantly reestablish connections in the olfactory bulb and from there, to areas in the cortex.  It’s like a new circuit gets built continuously.  One of the earliest signs of Alzheimer’s is an altered sense of smell, so it might be possible to use this system as a diagnostic tool, but also as a model in animals for testing the effects of treatments that are meant to delay or prevent the disease.  Cheng and colleagues were able to show that if they decreased the expression of the mutant amyloid protein, the cells could reestablish their connections and the mouse could smell normally again.  Researchers usually use aged rodents in order to examine disease progression, but this olfactory system provides a faster alternative.
I have unwittingly challenged myself a bit too much with the article I chose for this week’s post.  I decided to read this particular article because the abstract seemed to be written in a new language, which looked to me like a combination of classic behavioral terms and computational neuroscience.  I felt fairly confident that with enough concentration and contemplation I would understand the research enough to relate it to knitting and describe it to you.  Well, we’ll see if I can make sense of it. 

August 4, after trying all week to digest the article by Liljeholm and colleagues

Liljeholm and colleagues, from the Computational and Neural Systems Program at California Institute of Technology, examined choice behaviors in humans while they were being scanned for activity in specific brain regions.  There’s been a lot of research looking at choice behavior in humans, some of which has been described here in previous posts, but this study was unique in that they tested the degree to which people perceived differences in how likely it was that their choice would yield an appealing outcome.  Of course, they used food as the goal.  A person would have to choose between a banana and Milano cookies, but would need to depend on previous experience to know how likely would their choice actually yield the desired outcome.  If all was equal, you and I would choose the Milanos, but if the chance of actually getting the Milanos was very small, we might instead decide to have the banana, if it was more likely that our choice would be rewarded.  They had a long list of tasty treats in the experiment, including Godiva dark chocolate bars and peanut M&Ms.  

The authors were interested in determining what brain area was necessary to make the comparison of possible outcomes, with the idea that every decision we make is a complex computation of the expected immediate and long-term impacts of the decision, as well as how the situation affects the outcome of our actions.  They called this a “cognitive map” of goal-directed behavior.  In the choice above, it’s not just about how much tastier Milanos are compared to the banana, but when, where and under what conditions the decision is being made.  If it’s 7 a.m. and I’m well-rested and staring at a picture on the refrigerator of me at the age of 24, I will probably choose the banana.  However, I am more likely to choose the Milanos under any other circumstance.

Honestly, I understood very little about how the authors manipulated the cognitive maps of their participants, but I can tell you that they demonstrated that the anterior portion of the supramarginal gyrus of the inferior lobule of the parietal cortex is the hot spot for comparison of potential outcomes of a behavior.  I can also say that they confirmed the role of the medial prefrontal cortex in computing the expected value of the items (Milanos are far superior to bananas, everyone knows).  Their article deserves a much better description than that, but it’s going to take more time and additional background reading to make that possible. 

Here’s how I believe this relates to knitting:
Developing a pattern for mitts

The fair isle sweater is resting in the Pretend Yarn Store

The Albers Pullover (from Interweave Knits summer 2013) 10 rows from being done.
How do I decide which project to work on, the mitts, the fair isle sweater, or the Albers Pullover?  Based only on the appeal of the project, I would choose the fair isle sweater.  However, the cognitive map of my goal directed behavior is more complex than that.  It includes considerations such as: "there are only 10 rows left on the Albers Pullover" and "I need to write the pattern for the mitts in time to test it before the Camel Knitters need to use it" and "the weather is too warm to wear the Albers Pullover, even though it is meant to be a summer sweater" and "I have done enough of the fair isle sweater to know that the pattern is working".  This kept the inferior lobule of my parietal cortex busy and I'll let you know what that activity yielded in the next post.

1 comment:

Oogyknitter said...

Neat stuff, Neuro! Milanos vs. bananas is almost like qiviut vs. acrylic - a definite no-brainer! (haha). Great articles too - thanks for helping to spread cool science news around!