Cognitive Bias and Metacognition

Family Medicine is a Cognitive Specialty

...as in cognitive vs. proceduralist
...yes, feelings, emotions, communication, and personality are all important, of course...
But, information acquisition, information processing, information retrieval and informed decision-making are our most important skills
Evidence-based medicine in its original intention works on all these skills. "Evidence-based Medicine" as a label does not...

Metacognition is...

Thinking about thinking.

Used in psychology for different disorders.

We are talking about it in the educational psychology context.

Some Thought Questions to Begin...

Question 1

How should doctors think? What are our ideal cognitive processes?

Question 2

In medical education, what is the role of memory? Of reasoning?

Question 3

What are the "cognitive personality" traits that are more useful for family medicine?

Question 4

It's straightforward to assess for knowledge deficit, but what do improper cognitive processes look like? How do we diagnose those?

Diagnosing Thinking Problems

Separate into: Logic and economics
- both can lead to the "rational ideal"
- logic - conclusions, thinking
- economics - behaviors

Logical Fallacies

SchoolOfThought_FallaciesPoster_A3.pdf

Common Logical Fallacies that affect Medical Decision Making

appeal to authority
appeal to emotion
appeal to curiosity
bandwagon effect
anecdotalism

Cognitive Biases

SchoolOfThought_BiasesPoster_A3.pdf

Common Cognitive Biases that affect Medical Decision Making

anchoring
framing effect
dunning-kruger effect
confirmation bias
premature closure
availability heuristic

Other Phenomena that affect Medical Decision Making

Implicit Bias based on race/ethnicity/sex
Decision Fatigue
Physiological, emotional stress
Scarcity
Delay discounting

Dual Process Theory

Daniel Kahneman & Amos Tversky
System 1 - rapid, intuitive, pattern based, survival focused
System 2 - slower, deliberative, rational
multiple cognitive biases

Heuristics

Gerd Gigerenzer
Reaction to "behavioral economics"
Heuristics
- helpful to reduce complexity of information
- well-constructed heuristics can be very useful
Risk Communication
- bayesian reasoning, base-rate fallacy
medical examples

Nudge theory

Richard Thaler and Cass Sunstein
leveraging behavioral economics to achieve outcomes
- preparing systems so that when system 1 is used, it results in the desired outcome
examples
- default choice (presenting desired choice as default)
- social-proof (comparing your behavior to others)
- increasing salience of an option (healthy snacks at checkout)

Solutions

How can we teach residents to guard against logical fallacies, cognitive bias and other thinking problems?

Consider our own.
Diagnose the problem - system 1 thinking or system 2?
- is the resident in a hurry, stressed, angry, etc.? (situations in which system 1 might be helpful)
- Is the resident overthinking? drawing in too much data, unable to focus or decide? (too much system 2)
Slow down
Ask for supporting evidence (either signs and symptoms or studies/guidelines)
Re-broaden the differential or the therapeutic options - what else could be going on here? what else could be helpful?
Could a heuristic (clinical decision rule, best practice) help?

Examples

Resident presents a long case, lots of clinical details - arrives at a single diagnosis, or discusses only a single therapeutic option.
Resident is lost in details, stops short of assessment/plan, seems unable/unwilling to provide A&P.
ACLS and Evidence-based medicine

Structural Changes

work hour requirements
learning environment rules
implicit bias training
reminder systems to combat decision fatigue, etc.