View
4
Download
0
Category
Preview:
Citation preview
UX ResearchCognitive Biases
KP Ludwig John
UX Researchcognitive biases
UX ResearchCognitive Biases
KP Ludwig John
Estimate quickly
Task
1 x 2 x 3 x 4 x 5 x 6 x 7 x 8
for Group 1Group 2 turn around, please
UX ResearchCognitive Biases
KP Ludwig John
Estimate quickly
Task
What’s your guess?Write it down
Group 1
UX ResearchCognitive Biases
KP Ludwig John
Two modesof thinking
UX ResearchCognitive Biases
KP Ludwig John
operates automatically, quickly and with little or no effort
System 2 (slow thinking)
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 20f
System 1 (fast thinking)
To gather and process information our mind switches between two different modes of operation.
Daniel Kahneman (Psychologist) describes these modes in his book Thinking fast and slow and calls themfast thinking and slow thinking or simply:
Two Systems
allocates attention to the effortful mental activities that demand it, including complex computations.
Also in charge for monitoring and (self) control
UX ResearchCognitive Biases
KP Ludwig John
Two Systems Examples
Reference: Daniel Kahnemann - Interview with Guardian 22. Nov. 2011https://www.youtube.com/watch?v=RHmXPyX7czU
UX ResearchCognitive Biases
KP Ludwig John
operates automatically, quickly and with little or no effort
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 21f
Two Systems
Examples• detect that one object is more distant than another • Orient the source of a sudden sound • Answer to 2+2 = ? • Detect hostility in a voice • Understand simple sentences
System 1 (fast, automatic)
Several mental actions … are completely involuntary. For instance: You can’t refrain from understanding simple sentences in your own language.
UX ResearchCognitive Biases
KP Ludwig John
allocates attention to the effortful mental activities that demand it, including complex computations.
Also in charge for monitoring and (self) control
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 22
System 2 (slow, controlled)
Two Systems
Examples• Count the occurrences of the letter „a“ in this text • Tell someone your phone number • Fill out a tax form • Compare two washing machines for overall value • Check the validity of a complex argument
Operations of System 2 require attention and are disrupted when attention is drawn away!
UX ResearchCognitive Biases
KP Ludwig John
Both are active whenever we are awake.
System 1 runs automatically. It is in charge of the things we do most of the time. System 2 is normally in a comfortable low effort mode
When System 1 runs into difficulties, it calls on System 2 to support more detailed and specific processing that may solve the problem.
System 2 is mobilized when a question arises for which System 1 does not offer an answer.
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 24
Two Systems
System 1 + System 2
Compare 2 x 2 = ?17 x 24 = ?
UX ResearchCognitive Biases
KP Ludwig John
Two Systems Examples
Reference: H. Pishro-Nik from https://www.probabilitycourse.com/https://www.youtube.com/watch?v=qNV-c-95vTM
UX ResearchCognitive Biases
KP Ludwig John
Cognitive biases (kognitive Verzerrungen)
A cognitive bias refers to the systematic pattern of deviation from norm or rationality in judgment Individuals create their own "subjective social reality" from their perception of the input
Thus, cognitive biases may sometimes lead to perceptual distortion, inaccurate judgment, illogical interpretation, or what is broadly called irrationality.
Reference: https://en.wikipedia.org/wiki/Cognitive_bias
Some Examples:
UX ResearchCognitive Biases
KP Ludwig John
Estimate quickly
Task
for Group 2
8 x 7 x 6 x 5 x 4 x 3 x 2 x 1
UX ResearchCognitive Biases
KP Ludwig John
Estimate quickly
Task
What’s your guess?Write it down
for Group 2
UX ResearchCognitive Biases
KP Ludwig John
Estimates?
Task
Group 2
8 x 7 x 6 x 5 x 4 x 3 x 2 x 1
1 x 2 x 3 x 4 x 5 x 6 x 7 x 8Group 1
Result (calculated): 40 320What’s been your guesses?
UX ResearchCognitive Biases
KP Ludwig John
Examples
Reference: Laurie Santos (Yale University)https://www.youtube.com/watch?v=NFiDdbquWJY
Anchor bias
UX ResearchCognitive Biases
KP Ludwig John
Anchor bias
Anchoring Effect occurs when people consider a particular valuefor an unknown quantity before estimating that quantity. The estimates stay close to the number that people considered. Any number that you are asked to consider as a possible solutionto an estimation problem will induce an anchoring effect
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 119f
!
The spinning wheel Experiment
UX ResearchCognitive Biases
KP Ludwig John
Anchor biasThe spinning wheel Experiment
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 119f
Wheel of fortune - prepared to only stop at either 10 or 65. Test persons were asked to turn the wheel and write down the resulting number (which would be 10 or 65).
Then answered these two questions:
Is the percentage of African nations in the UNlarger or smaller than the number you just wrote? What is the best guess of the percentage of African nations in the UN?
Average Estimatessaw
estimated10 6525% 45%
UX ResearchCognitive Biases
KP Ludwig John
Examples
Reference: Procurement Academy (YouTube Channel) https://www.youtube.com/watch?v=83Y-B72BiYY
Anchor bias
UX ResearchCognitive Biases
KP Ludwig John
Confirmation bias
Influences how people gather information but it also influences how they recall and interpret it.
Reference: www.quora.com/What-are-some-good-examples-of-a-confirmation-bias
Confirmation bias is the tendency to accept evidence that confirms our beliefs and to reject evidence that contradicts them
Reference: www.thoughtco.com/what-is-confirmation-bias-1689786
Example
If you were a messy kid … mom always picked out the instances when the room was a mess and socks were all over the floor, but never acknowledged when you had spent an hour cleaning it. This was simply because she was looking for evidence to support her already formed thesis that you were a messy kid. (Jochen Gruenbeck)
UX ResearchCognitive Biases
KP Ludwig John
Confirmation bias Example
Reference: H. Pishro-Nik from https://www.probabilitycourse.com/www.quora.com/What-are-some-good-examples-of-a-confirmation-bias
UX ResearchCognitive Biases
KP Ludwig John
Confirmation bias
When asked: Is Sam friendly?
… different instances of Sam’s behavior will come in mindthan would if you been asked:
Is Sam unfriendly?
People seek data that are likely to be compatible with the beliefs they currently hold.
The confirmation bias … favors uncritical acceptance of suggestions.
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 81
How does this effect surveys?
UX ResearchCognitive Biases
KP Ludwig John
Confirmation bias
Reference: Concordance Channel at YouTube https://www.youtube.com/watch?v=u34BhEgO_es
Example
UX ResearchCognitive Biases
KP Ludwig John
Confirmation bias
Reference: IMS.UX 2018 Team Long
Example
Identical game, identical demonstration Different wording in Question!
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect
Reference: https://commons.wikimedia.org/wiki/File:Gandhara_Buddha_(tnm).jpeg
A halo (from Greek ἅλως, halōs;[1] also known as a nimbus, aureole, glory, or gloriole)
is a crown of light rays, circle or disk of light[2] that surrounds a person in art.
Reference: https://en.wikipedia.org/wiki/Halo_(religious_iconography)
Gandhara Buddha. 1st-2nd century CE. Tokyo National Museum.
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect
Reference: https://en.wikipedia.org/wiki/Halo_effect
When good-looking people, for example, are perceived as more intelligent, more successful and more popular, "that's the halo effect in psychology"[1]
It's caused by a cognitive bias,the tendency for an impression created in one area to influence opinion in another area.[3]
Reference: https://commons.wikimedia.org/wiki/File:Gandhara_Buddha_(tnm).jpeg
in Psychology
First use of this term in psychology was a 1920 paper by Edward Thorndike
This effect is named in reference to a personbeing perceived as having a halo
Gandhara Buddha. 1st-2nd century CE.
Tokyo National Museum.
A halo effect is an outcome in one area due to factors derived from another.
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect
Reference: https://en.wikipedia.org/wiki/Halo_effect
Gandhara Buddha. 1st-2nd century CE.
Tokyo National Museum.
Reference: https://commons.wikimedia.org/wiki/File:Gandhara_Buddha_(tnm).jpeg
in PsychologyObservers tend to bend their judgement according to one patent characteristic of the person (the "halo"), generalizing towards a judgement of that person's character
If the observer dislikes one aspect of something, they will have a negative predisposition toward everything about it.[8]
Effect works in both positive and negative directions.
If the observer likes one aspect of something, they will have a positive predisposition toward everything about it.
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 82
Example
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect / Primacy Effect
Psychologist Solomon Asch presented descriptions of two people and asked for comments on their personality.
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 82
In other situations, evidence accumulates gradually and the interpretation is shaped by
the emotions attached to first impression.
Alan
intelligent – industrious – impulsive - critical – stubborn - envious intelligent - arbeitsam - impulsiv - kritisch - dickköpfig - neidisch
Ben
envious – stubborn – critical - impulsive – industrious – intelligent neidisch – dickköpfig – kritisch - impulsiv – arbeitsam – intelligent
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect / Primacy Effect
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 82
Choose one of the two as your co-worker in the project team!!
Alan
intelligent – industrious – impulsive - critical – stubborn - envious intelligent - arbeitsam - impulsiv - kritisch - dickköpfig - neidisch
Ben
envious – stubborn – critical - impulsive – industrious – intelligent neidisch – dickköpfig – kritisch - impulsiv – arbeitsam – intelligent
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect / Primacy Effect
The initial traits (Eigenschaften) in the list change the very meaning of the traits hat appear later.
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 83
Stubbornness of an intelligent person is seen likely to be justified … but intelligence in an envious and stubborn person make him more dangerous.
The halo effect increases the weight of first impressions, sometimes to the point that subsequent information is mostly wasted
Alan
intelligent – industrious – impulsive - critical – stubborn - envious intelligent - arbeitsam - impulsiv - kritisch - dickköpfig - neidisch
Ben
envious – stubborn – critical - impulsive – industrious – intelligent neidisch – dickköpfig – kritisch - impulsiv – arbeitsam – intelligent
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect / Primacy Effect
Reference: http://www.intropsych.com/ch15_social/biasing_effects_in_person_perception.html
Half the students received a description that listed his technical qualifications then noted (at the end) that he was considered a "very warm" person.
The other half of the students received the same description but with an ending that said he was considered a "rather cold" person.
The lecturer gave the same talk to all the students.
Those who had warm in their descriptions rated the lecture more favorably and were more likely to ask questions and interact with the lecturer.
The students with cold in their descriptions rated the lecturer as aloof (abgehoben) and unsociable. Only 32% said they wanted to ask him a question or interact with him, compared to 56% in the group that heard him described as warm.
!
Experiment by Harold Kelley in 1950: He gave a class of students a paragraph-sized description of a visiting lecturer.
UX ResearchCognitive Biases
KP Ludwig John
Reference: https://www.simplypsychology.org/asch-conformity.html
Conformity
Solomon Asch - Conformity Experiment (1951)
to investigate the extent to which social pressure from a majority group could affect a person to conform.
Aim
Procedure
Using a line judgment task, Asch put a naive participant in a room with seven confederates.
The real participant did not know this and was led to believe that the other seven participants were also real participants like themselves.
„Vision Test“ as a Lab experiment with 50 male students.
The confederates had agreed in advance what their responses would be when presented with the line task.
UX ResearchCognitive Biases
KP Ludwig John
Conformity
Reference: https://www.youtube.com/watch?v=NyDDyT1lDhA
Dr. Philip Zimbardo about Conformity Experiment by Solomon Asch
UX ResearchCognitive Biases
KP Ludwig John
Reference: https://www.simplypsychology.org/asch-conformity.html
Conformity
Solomon Asch - Conformity Experiment (1951)
On average, about one third (32%) of the participants who were placed in this situation went along and conformed with the clearly incorrect majority on the critical trials
Results
about 75% of participants conformed at least once, and 25% of participant never conformed.
In the control group, with no pressure to conform to confederates, less than 1% of participants gave the wrong answer.
However, the subjects conformed much less if they had an „ally".Apparently, it is difficult to be a minority of one but not so difficult to be part of a minority of two.
Reference: http://www.age-of-the-sage.org/psychology/social/asch_conformity.html
UX ResearchCognitive Biases
KP Ludwig John
Reference: https://www.simplypsychology.org/asch-conformity.html
Conformity
Solomon Asch - Conformity Experiment (1951)
When they were interviewed after the experiment, most of them said that they did not really believe their conforming answers, but had gone along with the group for fear of being ridiculed or thought "peculiar".
Conclusion
Apparently, people conform for two main reasons: because they want to fit in with the group (normative influence) and because they believe the group is better informed than they are (informational influence).
UX ResearchCognitive Biases
KP Ludwig John
Halo Effect / Primacy Effect / Conformity
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 84f
How to avoidUse multiple sources of evidence,which are independent from each other
Allow individual (private) judgments
In Meetings: before to start a discussion let participants write down their position on the subject
In Evaluations: change sequence and order of test items throughout the process
To get personal evaluations of subsequent answers / resultsunbiased by previous results / evaluations / impressions (Halo)evaluate … without knowing previous results of the personor without knowing who the person is
UX ResearchCognitive Biases
KP Ludwig John
Availability heuristic
Reference: https://en.wikipedia.org/wiki/Availability_heuristic
Mental shortcut that relies on immediate examples that come to mind when evaluating a specific topic, concept, method or decision
Operates on the notion that if something can be recalled, it must be important, or at least more important than alternative solutions which are not as readily recalled
People tend to heavily weigh their judgments toward more recent information, making new opinions biased toward that latest news
UX ResearchCognitive Biases
KP Ludwig JohnReference: https://www.youtube.com/watch?v=2_wkv1Gx2vMYouTube Channel CogSai
Availability heuristic
UX ResearchCognitive Biases
KP Ludwig John
Availability heuristic
Task 1: List six instances in which you behaved assertively (durchsetzungsfähig)
Task 2: Evaluate how assertive you are
in Questionnaires
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 132
Experiment by Norbert Schwarz
Task 1: List twelve instances in which you behaved assertively (durchsetzungsfähig)
Variation
Research Question
Will people evaluate differently depending on the number of instances asked
UX ResearchCognitive Biases
KP Ludwig John
Availability heuristic in Questionnaires
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 132
Experiment by Norbert Schwarz
People who listed twelve instances rated themselves as less assertive than people who listed only six.
If you can not easily come up with instances for a certain behavioryou are likely to conclude you do not have that characteristic
Findings
• number of instances retrieved • the ease with which they come to mind
Two aspects to consider
Finding twelve instances is more impressive but more demanding (difficult) as well!
Which aspect will count more?
UX ResearchCognitive Biases
KP Ludwig John
Availability heuristic in Questionnaires
Reference: http://nbn-resolving.de/urn:nbn:de:0168-ssoar-67234
Experiment by Norbert SchwarzApplications of this experiment in other context found that people:
• believe that they used their bicycle less often after recalling many rather than few instances
• are less confident in a choice, when they are asked to produce more arguments to support it
• are less confident that an event was avoidable after listing more ways it could have been avoided
• are less impressed by a car after listing many of its advantages
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 132
UX ResearchCognitive Biases
KP Ludwig John
Availability heuristic in Questionnaires
Experiment by Norbert Schwarz
Professor (at UCLA) asked two different groups of students to list ways to improve the course.
… and … Varied the required number of ways to improve
As expected: the students who listed more ways to improve the classrated it higher!
Practical Exploitation of this experiment
Reference: Kahnemann, Thinking fast and slow, Penguin books 2011, p. 133
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect
Reference: https://psychologenie.com/meaning-examples-of-framing-effect
ExampleA glass of water which is either half-full or half-empty: both are equivalent truths.
When portrayed in a negative frame, you think that the glass is half-empty. If portrayed in a positive frame, you see the glass as half-full.
A displeasing event at the start of your day puts you in a negative frame of mind, and makes you feel negative.Everything goes wrong that day.
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect
Reference: https://psychologenie.com/meaning-examples-of-framing-effect
ExamplePositive FrameThe product has been proven effective in 80% cases.
Negative FrameThe product has failed to work on 2 out of every 10 cases.
People react in different ways to the same choice depending on the way and in which context it is presented.
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect
Reference: https://psychologenie.com/meaning-examples-of-framing-effect
The power of DefaultOrgan donation programs
People react in different ways to the same choice depending on the way and in which context it is presented.
UX ResearchCognitive Biases
KP Ludwig JohnDan Ariely about the power of Default (2008)Reference: https://www.youtube.com/watch?v=JhjUJTw2i1M
Framing Effect The power of Default
UX ResearchCognitive Biases
KP Ludwig John
Framing EffectReference: https://constantrenewal.com/design-for-default/
The power of Default
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect
Reference: Tanay Jaipuria https://medium.com/@tanayj/the-power-of-defaults-976bc8b015b7
The power of Default
Making an option the default among a set of choices increases the likelihood of it being chosen.
Reasons• Choosing something that isn’t the default requires some effort.
• Often the default is viewed as the recommended option.(Particularly important when choices are regarded as complex.)
Walt Disney WorldWhen restaurants changed the default choices in its kids’ meals from soda to juice and french fries to fruits and vegetables.
Kids consumed 21% fewer calories and 40% less fat and sodium.
Reference: https://wdwnews.com/releases/walt-disney-world-menus-evolving-with-healthful-choices/
UX ResearchCognitive Biases
KP Ludwig JohnDan Ariely about the power of Compare References (2008)Reference: https://www.youtube.com/watch?v=JhjUJTw2i1M
Framing Effect Compare References
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect Compare References
Dan Ariely
Reference: http://zsoltbabocsai.org/dan-ariely-predictably-irrational/#sthash.i89Ag6B2.dpbs
The Economist wanted to offer these subscription options:
3. Print-and-Internet subscription for $125.
1. Internet-only subscription for $59.2. Print-only subscription for $125.
But they realized that very few people would pay that much for print so they added a twist.
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect Compare References
Dan Ariely
Reference: http://zsoltbabocsai.org/dan-ariely-predictably-irrational/#sthash.i89Ag6B2.dpbs
3. Print-and-Internet subscription for $125.
1. Internet-only subscription for $59.2. Print-only subscription for $125.
By adding this third option … they changed the basis for comparison. You get the online for free … if you pay for the print, which makes the 3rd option much more attractive.
Two options for the same price makes it easy to compare those two and choose the better one, and disregard the first option even if it’s half the price.
UX ResearchCognitive Biases
KP Ludwig John
1. Internet-only for $59. – 16%2. Print-only for $125. – 0%3. Print-and-Internet for $125. – 84%
Framing Effect Compare References
Dan Ariely Reference: http://zsoltbabocsai.org/dan-ariely-predictably-irrational/#sthash.i89Ag6B2.dpbs
Subscription models chosenby Students in the experiment
1. Internet-only for $59. – 68% 2. Print-only for $125. – 32%
UX ResearchCognitive Biases
KP Ludwig John
Framing Effect Compare ReferencesReference: http://munchweb.com/predictably-irrational-summary-pdf-chapters-review-of-dan-arielys-bookDan Ariely
• A
• B (very distinct, but equally as attractive)
• and A- (similar to A, but inferior)
We will almost always choose A, because it is clearly superior to A-
Given three choices:
The addition of a third “A-” option, “Paris without a free breakfast”,will cause us to choose “Paris with a free breakfast”, the “A” option, over “Rome with a free breakfast”, the equally attractive “B” option.
UX ResearchCognitive Biases
KP Ludwig John
Summary Heuristics and Biases
Value present in the mind of a test personbefore asked to estimate a quantity, frequency, probability etc.
Anchor bias
Tendency to accept evidence that confirms our beliefs Confirmation bias
Outcome in one area due to factors derived from another.Halo effect
Conformity biasPeople want to fit in with the group (normative influence) and / or they believe the group is better informed than they are
Availability heuristicOperates on the notion that if something can be (easily) recalled, it must be important
People react in different ways to the same choice depending on the way and in which context it is presented.
Framing Effect
UX ResearchCognitive Biases
KP Ludwig John
PracticeLook for answers to your research questionsby conducting a survey
Determine 2 controversial desired results
Choose a research subject,and state your research questions
Imagine you have a personal interest that your survey yields a certain result
Create two versions of the surveyto produce the desired results
Apply at least three of the mentioned Biases when creating the versions
UX ResearchCognitive Biases
KP Ludwig John
Practice
Determine 2 controversial desired results
Create two versions of the surveyto produce the desired resultsApply at least three of the mentioned Biases when creating the versions
Come Together in Teams
Choose a test subject and research questions Determine measurements
Teamwork15 minutes
UX ResearchCognitive Biases
KP Ludwig John
Create a first draft of surveys
Task
• state the desired opposite outcomes of your surveys
• Create 10 Questions • State which Biases you want to use and how
Be ready to discuss your first drafts
next week
Determine the members of your teamand find a descriptive name for it
UX ResearchCognitive Biases
KP Ludwig John
Schedule + Material
21. January
delivery + presentation All Eyetracking and ObservationPDF
05. November
14:00 All short presentations (15 Min.) + deliverySurveys
hs-augsburg.de/homes/john
Recommended