Transcript Uncertainty
Physics 270 – Experimental Physics “The Scientific Method” Science as a Collection of Facts Fact 1 Fact 2 Fact 3 … One possible definition: activities aimed at understanding the natural world Scientists have shared values and perspectives that characterize a scientific approach to understanding nature: a demand for naturalistic explanations supported by empirical evidence that are testable against the natural world. Other shared elements include observations, rational argument, inference, skepticism, peer review and reproducibility of work. Observations of phenomena Experiments Empirical formulas Simplify the phenomena Models by creating a model system on which calculations Laws / Theory Let’s do some experiments! can be carried out to study the phenomena. Develop multiple approaches since you aren’t sure which one will work. Start Goal “This is not a pipe” Painting by Rene Magritte In science results are presented using precise (though technical) arguments, …with… testable consequences falsifiability reproducibility Experimental Verification And Reproducibility “Truth” in science Descriptions of some aspect of nature in terms of a model. Any view of the natural world that a scientist devises is just a model loaded with assumptions and approximations of that world. Models, in general, have limited applicability. As data and technology improve, models are replaced by others which explain a larger range of phenomena. Theory – the best available description of nature – as close to “truth” as we get. Theories are validated by experiments. ◦ There is no “truth-meter” in science. Experiments expose the limitations or incorrectness of theories. Something may only be known if it is proven to be true. Beliefs may be true or false. Rationality is the best test of truth. Our senses can easily be fooled! Reductionism versus Wholism Reduction Reduce a complicated problem into its constituents and aims to understand that complex problem through the study of its components Wholism a phenomena must be viewed as a whole in order to understand its structure Reductionist Example: The Structure of Proteins Proteins consist of amino acids. These are assembled into ribosomes. The order of assembly is determined by RNA after it is copied from DNA. DNA consists of 4 units called nucleotides. The structure of proteins is very complicated, but here the problem has been reduced to the assemblage of simpler building blocks. Holistic Example: An ant hill Complex physical, chemical, and biological structure built and sustained by millions of ants. Cannot be understood by braking the ants into tiny parts. Its essence is in the complexity of the whole. Deduction – logical development of the consequences of an explanation starts with theoretical model ⇒ testable prediction ⇒ observations under specific conditions ⇒ confirmation or rejection of the prediction and/or the model • Enrico Fermi proposed the existence of the neutrino in 1930 because the observed decay products from beta decay seemed to violate mass and energy conservation. • In 1956, Cowen and co-workers detected its existence. Deduction versus Induction Induction – generalization of observed patterns starts with observations ⇒observed patterns ⇒development of model ⇒testable predictions ⇒competition of models ⇒theory • John Snow in 1854 observed that patients who had contracted cholera had been drinking water from a particular pump in London. • He suspected that the cholera was spread by contaminated water. • Led to Louis Pasteur’s formulation of germ theory in 1857. • Bacteria and viruses were later confirmed by direct observation, establishing their connection to disease. http://espanol.video.yahoo.com/watch/327162/2140779 Fallacies Circular Reasoning – Begging the question Appeal to emotion Argument from authority Sweeping Generalization Irrelevant Conclusion Denying the antecedent For a given measureable parameter, there exists a true value of that parameter for a set of circumstances at a given time. We do not know what it is, nor do we have any independent means of knowing it. Precision versus Accuracy Probabilistic versus Deterministic Models