Course Objectives Research Methods Understand the nature and - PDF document

Course Objectives Research Methods ■ Understand the nature and practice Scientific Method of science in general Tim Weyrich ■ Learn how to identify and test http://www.cs.ucl.ac.uk/staff/t.weyrich/ scientific hypotheses Based on slides by Daniel Alexander ■ Understand publishing processes ■ Perform a literature review for your Master’s project ■ Obtain practice in written and oral presentation 1 2 Course Objectives Overview ■ A bit of philosophy ■ Learn how to work like a scientist ■ Some technical material (after all, it’s an M Sc …) – Statistical hypothesis testing • Frequentist and Bayesian – Experiment design • Modelling and sampling, experiment design. ■ Handling live subjects – Data from humans (questionnaires) – Ethics in science ■ Mechanisms of scientific discourse ■ Oral and written presentation 3 4 Assessment Today’s Outline ■ Coursework ■ What is this thing called science? • 2 pieces, 20% ■ Facts and experiments ■ Written Literature Review ■ Induction and Deduction • max.(!) 10,000 words, 60% ■ Falsifiability ■ Final Presentation on literature and ■ Paradigms and revolutions planned research ■ Other theories of science • 15 minutes, 20% 5 6

What is this thing called science? What is this thing called science? ■ Knowledge derived from facts! ? 7 8 Facts and experiments Facts and experiments ■ What are the facts ■ What are the facts 9 10 Facts and experiments Facts and experiments ■ What are the facts ■ What are the facts ■ What people observe depends highly ■ What experiments do you do? on their experience ■ Influence of theory on: • Botanist • What experiments • Radiologist • Experimental setup ■ Observations of facts are fallible 11 12

Deduction and induction Deduction and induction ■ How do we derive knowledge from Deductive logic is truth preserving ■ facts 1. All lectures on philosophy are boring ■ Deductive logic: Move from given 2. This lecture is on philosophy premises (by definition true) to 3. Therefore, this lecture is boring conclusions 1. All humans have 2 noses ■ Inductive logic: Move from particular 2. I am human facts to general theory 3. Therefore, I have 2 noses 13 14 Deduction and induction Deduction and induction When can we apply inductive logic ■ Inductive logic ■ • N is large • Metal x 1 expanded when heated at t 1 • N is varied • Metal x 2 expanded when heated at t 2 • The theory does not contradict any • Metal x 3 expanded when heated at t 3 existing evidence • Therefore, all metals expand when heated Inductive logic as applied by the ■ Christmas turkey (Bertrand Russel) 15 16 Deduction and induction Deduction and induction When can we apply inductive logic ■ Further problems with inductionism ■ • N is large • How can the existence of unobservable • N is varied concepts (DNA, atoms) be derived from • The theory does not contradict any observable facts existing evidence • How can the correctness of inductionism Problematic!! ■ be induced? • When is N large enough • What should we vary • There are always exceptions 17 18

Deduction and induction Falsifiability Theories Karl Popper (1902 – 1994) ■ Worried about all kinds of ‘scientific ■ theories” which were backed up by facts (Marxism, Freud) These theories can be made to ■ Facts New fit any data acquired predictions But do they rule out anything? ■ • Rainbow example 19 20 Falsifiability Falsifiability Science makes definite predictions ■ Eddington’s test of general relativity in ■ Science is therefore falsifiable 1919 ■ Observations are guided by and ■ presuppose theories Theories are tentative constructions ■ created by scientists to overcome problems existing with previous theories 21 22 Falsifiability Falsifiability Falsifiable statements A hypothesis is falsifiable if there ■ ■ exists a logically possible observation It never rains on Wednesdays 1. statement that is inconsistent with it Either it is raining or it is not raining 2. Example of falsification All points on a Euclidean circle are 3. ■ equidistant from the centre • Theory: All swans are white All metals expand when heated • Discovery of Australia 4. Heavenly objects fall in straight lines • Observation of black swans 5. towards the earth • Theory is falsified 23 24

Falsifiability Falsifiability Is string theory a valid scientific theory? Falsifiable statements ■ Luck is possible in speculation in 1. "For more than a generation, physicists have been chasing a willo � -the-wisp called string theory. The beginning of this chase sports marked the end of what had been three-quarters of a century of progress. Dozens of string-theory conferences have been held, When a ray of light is reflected from a hundreds of new Ph.D.s have been minted, and thousands of 2. papers have been written. Yet, for all this activity, not a single plane mirror, the angle of incidence is new testable prediction has been made, not a single theoretical puzzle has been solved. In fact, there is no theory so far—just a equal to the angle of reflection set of hunches and calculations suggesting that a theory might exist. And, even if it does, this theory will come in such a bewildering number of versions that it will be of no practical use: a Theory of Nothing. http://www.newyorker.com/archive/2006/10/02/061002crat_atlarge 25 26 Falsifiability Falsifiability A caricature of Adler’s psychology Degrees of falsifiability ■ ■ • All behaviour is guided by feelings of a) Mars moves in an ellipse around the sun inferiority of some kind b) All planets move in ellipses around the • Situation: Man is standing next to a sun treacherous river and suddenly a child The set of potential falsifiers for a) is ■ falls in. a subset of the set of potential • What does he do? falsifiers for b). 27 28 Falsifiability Falsifiability The enterprise of science involves the ■ Inductionism Falsificationism proposal of highly falsifiable Only theories that are Any theory hypotheses, followed by deliberate true are scientific (speculative or not) and tenacious attempts to falsify which is falsifiable is useful for the them. advancement of our Science is trial and error. knowledge ■ Falsification is the major growth path ■ The more precise a theory is the more falsifiable it is and thus the better 29 30

Falsifiability Falsifiability Progress of science Sophisticated falsification ■ Set of problems ■ Competing theories ■ Set of hypotheses to deal with them ■ New theory is accepted above ■ Falsification of many of these ■ previous ones if: hypotheses • It is more falsifiable than the previous Further testing of the remaining ■ one successful hypotheses • It predicts a new phenomenon not taken New problems arise ■ into account in previous theories Etc… ■ 31 33 Falsifiability Falsifiability The discovery of Neptune The proposal of highly risky new ■ ■ theories is not always a good thing • The orbit of Uranus did not fit with Newtonian mechanics Progress is especially made if: ■ • A new planet was proposed which A bold conjecture is confirmed . 1. disturbed Uranus’s orbit A cautious conjecture is falsified 2. • A range of possible orbits for the new Confirmation does have a role in the planet was derived ■ (intial) acceptance of theories • The new planet (Neptune) was observed in the range of predicted orbits! 35 36 Limitations of falsifiability Limitations of falsifiability Fact is inconsistent with theory Example: Tycho Brahé’s refutation of ■ ■ • Experimentalist: “Theory is wrong” Copernicus • Theorist: “Measurement is wrong” If the earth orbits the sun, the position ■ Observations are theory-driven and ■ of the stars should differ during the fallible. So if O is observed, which course of a year. falsifies T, then this will only mean He did not observe this because he that O| T is false ■ underestimated the distance to these The observation of O is ■ other stars supplemented by auxiliary theory (measurement apparatus) 37 38