Monday, February 14, 2011

Hypothesis and Evidence

Thoughts on hypothesis, evidence and science.

It was a Sunday afternoon or was it evening? Not quite sure of the time element but I do recall the dent on the rear fender of the car. It was a small indentation, something that I thought might have been the outcome of a parking lot minor mishap. But it gnawed at me for a while and so I decided to investigate it. That little journey took me into the realm of “evidence,” as proof began to reveal itself. Slowly in stepladder fashion, I climbed my way to a new understanding. The hypothesis of the mishap was correct but when and where?

So what is evidence really? Is it a thing, an object, a thought or an imagined corollary without permanence of form in material? Interesting wouldn’t you say? It does have that mind-jangling, asymptotic, confusing ring to it.

Philosophy of Evidence:
Evidence, it appears, is a philosophical construct. Now lets not get too far ahead and collapse this argument by closing our minds. Seriously, stay with me on this one. Evidence can be a bloody knife, a set of fingerprints, a historical construct based on previously accepted principles, a color of an object, a sound wave, a mentioned reasoning in a book, the smell of a plant or a golf glove. You see all these so-called evidences can be used in science, law, biology, medicine, engineering and related fields. So “evidence” by definition has no limitations on the type of construct. It is an informational piece of relevance that is reasoned to agree or disagree with a hypothesis.

So it becomes clear that evidence, is a secondary phenomenon, required to prove a hypothesis. Without one of these you cannot prove one of those. But to generate a hypothesis one has to have a thought construct based on some relevant information that allows you to undertake such an enterprise, doesn’t it?

Carl Frederich Gauss
Mathematician
So evidence can be gained prior to the hypothesis being developed but here in lies the poverty in all scientific disciplines. If one finds certain data and develops a hypothesis to fit the data then one has committed the act of “observer bias.” For, there the question remains, was the hypothesis just a ruse based on the data accidentally at hand and thus a conjecture made and arrived at by forcing a square peg in a round hole, or in some cases a round peg into a round hole. Or is it simply like shaking a tree and the apples fall down thus making us declare that it is an apple tree? There hides a lot of intent in these revelations.

Let us test a hypothesis then to clarify the subject; Man can fly. We both know that is impossible. Yet given the evidence that exists today with thousands of aircraft flights man does indeed hurtle through space albeit in an aircraft designed for such purpose. Yet he himself cannot fly. So the evidence seems to prove the hypothesis that man can fly, yet it does not quite prove that. Does it? Another corollary would be if someone postulated that clouds occur as a result of rain. Well here the reverse is true but the premise postulated is obviously in peril since our observations hold that clouds occur without rain all the time.

Similarly in a criminal case a bloody knife found near the victim does not convict the perpetrator of the crime, unless his fingerprints are found on the knife. Even then the evidence is limited since the context of the fingerprints’ presence on the knife has to be taken into account; the real perpetrator may have stolen someone else’s knife. And on and on the lawyers pick through the haystack to get at the needle.

Thomas Bayes, Mathematician
Evidence supporting a hypothesis is usually static in time. It is the governing principle of that day, week, month, year, decade or even sometimes a century. Today’s evidence may not be relevant for tomorrow and the hypothesis from today may not carry weight the next month let alone the next hour given new evidentiary information. Thus a true proof of a hypothesis is the epistemological data from the past and present as it relates to the truths, beliefs and justifications.

In evidentiary proof of a given hypothesis, rational explanations or justifications are not a guarantee of being correct. It may be rational to think that “reason to believe” and “evidence” are synonymous although the former is a mental construct and the latter may reference material substance. Thus “evidence” which justifies belief may be rational but not proof for a given hypothesis. Similarly being mistaken and doggedly representing reasoned rationality does not determine qualified proof. In the book “Proofiness,” Charles Seife mentions, “Hamlet Evaluation System” utilized by the U.S. Military to justify the purported sham victory in Vietnam that later became labeled as the “Five-o’clock Follies” for their misrepresentation of the actual events.

The term used for judicious and careful thinking and scrupulously obtaining evidence is “Cognitive Prosperity” which implies the holding of large number of true beliefs and relatively few false beliefs is used for individuals who use tedious and time-labored endeavors in uncovering the proofs for a given hypotheses.

"I thought we should require physical determinations, and not abstract integrations. As pernicious taste begins to infiltrate, from which real science will suffer far more than it will progress, and it would be often better for the true physics if there were no mathematics in the world.”
Daniel Bernoulli
Mathematical Hypotheses:
The use of the Probability calculus as the key to all evidence is glazed over by the sweet syrup of statistical nuances whereby manipulation of the numbers can relate proof to some hypothesis. Looking at the Probability of Confidence Intervals in a Gaussian distribution suggests that a 95% probability indicates a defacto proof for any proposed hypothesis, especially in medicine and that the remaining 5% (2.5+2,5) on either side of this normal distribution can be considered outlier and therefore not representative of the factual evidence. Not only this thinking obscures the truth, it also misrepresents the entire evidence. The real truth in mathematical modeling comes from the “law of large numbers” where the “Normal Distribution” will give a 99.7% confidence interval at the 3rd standard deviation, which one might argue is still not 100% and therefore not the real “truth.” In other words the larger the “x” in N=(x) the better the evidence to convict the hypothesis.

Daniel Bernoulli
The rampant use of Bayesian rules holds that it is reasonable for one to believe that the evidence to which one is exposed is related to one’s prior probability distribution. So in fact that if one has had previous encounters, the relatedness to those encounters and the probability derived from them can lay the foundation for the future evidentiary proofs for hypotheses. Now any reasonable, thinking and fair-minded person would vociferously object to that line of reasoning. Yet it is in use today and accepted widely.
“What is Creditable… is not the mere belief in this or that, but the having arrived at it by a process which, had the evidence been different, would have carried one with equal readiness to a contrary belief.”
Blanshard, Reason and Belief

Proving a hypothesis thus must avail of all evidentiary information both pro and con before being labeled as fact. “Evidence thus must become a neutral arbiter of all proposed theoretical commitments including axioms and hypotheses”.

The objectivity of science is secured by the shared relevant evidence conceived, discovered or brought forth in the discovery process prior to rendering a hypothesis as a fact.

Astronomical Hypotheses:
Claudius Ptolemy
Galileo Galilei
The Geocentric theory espoused by the Greek Astronomers for centuries and explained by Claudius Ptolemy (c90-148AD) was rendered moot by Galileo Galilei (1564-1642AD), who proposed the Heliocentric theory, championing the Copernicunism of the day, that still survives today, based on a telescopic evidence of the movement of the planets and the stationary status of the sun. The Ptolemaic arguments were based on a complex mathematical modeling derived from visual evidence only. Of course the heliocentrism is now in slumber mode since the knowledge that the solar system revolves around the galactic center and the galaxies roar by at light speed to the far edges of space in an ever-expanding universe ever increasing in distance from their neighbors. This expansion has been in progress since the Big Bang.

Arno Penzias
The serendipitous discovery of the (CMB) Cosmic Microwave Background Radiation by Arno Penzias and Robert Wilson in 1964 at Princeton, NJ was the first evidentiary proof of the Big Bang. When all possible reasons for the cosmic microwave chatter received via the radio telescope including “bird-droppings” had been considered and rejected, only then the hypothesis was generated and this information rendered as proof since the CMB was universal in its presence. So the Big Bang did happen based on the CMB.

Medical Hypotheses:
Ignaz Philipp Semmelweis
Ignaz Philipp Semmelweis (1818-1865) a Hungarian physician described as the “savior of mothers” imposed strict hand-washing disinfection with chlorinated lime in obstetrical clinics to prevent “puerperal fevers.” He did not have proof of the germs but intuitively knew that “something” was causing the fever that could be prevented with this simple solution. Preventing maternal death, it did!

Edward Anthony Jenner (1749-1823) inoculated an 8 year-old child named James Phipps in 1796 with Cowpox blister material to successfully vaccinate him against Smallpox. There are sketchy references that other non-physicians had utilized this vaccinating procedure before him. The hypothesis therefore existed and Edward Jenner was the first to lay the scientific proof as the foundation of vaccination against smallpox and earning the endearing term, “father of vaccination.”

Barry James Marshall
Barry James Marshall (September 1951-) from University of Western Australia proved that Helicobacter Pylori was the causative agent for peptic ulcers. Hitherto the reasoning had been that stress, spicy foods and excess acidity were the leading causes. However he proved his premise after a monumental struggle against the established rigidity of embedded medical reasoning, by drinking water teeming with H.Pylori followed by an endoscopic evaluation that showed the development of ulceration in his stomach and then treated it with an antibiotic to cure the illness without repercussions all without the aid of antacids. Dr. Marshall received a Nobel Prize in Medicine for his efforts.
Marie Curie

Marie Sklodowska Curie (1867-1934) the first recipient of two Nobel Prizes in Physics and Chemistry discovered radioactivity from Radon. She, based on her inspiration established the use of radiation therapy for cancers. The premise of radioactivity and its potential use to benefit humans was her hypothesis and when the therapeutic use was undertaken, it showed success in patients, proving the hypothesis with proof.

Germ Theory:
Robert Koch
In spite of greats like Girolamo Fracastoro, Agostino Bassi, Anton van-Leeuwenhoek and Ignaz Semmelweis, the first individual to formulate that germs were the cause of illness was John Snow. He discovered that Cholera outbreak in Soho London in 1854. Filtering and boiling it to prevent further spread of the illness proved the hypothesis of dirty water, but it did not get proven until Koch’s Postulate was established that Vibrio Cholerae, the organism was the proximate cause of the illness.
The incredible diligence in the yesteryears towards the etiology of disease has slowly disintegrated into mathematical probability events. Where once absolute proof was a requirement now probability suffices. Cases in point are the various “Epidemiological proofs” that are arrived at with incongruous data. For instance there were initial studies that maintained that carbohydrate diets were heart healthy, finding that a high carbohydrate diet leads to high triglycerides led to mathematical modeling to suggest increased protein was better, except if one’s kidneys cannot process the nitrogen from the proteins, and now we find ourselves with the old commonsensical grandma’s tenet that less is better and fruits and vegetables are extremely important for health. There are further proofs that ingestion and retention of excess fat (obesity), smoking and alcohol usage are deleterious to human health. The changeable thinking shows the gravitas placed on cigarettes, when in 1893, smoking was actually promoted as a treatment for bronchitis. Now we know that it is a curse of/on civilization. We laugh at that original therapeutic premise, yet it was a practical suggestion based on faulty reasoning then. Smoking temporarily suppressed the coughing induced by bronchitis, therefore it was hailed as a panacea.

Genetics:
Barbara McClintock
It was once postulated not to distant in the past when the human genome project was completed that disease especially cancer would be cured since discovery of the 25,000 active genes allowed us to peer into the machinations of what makes humans tick. The one-gene-one-disease concept, since has been discarded as information regarding epi-genetics and inter-genetic workings of the DNA helix has come to light. Genes work through varying degrees of modulation under the governing of other molecules like the interference RNA (miRNA). These agents can over-express, or suppress a gene’s function. Not only does the human DNA play musical chairs to Barbara McClintock’s “Jumping Genes” it also can through the various SNPs (Single Nucleotide Polymorphisms) that occur randomly allow the process of evolution for survival through making us “fitter.” These changes have unintended consequences. Sure we have the DNA mismatch repair mechanisms to protect ourselves, but if inundated with errors, some can pass through and wreak havoc on the body. This relative change on the gene makeup can shut off the protein production that gene produces  conversely it can allow production in excess amounts. The varying degrees of function thus changes the growth potential of the cell in which this corrupted influences is being waged. A wayward cell with a renegade genetic mechanism is the hallmark of cancer. The hypothesis that knowing the genetic makeup will yield cure for cancer remains unfulfilled. More evidentiary information is needed. And more is to come.

Unintended Consequences:
MRSA bacteria
The rampant use of antibiotics is creating a dangerous scenario that finds the pharmaceutical industry on its heels to find newer and newer generations of medicines. This ambitious project is governed by both the desire to have saleable product and also because the organisms are evolutionally quite expedient in their mutational edict to escape the death threat imposed by such agents. This broad and generous use of antibiotics has caused the emergence of newer strains of bacteria eg. MRSA (Methicillin Resistant Staphylococcus Aureus), VRE (Vancomycin Resistant Enterococcus), Resistant strains of TB (Tuberculosis) and AZT resistant strains of HIV. Additionally the constantly mutating and evolving Hepatitis B virus ten years ago had 4 sub genotypes now the HBV has evolved into eight, due to selection pressures imposed upon it by vaccination. More importantly a new mutant Poliovirus was recently discovered in Africa that has caused over 450 deaths in children.

Poliovirus
Our understanding of the germ theory has taken on a whole different premise. Having fought the battle against the germs with products that allow humanity to survive has brought consequences of a new emergence of evolutionally mutated agents with newer tools to fight against us. The population of the viruses is in the trillions and humanity is in the billions. Victory in any battle belongs to the agile, constantly evolving and deceptive force.

Evidence based Medicine:
“Evidence based Medicine” is a terminology that has found a haven in medical circles. The intent in the statement is to provide proof for all the guidelines and dicta invoked in the premise of “how-to and what-to” in practicing the art of medicine. Those guides and dicta become concrete with the use of the term “evidence based medicine” and when used in that manner, no questions may be asked. It is a method by some to stifle scientific thought and therefore progress. Medicine is constantly changing because the environment changes too. To place a burden on the practice of medicine with past tense information is deliberately preventing the use of newer methodologies for the betterment of human health. The failure of this logic is in evidence throughout history. Agility and changeable thinking with newer hypotheses and newer proofs will lead to a better outcome. Today is in flux due to the understanding of the past and the unknowns of a hypothetical future.

… Oh, and not forgetting the minor mishap that started this whole thinking, the evidence of a bluish color present in the dented fender matched with a child’s blue colored metal tricycle. Faced with irrefutable evidence of matched colors between the dent and the bike Junior had nowhere to hide. QED.


Parvez Dara, MD FACP
Twitter: JediPD

References:

1. Kelly, Thomas, "Evidence", The Stanford Encyclopedia of Philosophy (Fall 2008 Edition), Edward N. Zalta (ed.)
2. Blanshard, Brand (1974). Reason and Belief. (London: Allen and Unwin).
3. Carnap, Rudolf (1950). Logical Foundations of Probability (Chicago: University of Chicago Press).
4. Carnap, Rudolf (1966). An Introduction to the Philosophy of Science. Edited by Martin Gardner. (New York: Dover).
5. Fisher, R.A. (1930). ‘Inverse Probability’, Proceedings of the Cambridge Philosophical Society 26(4): 528-535.
6. Goldman, Alvin (1979). ‘What Is Justified Belief?’ in George Pappas (ed.). Justification and Knowledge (Dordrecht: Reidel Publishing Company): 1-23.
7. Horwich, Paul (1982). Probability and Evidence (Cambridge: Cambridge University Press).
8. Silins, Nico (2005). ‘Deception and Evidence’, in Philosophical Perspective, vol.19: Epistemology (Malden, MA: Blackwell Publishers)
9. Williamson, Timothy (2000). Knowledge and Its Limits. (Oxford: Oxford University Press).
10. Seife, Charles Proofiness, The dark art of Mathematical Deception,  pub Viking Press 2010.

No comments: