10 Things Our Readers (Scientists) Suspect is true but not yet supported by data.

So long time ago, we asked our readers via mail, Scientists: What are craziest or weirdest things in your field that you suspect is true but is not yet supported fully by data? We got a lot of interesting replies from scientists and researchers. Here are their responses and their respective field of expertise. Most of them have requested to remain anonymous, so their names will not be published.

1. Link between Hormonal Birth Control and Divorces

[Psychology/Biology]

This is kind of out there, but I believe hormonal birth control plays a role in a subset of divorces and in the birth of children with poor immune systems. I think most people are familiar with the t-shirt study that found women were attracted to the smell of men with dissimilar immune systems and that they felt t-shirts with similar immune systems smelled like their father or brother.

Women on hormonal birth control did not perform well in that study – they often picked similar immune systems as more attractive. A massive percentage of women are on birth control during their dating years and when they meet their husbands. After marriage, another percentage of those women will stop taking hormonal birth control in order to have children. What happens then?

If you have a woman who marries a man who is a poor genetic match – someone she may not have wanted to be with if she wasn’t on birth control – when she goes off of her birth control, it stands to reason that her perception of her husband may change. Given that her body and personality will be going through a change while her hormones are re-regulated, her husband’s view of her may change as well. It would not surprise me if this subconscious change contributes to divorce rates. In addition, should those hypothetical people have children, the child may have a poorer immune system due to the lack of genetic variance between its parents.

Ladies, please don’t let this scare you off of your hormonal birth control. The point of this question was to share something weird that isn’t fully supported by data.

2. Cure Blindness

[Ophthalmology]

Within the next 10 years, we will likely finish creating a polymer optoelectric interface that can restore light sensitivity in blind retinas.

In other words, we will be able to cure blindness due to retinal degeneration.

This actually already is on track to working (labs up at Stanford have a version that can operate if you wear a pair of glasses that can supply the interface with energy by actually projecting what you’re seeing onto the retina), but the goal is to make it organic and wireless/without glasses. It’s coming though. I hope.

3. Fusion Power is Achievable

[Physics]

We can, and will eventually, achieve Inertial Confinement Fusion as a cheap and renewable source of energy. Here is the website to the NIF project. I can’t say much more than the website tells, but suffice it to say I’m convinced it will work, and it will work soon.

Inertial confinement fusion is essentially bombarding hydrogen fuel pellets with extremely high power lasers in the hopes of getting net positive energy.

The energy of the lasers causes hydrogen isotopes (deuterium and tritium) to undergo fusion reactions, which should create helium and a neutron. In this reaction, a small amount of mass (about 0.4%) is converted directly into energy (as per E=MC²), which is where the net energy comes from. (yes the helium and neutron results will weigh less than the original deuterium and tritium inputs). None of this is controversial, nor is it a perpetual energy machine. ICE (and most other types of conventional fusion) would use huge amounts of power to “ignite” the fuel, which would theoretically release more energy than you put in. But then the fuel is spent and you need to keep doing it over and over again.

4. Chestnuts

Chestnut blight was introduced to the United States in the early 1900’s and since then, all American chestnuts in much of the U.S. cannot mature properly and will die at a nearly uniform stage in development. Much of the work I did in college was looking at the issue of the blight, my masters took me to propose that the chestnut gall wasp has a direct link to developmental challenges introduced by hormonal changes within the chemistry of C4 pathway photosynthesis. These changes are directly influencing a protein coupling mechanism failure and aiding the blight of the chestnut. My doctoral thesis is showing this without a doubt and will soon present ways to offset the population growth of the gall wasp and will soon allow chestnuts to be regrown naturally. They will then dominate our landscape again as they did early in our history. I have a personal specimen in the Smithsonian credited to my professor and I for being the first ones to document the gall wasp this far north on the east coast.

5. Darwin’s Dilemma

[Paleontology]

“Darwin’s Dilemma” refers to the absence of known macroscopic fossils (i.e. excluding microbial stuff) from beneath the base of the Cambrian period (ie. older than 541 million years) when he was first describing his theory of evolution. Everyone (including Darwin) expected that when fossils were found from before the Cambrian period that they would be the ancestors of modern animals.

Then when those fossils were found (Ediacaran biota we call them, after the Ediacara Hills in South Australia), they were enthusiastically classified as ancient jellyfish, ancient worms, etc.

But it turns out that’s a load of crap. The consensus is growing that almost none of those classifications were accurate. Instead, most Ediacaran organisms seem to have belonged to extinct groups unrelated to anything alive today, and they lived in a world that was so different from the following 540 million years, they had modes of life distinct from any living macroscopic organisms. We were so eager to make everything fit into our neat little evolutionary narrative that for decades we basically glossed over these crazy, alien sort of ecosystems that could be critical to understanding the origins of complex life on other planets as well as our own. We’re still just beginning to really understand it though. It’s a very controversial field of paleontology.

6. Root of Psychiatric Disorders

[Psychiatry]

There was a recent finding that 5 psychiatric disorders share some common genetic mutations related to voltage gated calcium channel subunits in the brain.

Voltage-gated calcium channels play a role in neurotransmitter release, which is why some medications that increase the amount of neurotransmitter in the synapse (such as SSRIs) seem to “cure” some psychiatric disorders for some people.

This also might pose why electroconvulsive shock therapy works for some patients in treating their previously treatment-resistant depression. The shock activates the voltage-gated channels that are affected by the mutation.

Psychiatry is probably one of the least understood areas of medicine. We still have no idea what truly causes mental disorders.

7. Computers will make Human Decisions

[Computer science/machine learning]

There is a reasonable chance that, within the next 10 years, computers that analyze large bodies of data and make statistical predictions will be able to make much better routine strategic decisions at almost every layer of society than actual humans. This means many of the decisions that pertain to economics, politics, etc. At some point this will likely start having a major impact on the structure of society, as there won’t be all that much room for argument when the computer says that some decision is or is not optimal.

Some current signs that point to this, for example, are the story about an automated Target marketing campaign that detected that a teenage girl was pregnant before her own father knew about it. Automated data analysis (when data is available) is simply better at predicting common events than real people. Why then would a corporation, for example, rely on human executives to make decisions about acquisitions and mergers? Why would a politician use human strategists to choose policy positions? And if a politician makes decisions by a computer, why do we need the politician?

This will raise some interesting questions. Should we still use human juries and judges when computers can determine guilt or innocence with much greater accuracy? What happens when a human jury convicts a man, and the innocence project makes an appeal on the basis that the judge-o-matic predicts that the man’s chance of having actually committed the crime is 0.000001%?

Should we have politicians set the tax rate when computers can do it better to optimize both economic growth and government performance?

I would emphasize that this has little to do with “AI” — it’s not intelligence, just prediction. And we’ll have to contend with this issue long before computers become “intelligent” in the conventional sense. It also doesn’t apply to rare, unprecedented events (so the computer won’t tell you whether we’ll invent cold fusion), but it will very much apply to events that have happened before (recessions, political unrest, the effect of taxation on the economy, etc).

8. We don’t understand how reactions actually occur

[Chemistry]

A lot of the accepted mechanisms for the way reactions occur are incomplete at best. Example, water and alkali metals go boom. The accepted mechanism is that the metal dissolves and loses an electron which goes to an H+ to make H which combines with another H to make H2. Then the heat causes the H2 to combust. However recent observations of various colors associated with metal vapor evolving during the reaction suggest the metal may be vaporizing at the interface with water and that this reaction is occurring in the gas phase rather than a simple dissolution of the solid and subsequent electron exchange.

9. Bacteriophage will replace antibiotics

[Medical Microbiology]

I believe very soon (in terms of development of new treatments) that bacteriophages will replace antibiotics. Bacteriophages are essentially viruses that attack bacteria and bacteria only. Some of you might argue, well there’s good bacteria in our digestive system, right? True. Bacteriophages have surface proteins that work like a lock and a key, and can only combine with certain types of bacteria.

Antibiotics often cause many side effects and over the last 10 years have decreased in efficacy due to bacteria becoming resistant to the treatments. Also, antibiotics are delivered to the bloodstream and only happen upon the bacteria. Whereas bacteriophages can reach the bacteria and take over its replication machinery and replicate exponentially (10⁸).

Currently, bacteriophages are a relatively small part of medical research. Many studies are being conducted on the possibility of bacteriophages being utilized concurrently with antibiotics to increase the drug efficacy (mainly by turning off the bacteria’s “defenses”).

10. There genomic database of the entire humankind in the future

[Biology]

When personal genomic sequencing becomes readily accessible, even as a routine medical service (much like a physical), we should be able to place the data into a controlled anonymous database and use statistical analysis to essentially link most disease states and predispositions to their genotypic bases. There are loads of “combinitoric blackholes” in the genome data right now, since we haven’t been able to collect large-scale, high-throughput disease/genotype relationships on a genomic scale yet due to the expense.

Have a look at the company 23andme for an example of a smaller scale version of this. They represent a good first step towards breaking the cost down and making personal genomics accessible regardless of socioeconomic status.