When you have completed this section, you should be able to
• describe the inductive and hypothetico-deductive methods of obtaining scientific knowledge;
• describe some aspects of experimental design that help to ensure objective and reliable results; and
• explain what is meant by hypothesis, fact, law, and theory in science.
Prior to the seventeenth century, science was done in a haphazard way by a small number of isolated individuals. The philosophers Francis Bacon (1561-1626) in England and René Descartes (1596-1650) in France envisioned science as a far greater, systematic enterprise with enormous possibilities for human health and welfare. They detested those who endlessly debated ancient philosophy without creating anything new. Bacon argued against biased thinking and for more objectivity in science. He outlined a systematic way of seeking similarities, differences, and trends in nature and drawing useful generalizations from observable facts. You will see echoes of Bacon's philosophy in the discussion of scientific method that follows.
Though the followers of Bacon and Descartes argued bitterly with each other, both men wanted science to become a public, cooperative enterprise, supported by governments and conducted by an international community of scholars rather than a few isolated amateurs. Inspired by their vision, the French and English governments established academies of science that still flourish today. Bacon and Descartes are credited with putting science on the path to modernity, not by discovering anything new in nature or inventing any techniques—for neither man was a scientist—but by inventing new habits of scientific thought.
When we say "scientific," we mean that such thinking is based on assumptions and methods that yield reliable, objective, testable information about nature. The assumptions of science are ideas that have proven fruitful in the past—for example, the idea that natural phenomena have natural causes and nature is therefore predictable and understandable. The methods of science are highly variable. Scientific method refers less to observational procedures than to certain habits of disciplined creativity, careful observation, logical thinking, and honest analysis of one's observations and conclusions. It is especially important in health science to understand these habits. This field is littered with more fads and frauds than any other. We are called upon constantly to judge which claims are trustworthy and which are bogus. To make such judgments depends on an appreciation of how scientists think, how they set standards for truth, and why their claims are more reliable than others.
Chapter 1 Major Themes of Anatomy and Physiology 7
The inductive method, first prescribed by Bacon, is a process of making numerous observations until one feels confident in drawing generalizations and predictions from them. What we know of anatomy is a product of the inductive method. We describe the normal structure of the body based on observations of many bodies.
This raises the issue of what is considered proof in science. We can never prove a claim beyond all possible refutation. We can, however, consider a statement as proven beyond reasonable doubt if it was arrived at by reliable methods of observation, tested and confirmed repeatedly, and not falsified by any credible observation. In science, all truth is tentative; there is no room for dogma. We must always be prepared to abandon yesterday's truth if tomorrow's facts disprove it.
Most physiological knowledge was obtained by the hypothetico-deductive method. An investigator begins by asking a question and formulating a hypothesis—an educated speculation or possible answer to the question. A good hypothesis must be (1) consistent with what is already known and (2) capable of being tested and possibly falsified by evidence. Falsifiability means that if we claim something is scientifically true, we must be able to specify what evidence it would take to prove it wrong. If nothing could possibly prove it wrong, then it is not scientific.
_Think About It_
The ancients thought that gods or invisible demons caused epilepsy. Today, epileptic seizures are attributed to bursts of abnormal electrical activity in nerve cells of the brain. Explain why one of these claims is falsifiable (and thus scientific), while the other claim is not.
The purpose of a hypothesis is to suggest a method for answering a question. From the hypothesis, a researcher makes a deduction, typically in the form of an "if-then" prediction: If my hypothesis on epilepsy is correct and I record the brain waves of patients during seizures, then I should observe abnormal bursts of activity. A properly conducted experiment yields observations that either support a hypothesis or require the scientist to modify or abandon it, formulate a better hypothesis, and test that one. Hypothesis testing operates in cycles of conjecture and disproof until one is found that is supported by the evidence.
Doing an experiment properly involves several important considerations. What shall I measure and how can I
Saladin: Anatomy & I 1. Major Themes of I Text I I © The McGraw-Hill
Physiology: The Unity of Anatomy and Physiology Companies, 2003 Form and Function, Third Edition
8 Part One Organization of the Body measure it? What effects should I watch for and which ones should I ignore? How can I be sure that my results are due to the factors (variables) that I manipulate and not due to something else? When working on human subjects, how can I prevent the subject's expectations or state of mind from influencing the results? Most importantly, how can I eliminate my own biases and be sure that even the most skeptical critics will have as much confidence in my conclusions as I do? Several elements of experimental design address these issues:
• Sample size. The number of subjects (animals or people) used in a study is the sample size. An adequate sample size controls for chance events and individual variations in response and thus enables us to place more confidence in the outcome. For example, would you rather trust your health to a drug that was tested on 5 people or one tested on 5,000?
• Controls. Biomedical experiments require comparison between treated and untreated individuals so that we can judge whether the treatment has any effect. A control group consists of subjects that are as much like the treatment group as possible except with respect to the variable being tested. For example, there is evidence that garlic lowers blood cholesterol levels. In one study, a group of people with high cholesterol was given 800 mg of garlic powder daily for 4 months and exhibited an average 12% reduction in cholesterol. Was this a significant reduction, and was it due to the garlic? It is impossible to say without comparison to a control group of similar people who received no treatment. In this study, the control group averaged only a 3% reduction in cholesterol, so garlic seems to have made a difference.
• Psychosomatic effects. Psychosomatic effects (effects of the subject's state of mind on his or her physiology) can have an undesirable impact on experimental results if we do not control for them. In drug research, it is therefore customary to give the control group a placebo (pla-SEE-bo)—a substance with no significant physiological effect on the body. If we were testing a drug, for example, we could give the treatment group the drug and the control group identical-looking starch tablets. Neither group must know which tablets it is receiving. If the two groups showed significantly different effects, we could feel confident that it did not result from a knowledge of what they were taking.
• Experimenter bias. In the competitive, high-stakes world of medical research, experimenters may want certain results so much that their biases, even subconscious ones, can affect their interpretation of the data. One way to control for this is the doubleblind method. In this procedure, neither the subject to whom a treatment is given nor the person giving it and recording the results knows whether that subject is receiving the experimental treatment or placebo. A researcher might prepare identical-looking tablets, some with the drug and some with placebo, label them with code numbers, and distribute them to participating physicians. The physicians themselves do not know whether they are administering drug or placebo, so they cannot give the subjects even accidental hints of which substance they are taking. When the data are collected, the researcher can correlate them with the composition of the tablets and determine whether the drug had more effect than the placebo.
• Statistical testing. If you tossed a coin 100 times, you would expect it to come up about 50 heads and 50 tails. If it actually came up 48:52, you would probably attribute this to random error rather than bias in the coin. But what if it came up 40:60? At what point would you begin to suspect bias? This type of problem is faced routinely in research—how great a difference must there be between control and experimental groups before we feel confident that the treatment really had an effect? What if a treatment group exhibited a 12% reduction in cholesterol level and the placebo group a 10% reduction? Would this be enough to conclude that the treatment was effective? Scientists are well grounded in statistical tests that can be applied to the data. Perhaps you have heard of the chi-square test, the t test, or analysis of variance, for example. A typical outcome of a statistical test might be expressed, "We can be 99.5% sure that the difference between group A and group B was due to the experimental treatment and not to random variation."
When a scientist applies for funds to support a research project or submits results for publication, the application or manuscript is submitted to peer review—a critical evaluation by other experts in that field. Even after a report is published, if the results are important or unconventional, other scientists may attempt to reproduce them to see if the author was correct. At every stage from planning to postpublication, scientists are therefore subject to intense scrutiny by their colleagues. Peer review is one mechanism for ensuring honesty, objectivity, and quality in science.
Facts, Laws, and Theories
The most important product of scientific research is understanding how nature works—whether it be the nature of a pond to an ecologist or the nature of a liver cell to a physiologist. We express our understanding as facts, laws, and theories of nature. It is important to appreciate the differences between these.
Saladin: Anatomy & 1. Major Themes of Text © The McGraw-Hill
Physiology: The Unity of Anatomy and Physiology Companies, 2003
Form and Function, Third Edition
A scientific fact is information that can be independently verified by any trained person—for example, the fact that an iron deficiency leads to anemia. A law of nature is a generalization about the predictable ways in which matter and energy behave. It is the result of inductive reasoning based on repeated, confirmed observations. Some laws are expressed as concise verbal statements, such as the first law of thermodynamics: Energy can be converted from one form to another but cannot be created or destroyed. Others are expressed as mathematical formulae, such as the law of Laplace: F = 2T/r, where F is a force that tends to cause a microscopic air sac of the lung to collapse, T is the surface tension of the fluid lining the sac, and r is the sac's radius.
A theory is an explanatory statement, or set of statements, derived from facts, laws, and confirmed hypotheses. Some theories have names, such as the cell theory, the fluid-mosaic theory of cell membranes, and the sliding filament theory of muscle contraction. Most, however, remain unnamed. The purpose of a theory is not only to concisely summarize what we already know but, moreover, to suggest directions for further study and to help predict what the findings should be if the theory is correct.
Law and theory mean something different in science than they do to most people. In common usage, a law is a rule created and enforced by people; we must obey it or risk a penalty. A law of nature, however, is a description; laws do not govern the universe, they describe it. Laypeople tend to use the word theory for what a scientist would call a hypothesis—for example, "I have a theory why my car won't start." The difference in meaning causes significant confusion when it leads people to think that a scientific theory (such as the theory of evolution) is merely a guess or conjecture, instead of recognizing it as a summary of conclusions drawn from a large body of observed facts. The concepts of gravity and electrons are theories, too, but this does not mean they are merely speculations.
_Think About It_
Was the cell theory proposed by Schleiden and
Schwann more a product of the hypothetico-
deductive method or of the inductive method?
Explain your answer.
Before You Go On
Answer the following questions to test your understanding of the preceding section:
4. Describe the general process involved in the inductive method.
5. Describe some sources of potential bias in biomedical research. What are some ways of minimizing such bias?
6. Is there more information in an individual scientific fact or in a theory? Explain.
Chapter 1 Major Themes of Anatomy and Physiology 9
Was this article helpful?
Did you ever think feeling angry and irritable could be a symptom of constipation? A horrible fullness and pressing sharp pains against the bladders can’t help but affect your mood. Sometimes you just want everyone to leave you alone and sleep to escape the pain. It is virtually impossible to be constipated and keep a sunny disposition. Follow the steps in this guide to alleviate constipation and lead a happier healthy life.