EPI 390 Lecture 2Outline of Last Lecture I. VocabularyII. Understanding BiasIII. Inductive vs. Deductive ReasoningIV. Defining CausalityV. Designing Epidemiological ExperimentsOutline of Current Lecture I. Cognitive ThinkingII. Occam’s RazorIII. Einstein and Needle in HaystackCurrent LectureI. Cognitive Thinkinga. Problems associated with Inductive/Instinctive reasoningi. Confirmation bias – people are (subconsciously) driven to confirm their beliefs or opinions; they are more prone to remember supportive evidence and discredit or ignore contradictory evidenceii. Fallibility of Human Memory – memories are easily lost or confused; people can also generate false memories of events that never occurred.1. Present beliefs or opinions can alter past memories.iii. Stories over Statistics – Humans relate more to anecdotal evidence than numerical/statistical evidence, though this is prone to flaws like leaving out information and is biased.iv. Observation vs. Reality – Humans perceive the world based on what they expect to see, not what is actually in the external world.1. Perceptual blindness is often used to discredit eyewitness testimonyv. Oversimplifications – The amount of information needed for decision-making is overwhelming, so humans tend to simplify it to make it manageable but important facts are lost that way. vi. Chance and Coincidence – Humans try to find causes for and explain chance events, even if there are no underlying cause-and-effect relationships.These notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.II. Occam’s Razora. “Don’t multiply the agents in a theory beyond what’s necessary” – William of Ockham, medieval theologian and logician.i. Aka with 2 competing theories, the simplest theory is more likely the correct one.III. Einstein and Needle in Haystacka. Einstein looked for all possible inductive explanations, instead of the first, most obvious, one.i. He would also look for the counter-intuitive answers and relationships between new and old datab. Used drawings, charts, etc., to find any patterns that might have been overlooked.c. Used analogies to make a concept easier to explain and understand.d. Tried different combinations of new and old information to generate new
View Full Document