PS2111 - Reasoning2024 Part I_BB.pptx

PS2111 – REASONING I DR CAREN FROSCH TopHat PS2111 (2023-24) - Information Processing and Cognition Join: 053994 Problem Solving Judgment & decision making Reasonin g PS3109 Inductiv e reasonin g Deductiv e reasonin g Hypothes is testing Definitions Inductive reasoning Forming generalisations (that...

PS2111 – REASONING I DR CAREN FROSCH TopHat PS2111 (2023-24) - Information Processing and Cognition Join: 053994 Problem Solving Judgment & decision making Reasonin g PS3109 Inductiv e reasonin g Deductiv e reasonin g Hypothes is testing Definitions Inductive reasoning Forming generalisations (that may be probable but are not certain) from examples or sample phenomena Hypothesis testing Confirmation or disconfirmation (falsification) of hypothesis Deductive reasoning Reasoning to a conclusion from a set of premises or statements where that conclusion follows necessarily from the assumption that the premises are true. Learning outcomes Explain the difference between inductive and deductive reasoning Explain different hypothesis testing strategies Be able to discuss the value of different hypothesis testing strategies Outline some deductive reasoning tasks Consider the evidence for different theoretical perspectives put forward to explain deductive reasoning Explain how performance on reasoning tasks can be improved Consider the extent to which reasoning tasks assess human rationality Overview of Reasoning I Inductive reasoning  focus on hypothesis testing Wason 2-4-6 task What does it tell us about hypothesis testing strategies? How do modifications to the task change strategies/enhance performance? Are people rational? Klayman& Ha Hypothesis testing outside of the laboratory Why use abstract tasks? Inductive reasoning Forming generalisations (that may be probable but are not certain) from examples or sample phenomena Phil Johnson-Laird 2006 Also PS2103 – thematic analysis Can you discover my rule? Generate number triples to find out This triple conforms to my rule: Number i 2 – 4 – 6 s e h t o p y H s Triple My rule is… Don’t reveal the rule if you have not attempted the task! Ascending numbers Consider the role of prior knowledge! See video on Blackboard for full illustration of task Hypothesis testing strategies Confirmation Search for evidence which verifies hypothesis Falsification Search for evidence which disconfirms hypothesis Falsifiability (the potential for falsification) separates scientific from unscientific activities such as religion and pseudo-science* Karl Popper *he probably did not use these exact words (see Eysenck & Keane, 2015 instead) Wason’s 2-4-6 task (1960) Most participants show confirmation / verification bias Only 21% correct with first triple 28% never discovered the rule Are people not rational? Confirmation Bias https://youtu.be/yxDDrEA49 7E Examples of confirmation bias 1. A student who is going to write a research paper may primarily search for information that would confirm their beliefs. The student may fail to search for or fully consider information that is inconsistent with their beliefs. 2. A reporter who is writing an article on an important issue may only interview experts that support their views on the issue. 3. An employer who believes that a job applicant is highly intelligent may pay attention to only information that is consistent with the belief that the job applicant is highly intelligent. Enhancing performance on 2-4-6 task Michelle Cowley Ryan Tweney What happens if the hypothesis you are testing is not your own? 62% abandoned other person’s hypothesis compared to 25% who abandoned their own (Cowley & Byrne, 2005) Get participants to infer experimenter has two rules in mind (DAX & MED) Over 50% got it right on first attempt (Tweney et al, 1980) Make it easy to identify relevant dimension Maggie When MED example was 444 -> 23% success Gale When MED example was 642 -> 75% success (Gale & Ball, 2012) Generation vs testing Paolo Cherubini People try to preserve information in original triple 1. Participants given examples 2-4-6 & 1618-20 2. Participants given examples 6-8-10 & 914-15  Group 1: 30% success rate  Group 2: 70% success rate (Cherubini et al, 2005) So are people really not rational? Klayman & Ha, 1987 ascendin g All triples Confirmation bias or positive test strategy? by e s a e r Inc 2 8, 10, 12 2, 4, 7 8, 10, 12 In c re as 2 e by ven e e v i t u consec In c re as 2 e by 3 even descending ascendin g Key conclusion from Klayman & Ha A positive test strategy is not always inappropriate as it can also lead to conclusive falsification if the circumstances are right Hypothesis testing in science What does it mean if a hypothesis is not supported in your practical report analysis? a) Hypothesis incorrect b) Problem with design c) Problem with data Hypothesis testing outside the laboratory Hypothesis: Insomnia caused by caffeine  positive test strategy: remove caffeine and see if it solves the insomnia  Positive testing is not irrational but an economical and diagnostic strategy of information search in many circumstances (e.g., Klayman & Ha, 1987; Oaksford & Chater, 1994) Why then use abstract tasks to investigate reasoning? Control by removing extraneous variables and background knowledge or exploiting it in a very targeted way, e.g. 2-4-6/2-4-8 All participants have exactly the same information Easier to manipulate aspects of the task Stripping reasoning task down to basic components  then you can examine the impact of context Summary Popper’s emphasis on falsification is oversimplified More appropriate to think of it as positive test strategy rather than confirmation bias This positive strategy is often appropriate in real world environments References Eysenck, M.W. & Keane, M.T. (2020). Cognitive psychology: A student's handbook. (8th ed.) Abingdon: Routledge. (ch.14 pp. 666 –672) Cherubini, P., Castelvecchio, E., & Cherubini, A. M. (2005). Generation of hypotheses in Wason’s 2–4–6 task: An information theory approach. The Quarterly Journal of Experimental Psychology Section A, 58(2), 309–332. Cowley, M. & Byrne, R. M. J. (2005). When Falsification is the Only Path to Truth. In B. G.Bara, L. Barsalou, & M. Bucciarelli (eds.). Proceedings of the TwentySeventh Annual Conference of the Cognitive Science Society. pp. 512-517. Mahwah, NJ: Erlbaum References Gale, M., & Ball, L. J. (2012). Contrast class cues and performance facilitation in a hypothesis-testing task: evidence for an iterative counterfactual model. Memory & Cognition, 40(3), 408-419. Klayman, J, & Ha, Y. (1987). Confirmation, disconfirmation, and information in hypothesis testing. Psychological Review, 94(2), 211-228. Tweney, R. D., Doherty, M. E., Worner, W. J., Pliske, D. B., Mynatt, C. R., Gross, K. A., & Arkkelin, D. L. (1980). Strategies of rule discovery in an inference task. Quarterly Journal of Experimental Psychology, 32, 109-123.