Devopsdays Paris 2015: Cognitive biases and our poor intuitions around probability

It's a wonder we can make decisions at all, as technology workers or otherwise. Unlike language and grammar, our intuitive grasp of probability and statistics is atrocious. We're also subject to all sorts of common cognitive biases and bad heuristics that lead to errors and missteps in decision making.

Why are we so easily fooled by retail pricing? Why do we overestimate the chance of memorable disasters and successes recurring? Why do many post-mortems fail at providing true understanding due to their reconstruction as a sequence of predictable events?

In this talk we’ll be covering the cognitive biases behind these questions and others, along with relevant examples from the life of a technology worker such as candidate interviewing, post-mortem analysis and service availability. Presented at Devopsdays Paris - 14/15 April 2015.

Video: http://www.dailymotion.com/video/x2njo2h_devopsdays-paris-2015-nigel-kersten-cognitive-biases-and-our-poor-intuitions-around-probability_tech

Transcript

1. Cognitive biases and our poor intuitions around probability Nigel Kersten CIO, VP Operations at Puppet Labs @nigelkersten Devopsdays Paris, April 2015. Image: In The Dice Box by Tomio Tapio K, (Flickr, CC BY-SA 2.0)
2. What are Cognitive Biases anyway? • Systematic errors in how we process and interpret information with an apparently irrational result • Often due to heuristics that simplify information processing • Result from a deficiency in our thinking rather than a logical fallacy
3. The Xerox Bad Copy Scanning Bug • Discovered by David Kriesel in 2013 • 8 years old… • http://www.dkriesel.com/en/ blog/2013/0802_xerox- workcentres_are_switching_writ ten_numbers_when_scanning
5. Originals Copies
6. Originals Copies
7. Original Copy “If using pattern matching upon compression there is no guarantee that parts of the scanned image actually come from the corresponding place on the paper.”
8. Heuristics Gone Wild
9. Two Systems • System 1 • Fast • Instinctive, automatic • Emotional • Subconscious • System 2 • Slower, requires effort • Deliberate • Logical • Lazy, easily exhausted • Conscious
10. System 1 and System 2 2 + 2 = ?? 2 ^ 8 = ?? 637 x 213 = ??
11. System 1 and System 2 LEFT left right RIGHT RIGHT left LEFT right upper lower LOWER upper UPPER lower LOWER upper 1. Go down both columns, silently call out whether each word is in lowercase or uppercase by saying “upper” or “lower”. 2. Repeat, but this time call out whether each word is to the left or the right of the center by saying “left” or “right”.
12. Our intuitive grasp of probability and statistics sucks "Levy distributionPDF" by User:PAR - Own work. Licensed under Public domain via Wikimedia Commons - http://commons.wikimedia.org/wiki/ File:Levy_distributionPDF.png#mediaviewer/ File:Levy_distributionPDF.png
13. Intuition and Probability: The Law of Small Numbers A study of the incidence of kidney cancer in the 3,141 counties of the United States reveals a remarkable pattern. The counties in which the incidence of kidney cancer is lowest are mostly rural, sparsely populated, and located in traditionally Republican states in the Midwest, the South, and the West. Why?
14. Why? Possibly because of the clean lifestyle in rural areas. Fresh food, clean air, and lower all-round pollution. "MaryJane harvests corn" by MaryJane Butters - Licensed under Creative Commons Attribution-Share Alike 3.0 via Wikimedia Commons - http://commons.wikimedia.org/wiki/ File:MaryJane_harvests_corn.jpg#mediaviewer/ File:MaryJane_harvests_corn.jpg
15. Intuition and Probability A study of the incidence of kidney cancer in the 3,141 counties of the United States reveals a remarkable pattern. The counties in which the incidence of kidney cancer is highest are mostly rural, sparsely populated, and located in traditionally Republican states in the Midwest, the South, and the West.
16. Why? Possibly because of the poverty in rural areas. Poor access to medical care, lower education levels. "Poor mother and children, Oklahoma, 1936 by Dorothea Lange" by Dorothea Lange - Library of Congress LC- USF34- 009694-E. Licensed under Public domain via Wikimedia Commons - http://commons.wikimedia.org/wiki/ File:Poor_mother_and_children,_Oklahoma,_1936_by_Dorot hea_Lange.jpg#mediaviewer/ File:Poor_mother_and_children,_Oklahoma,_1936_by_Dorot hea_Lange.jpg
17. How did we get here? • rural • sparsely populated • located in traditionally Republican states
18. How did we get here? • rural • sparsely populated • located in traditionally Republican states
19. How did we get here? • rural • sparsely populated • located in traditionally Republican states
20. How did we get here? • rural • sparsely populated • located in traditionally Republican states
21. How did we get here? • rural • sparsely populated • located in traditionally Republican states
22. We are not intuitively good at statistics 1. The Law of Large Numbers: Large samples are more precise than small samples
23. We are not intuitively good at statistics 1. The Law of Large Numbers: Large samples are more precise than small samples 2. Small samples yield more extreme results more often than large samples do
24. We are not intuitively good at statistics 1. Large samples are more precise than small samples 2. Small samples yield more extreme results more often than large samples do (1) is well known and feels obvious. (2) is often not.
25. System 1 is the problem • Eliminates doubt, suppresses ambiguity to construct a coherent story • Embellishes information to construct a complete story • Jumps quickly to conclusions based upon constructed story
26. Small Sample Size Example You’re hiring for a new job at work, with internal and external applicants, and have just received the feedback from all the interviewers, who are current employees. Two applications got overall positive feedback from the six interviewers. 1. One of your current employees got reasonably positive feedback from all interviewers. 2. An external candidate who none of the interviewers have ever worked with got absolutely outstanding feedback from two interviewers and neutral feedback from the other four.
27. Intuition and Probability: Clustering Illusion • We under-predict the variability that’s likely to appear in a small sample
28. Intuition and Probability: Clustering Illusion • We under-predict the variability that’s likely to appear in a small sample • We expect random events that occur in clusters to not really be random events
29. Intuition and Probability: Clustering Illusion • We under-predict the variability that’s likely to appear in a small sample • We expect random events that occur in clusters to not really be random events • If you toss a coin 20 times in a row, your chance of getting 4 heads in a row is 50%.
30. Animated "RandomPoints" by CaitlinJo - Own work. Licensed under CC BY 3.0 via Wikimedia Commons - http:// commons.wikimedia.org/wiki/ File:RandomPoints.gif#/media/ File:RandomPoints.gif
31. Can you see objects in these pictures?
32. Can you see objects in these pictures? There’s no object on the left. If you “saw” one, you might be feeling under stress or out of control. J. A. Whitson, A. D. Galinsky (2008). Lacking Control Increases Illusory Pattern Perception Science, 322 (5898), 115-117 DOI: 10.1126/science.1159845
33. Clustering Illusion - stronger when frustrated • When stressed or feeling out of control, we’re more likely to see patterns where none exist
34. Clustering Illusion - stronger when frustrated • When stressed or feeling out of control, we’re more likely to see patterns where none exist • More likely to believe conspiracy theories and to embrace superstition
35. Clustering Illusion - stronger when frustrated • When stressed or feeling out of control, we’re more likely to see patterns where none exist • More likely to believe conspiracy theories and to embrace superstition • Restoring feelings of control and affirmation almost entirely negates this effect
36. Intuition and Probability: Availability Heuristic • What happens when we estimate the frequency of a category? • Examples: • How often are you rude to someone? • How often are your co-workers late to work? • How often does Amazon have an outage?
37. Intuition and Probability: Availability Heuristic • What happens when we estimate the frequency of a category? • First we retrieve instances of that category from memory • Then we replace the question about frequency with one about how easy it was to remember the instances
38. Norbert Schwarz experiments (1995) Two groups of test subjects • List 6 instances where you behaved assertively • Now judge how assertive you are • List 12 instances where you behaved assertively • Now judge how assertive you are
39. Norbert Schwarz experiments (1995) Two groups of test subjects • List 6 instances where you behaved assertively • Now judge how assertive you are • List 12 instances where you behaved assertively • Now judge how assertive you are Result: The difficulty of remembering more instances causes people to judge themselves as being less assertive!
40. Norbert Schwarz experiments (1995) • When subjects were asked for 12 instances where they were not assertive, they judged themselves afterwards to be more assertive. • We frown when thinking hard — symmetric effect • When subjects were asked to frown, same results as being asked to think of more instances • When subjects were told background music would make listing 12 items more difficult, the effect was mitigated
41. Norbert Schwarz experiments (1995) • Crazy paradoxical results • We’re less confident in a choice when we have to produce more arguments to support it! • We’re less confident an event was avoidable after listing more ways it could have been avoided! • Relevant: Job Interviews, Post-mortems, Architectural designs
42. Biases in Postmortems • Availability Heuristic • We consider easily recalled information to be more important. • Hindsight Bias • We see events as predictable after they have occurred, regardless of whether they were or not at the time. • Outcome Bias • Our assessment of actions is heavily affected by the consequence of those actions. • Fundamental Attribution Error • We place too much emphasis on people’s internal characteristics rather than external factors when trying to explain their behavior.
43. Mitigation for postmortems • Hindsight Bias • Record predictions prior to results, review after results • Outcome Bias • Focus on and reward quality of judgements, not outcomes • Availability Heuristic • Examine the data that will be used to make a decision before making it • Distribute the load when collecting lists of actions and reasons • (e.g. Ask each person to produce one item rather than a complete list)
44. Mitigating intuitive probability • Work with frequencies rather than probabilities • Think diagrammatically • Internalize that subsets of random data will contain predictable looking sequences
45. Mitigating Ego Depletion - System 2 gets tired easily • Make critical decisions early in the day • Minimize unimportant decisions early in the day • Improve your mood • Doesn’t improve the ability of System 2 to function • Doesn’t slow down ego depletion • Does counteract ego depletion
46. Thanks! - References and Attributions “Thinking, Fast and Slow” - Daniel Kahneman, ISBN: 9780374275631 “The Human Side of Postmortems” - Dave Zwieback, ASIN: B00CLH38CM “Extraneous factors in judicial decisions” - Danzigera, Levavb, Pessoa. http://www.pnas.org/content/108/17/6889.full.pdf "Ease of retrieval as information: Another look at the availability heuristic” - Schwarz, Norbert; Bless, Herbert; Strack, Fritz; Klumpp, Gisela; Rittenauer-Schatka, Helga; Simons, Annette (1991). Journal of Personality and Social Psychology 61 (2): 195–202. doi:10.1037/0022-3514.61.2.195. All images licensed for reuse without attribution or memes by unknown authors unless otherwise noted.
47. Anchoring Effect When we consider a specific value for an unknown quantity before estimating it, our estimation is heavily influenced by the specific value.
48. Anchoring Example 1. Is the height of the tallest redwood more or less than 1,200 feet? • What is your best guess about the height of the tallest redwood? vs 2. Is the height of the tallest redwood more or less than 180 feet? • What is your best guess about the height of the tallest redwood?
49. Anchoring Example 1. Is the height of the tallest redwood more or less than 1,200 feet? • What is your best guess about the height of the tallest redwood? vs 2. Is the height of the tallest redwood more or less than 180 feet? • What is your best guess about the height of the tallest redwood? 1. Mean Answer: 844 feet 2. Mean Answer: 282 feet
50. Anchoring Effect • One of the strongest, most repeatable effects in experimental psychology • Doesn’t matter whether you believe the initial information is relevant or not • Doesn’t matter how motivated you are to produce a correct estimate • Experts are still susceptible, although more resistant • Inconclusive whether smarter people are less susceptible
51. Anchoring Effect relevance • We estimate all the time in operations and development • System Performance • SLA design • Monitoring checks • Purchasing software and hardware • Project Planning
52. Mitigating the Anchoring Effect • Role-play opponent’s moves when negotiating • Devil’s Advocate • Develop more expertise • But… it’s a really robust effect