Author: jlolds

Native Angeleno, professor of Neuroscience and Public Policy at George Mason University

Another academic year

We are on the cusp of commencement here–it’s 12 days away–and yet another academic year is about to end. I did a walk through of the building with my associate director, Ken DeJong, this morning and was astounded to see some of the places that I rarely frequent. There was a time when I knew where everything was–we’re too large for that now, so there are funny surprises for me, as well as our students and faculty who have been here for less time.

On thinking back, this has been a fine year for Mason’s Institute for Advanced Study, in spite of the much vaunted Great Recession. I continue to be impressed by our incredibly bright trainees, our productive faculty and the general willingness of all to reach out across their disciplinary boundaries to new colleagues, all while maintaining a vibrant collegiality.
That and the extant willingness to take on scientific risk, makes this a dream job.
This weekend, I went to the ballpark and watched as the hapless Nationals easily disposed of the St. Louis Cardinals. It seemed an apt metaphor for those of us living here, in proximity to Jefferson’s and Lincoln’s monuments: in the worst of times come the most pleasant of surprises.
Greenshoots,
Jim

Old rivalry–Amherst vs. Williams

I couldn’t help but smile at this mention of my Alma Mater’s ancient grudge match with Williams College reported in today’s Chronicle of Higher-Education on-line.

For those who can’t get behind the fire-wall, money quote:

A century and a half after an Amherst College senior, James L. Claflin, challenged the boys at nearby Williams College to a game of Massachusetts-rules “base ball,” the two colleges will square off once again on Sunday, marking the 150th anniversary of the first college baseball game ever played.

Swine flu–maybe not so bad

From Clive Cookson’s science blog over at Financial Times….

Money quote:

The Mexican strain currently lacks some of the molecular characteristics associated with the most virulent viruses – adding to the emerging epidemiological evidence that it causes mainly mild illness. Experts believe the apparently high mortality rate in Mexico is due to the result of vast under-reporting of less severe cases.

Neuroscience and education

Over at the Wall St. Journal, a review of Daniel T. Willingham’s new book, Why Don’t Student’s Like School.


Money quote:

So why don’t students like school? According to Mr. Willingham, one major reason is that what school requires students to do — think abstractly — is in fact not something our brains are designed to be good at or to enjoy. When we confront a task that requires us to exert mental effort, it is critical that the task be just difficult enough to hold our interest but not so difficult that we give up in frustration. When this balance is struck, it is actually pleasurable to focus the mind for long periods of time. For an example, just watch a person beavering away at a crossword or playing chess in a noisy public park. But schoolwork and classroom time rarely keep students’ minds in this state of “flow” for long. The result is boredom and displeasure. The challenge, for the teacher, is to design lessons and exercises that will maximize interest and attention and thus make students like school at least a bit more.

Experiment planning

I was having one of my regular meetings with one of our junior faculty members in Molecular Neuroscience this morning and we started talking about how to prove causality in some intriguing hippocampal physiology data.

And that’s when we both launched into a very common technique that neuroscientists often use when designing a crucial experiment. I like to call it the “god-like gedanken experiment”. Basically it involves creating a thought experiment in which all of the methodological constraints are removed–as are the need to think about controls.
So for example, if we we thinking about the phenomenon of lightening preceding thunder, and we wanted to get at the question of causality–i.e. does lightening actually cause thunder–we would create a thought experiment in which we would artificially induce a lightening bolt and attempt to cause thunder.
Obviously there are practical problems in creating wild-type lightening bolts (at least as far as I know, not being particularly up to date in the weather engineering business).
And clearly in this case, the thought experiment doesn’t really advance science very far–the phenomenon of lightening having been well studied since at least the time of Dr. Franklin.
But this type of thought experiment can be of great utility for bench-top bioscientists, because it facilitates the mental process of stripping the phenomenon being studied down to its mechanistic parts.
When we don’t have to worry about how a drug might cross the blood-brain barrier or how to engineer a reporter gene into a particular type of neuron in mice of a particular age, then we can focus on concentrating instead solely upon the biological process. This process is, from the standpoint of experimental design, a series of mechanistic steps, which the scientist is attempting to reveal through experimental measurements.
So first, the god-like gedanken experiment, then worry about the practicalities and the controls.
Jim

The fMRI analysis story continued

From the Neuroskeptic blog an interesting post on Kriegeskorte et al. in Nature Neurosciences….once again the very difficult problems inherent in doing statistics on this type of data.

Money quote:

But it would be wrong to think that this is a problem with fMRI alone, or even neuroimaging alone. Any neuroscience experiment in which a large amount of data is collected and only some of it makes it into the final analysis is equally at risk. For example, many neuroscientists use electrodes to record the electrical activity in the brain. It’s increasingly common to use not just one electrode but a whole array of them to record activity from more than brain one cell at once. This is a very powerful technique, but it raises the risk the non-independence error, because there is a temptation to only analyze the data from those electrodes where there is the “right signal”, as the author’s point out:

In single-cell recording, for example, it is common to select neurons according to some criterion (for example, visual responsiveness or selectivity) before applying
further analyses to the selected subset. If the selection is based on the same dataset as is used for selective analysis, biases will arise for any statistic not inherently independent of the selection criterion.

More on the modern university

From this morning’s NY Times–the chair of the religion department at Columbia University rails against US graduate education, which he likens to the US auto industry in terms of its desperate need for restructuring.

I’m not at all sure I agree with him, but then the situation for science is very different than the situation with humanities and especially departments of religion.
For one thing, doctoral students in the sciences tend to receive higher stipends, they produce dissertations which are separable easily into journal papers and their job prospects aren’t bad–especially when you consider non-traditional tracks to academia.
Jim