Science: Muon, Neanderthals, IBM Project Debater

Attending – Barbara, Howard, Bruce and Bob

This was a disturbingly small meeting, although no one was to blame. I believe that we all
enjoyed ourselves; and we had time to discuss scientific issues in depth. Because of the
small attendance, this a shorter summary than usual. At least the Ingenuity helicopter lived
up to expectations.

Barbara spoke first and showed material from two Scientific American papers. The first
concerned the inner ears of Neanderthals, which can be studied in detail by CT scans of
skulls, even those thousands of years old. From such scans, it is possible to construct
educated guesses about the hearing range of any animal. Those scans showed no similarity
of the inner ears of chimpanzees and ancient hominids to those of Neanderthals and
modern humans, which are close together. This makes it likely that Neanderthals soke,
even though presumed differences in the larynx suggested in the past that Neanderthals
didn’t speak, consistent with a general tendency to underestimate Neanderthal function.
Her second paper concerned brain organoids formed in culture from human stem cells
compared with organoids grown from other stem cells in which the important brain gene
NOVA1, which differs by one base pair between humans and Neanderthals, is replaced by
the Neanderthal form of NOVA1, using CRISPR techniques. It takes months to grow the pea-
sized organoids in culture, Those with the Neanderthal form of NOVA1 differ in appearance
and electrical behavior. It’s very difficult to know what to make of this. Many genes differ
between modern humans and Neanderthals, and we know that sensory input has a
profound effect on the organization of the growing mammalian brain. Organoids are
deprived of all sensory input from the beginning.

Howard then spoke of the much publicized aberrations found in detailed studies of the
behavior of muons in two major centers. As he stressed, that these detailed studies of such
a transient particle are done and agree in different labs is amazing, because the average
half-life of the muon is only 2.2 useconds (millionths of a second) and yet FermiLab now
and the Brookhaven National laboratory in 2001 both obtained similar anomalous
magnetic moments for the muon, anomalous because significantly different than predicted
by the standard model. Muons are elementary particles similar to the electron, with an
electric charge of −1 e and a spin of 1/2, but with a much greater mass , Muons, electrons
and the tau particle make up the family of leptons, which are not known to contain smaller
particles. Some subatomic particles have even shorter lives, the top quark being the
shortest. We’ll have to wait for a third lab to conform this anomalous behavior; if confirmed
it will not mean that the whole realm of subatomic particles is turned upside down,
important though it is.

Bob then spoke about the IBM Project Debater, begun in 2012, which has debated 5
premier human debaters in the last 2 ½ years. Conceived as a sequel to Deep Blue amd
Watson, machines that convincingly defeated the best humans at chess and Jeopardy,
Project Debater faced a more open ended challenge. The four main modules are: argument
mining, an argument knowledge base (AKB), argument rebuttal and debate construction. The
system has not been able to beat a human debater so far, but has performed surprisingly well.
Because of the open-ended nature of debates- debaters can’t invent words, but they may combine words and emotion in unusual ways, I expect human debaters to remain supreme for many years.

You can access one debate on YouTube and see what you make of it,
Our next meeting will be Monday, May 10, at 4 PM. Here are two scary stories, at least scary for
me:
Story 1
Story 2

Blogs Topics Posts about this Topic