The Importance of Being Stupid in Science

Science articles give the impression that the research is laid out in a logical order and that all the data fit neatly to confirm a hypothesis.  All the false starts are left out by convention.  All the embarrassing lab errors and the data that don’t fit are

The truth is, scientists really don’t know what we’re doing.  We’re wallowing in the unknown, flailing, trying to keep our heads above the muck and trying to find some clear message from the morass.

Most scientists feel stupid most of the time.  It comes with the territory.  The exceptions are people who aren’t really trying to do science, but are engaged in the vast enterprise of collecting the data on which modern science is built.

(This is summarized (with editorial freedom) from an article by Martin Schwartz in the Journal of Cell Science, The Importance of Being Stupid.)

High Quality Stupid Scientist Blank Meme Template

One of the beautiful things about science is that it allows us to bumble along, getting it wrong time after time, and feel perfectly fine as long as we learn something each time…The more comfortable we become with being stupid, the deeper we will wade into the unknown and the more likely we are to make big discoveries.

 

Advertisements

Controlling Computers with our Thoughts

Keyboards and mice are convenient and precise, but slow and especially difficult for people with some kinds of disabilities.  Voice control of computers has advanced greatly in the last decade, but we’re still nowhere near the place where we can talk to Hal the way Dave did.

Most convenient of all would be if we can control computers with our thoughts, the same way we decide to reach out a hand or a foot and our bodies comply with our wishes.  Control of our bodies is mediated by a network of nerves that connect our brains to the extremities and everywhere in between.  Computers don’t have that kind of access to details of the neural activity in our brains.  Most of us would rather not have electrodes implanted in our skulls.  So work has proceeded on how to decode voltage signals on the scalp.

This has been surprisingly effective.  The key to success has been that the brain is wired in a flexible way that facilitates learning.  With feedback from the computer, people learn to create just the electric patterns that the computer is programmed to look for, and to control the computer with their thoughts.

This week there is a report of a more efficient control system that requires no training.  A multinational team based in Japan has been working to get the computer trained how to read the human brain rather than having the human learn how to control the computer.  Preliminary success is encouraging.  The first applications will involve wheelchair control by quadriplegics.

Univ of S Calif Press Release                                             Article in Science Advances

A Rational Approach to Science Funding

The problem with science research today is that everyone wants to fund the next Einstein, and no one wants to fund a thousand crackpots whose ideas will lead only to dead ends—but none among us is smart enough to tell the difference.

We have to give up on the idea that we can manage research the way we manage an efficient business.

We have to give up on the idea that we have a solid foundation or understanding nature’s workings, and the job of scientists is to fill in the details.

— Josh Mitteldorf

image: https://upload.wikimedia.org/wikipedia/commons/thumb/2/20/Nikola_Tesla%2C_with_his_equipment_Wellcome_M0014782.jpg/318px-Nikola_Tesla%2C_with_his_equipment_Wellcome_M0014782.jpg

Nikola Tesla, with his equipment

Biodegrading Plastic Bottles

About 1 million plastic bottles are purchased every minute.  Only 14% of these are recycled worldwide.  The rest end up in landfills or, worse, pollute the ocean.  PET is inert by design, and lasts millions of years.

In 2016, a species of bacteria was genetically engineered to eat plastic.  The bacteria use an enzyme (a biochemical catalyst) that breaks down the plastic and turns it into liquids that can be used like kerosene or gas.

Tweaking the enzyme, a group of University of Portsmouth scientists stumbled on a form that is more efficient than what the bacteria used.  It can break down PET in just a few days.

“It is a modest improvement – 20% better – but that is not the point,” said McGeehan. “It’s incredible because it tells us that the enzyme is not yet optimised. It gives us scope to use all the technology used in other enzyme development for years and years and make a super-fast enzyme.”

Guardian article

4500

Sentience of Trees

Trees do most of the things you do, just more slowly. They compete for their livelihoods and take care of their families, sometimes making huge sacrifices for their children. They breathe, eat and have sex. They give gifts, communicate, learn, remember and record the important events of their lives. With relatives and non-kin alike they cooperate, forming neighborhood watch committees — to name one example — with rapid response networks to alert others to a threatening intruder. They manage their resources in bank accounts, using past market trends to predict future needs. They mine and farm the land, and sometimes move their families across great distances for better opportunities. Some of this might take centuries, but for a creature with a life span of hundreds or thousands of years, time must surely have a different feel about it.

 

 

— Barbara Kingsolver, writing in the NYTimes,
reviewing The Overstory by Richard Powers

 

Bacteria can Learn

Bacteria have individual lives, and also collective lives. They can form films that support a communal existence, protecting one another at a cost in individual autonomy.

UCLA press release

In this study published last month, bacteria remember the surface they were attached to and pass this information along to their offspring. This is learning, combined with inter-generational memory. Bacteria are the smaller kind of one-celled orgnisms, and science has no understanding of how memory is stored. The article doesn’t describe a mechanism. The most likely candidate would be epigenetic. In other words, markers on the DNA that tell what genes to turn on and off are persistent, and can be passed to offsring when the DNA is copied.

James Shapiro has written a book describing how bacteria modify their own DNA. This is full Lamarckian adaptation.