Diagnostic Reasoning
9 important questions on Diagnostic Reasoning
What heuristics may explain why doctors make mistakes in diagnosing their patients?
- Availability heuristics
- Confirmation bias
What is a limitation with research in this area?
-> To what extent does this research (that generally presents participants with hypothetical scenarios) tell us anything useful abut diagnostic reasoning in the real world?
What is the Subadditivity/ Unpacking Effect?
--> subadditivity/ unpacking effect has been attributed to availability heuristic i.e. the ease that information/examples come to mind can affect your judgment
The subadditivity/ unpacking effect suggests that if you get people to think about more alternatives to the hypothesis, this changes the probability they may assign to different diagnosis
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
What is Bayes’ Theorem?
Bayes' theorem is a normative model which links the degree of belief in a hypothesis (and its alternative(s)) before and after accounting for new evidence.
-> The posterior odds of a hypothesis being true (H) or false (-H) is determined by combining the prior odds of the hypothesis being true or false with the likelihood ratio of the evidence
What does base rate neglect studies tell us about how people make diagnostic judgements?
Tversky & Kahneman (1974) argued that base rate neglect examples indicate that when people are making probability judgements, they use a representativeness heuristic - people are looking at the extent to which the symptom/illness resembles the stereotypical case
-> this can lead people to neglect the base rate information
What is the representativeness heuristic? Who was it proposed by?
Tversky & Kahneman (1974)
The representativeness heuristic is the tendency to judge the frequency or likelihood of an event by the extent to which it resembles the typical case
How does confirmation bias has some overlap with the pseudodiagnosticity effect?
- you fail to think about how likely those symptoms are in people who have an alternative disease/ don't have the disease
What did Cahan et al. (2003) study?
They gave physicians descriptions of patients with a range of symptoms, test results etc. They also gave the physicians a list of possible diagnoses and asked them to assign probabilities to each diagnosis
Results:
65% of physicians probabilities that they gave for each diagnosis, if added together, equaled more than 100% (subadditivity problem) i.e. their total probability estimate was greater than 100%
-> suggests that numbers don’t mean anything to Ps - the percentages are a reflection of the degree of belief someone has for a particular diagnosis
What theory supports/relates to the Subadditivity/ Unpacking Effect?
- reflects the idea that numerical probabilities are not that meaningful in absolute terms because they are subjective probability judgements
•Subjective probabilities reflect the degree of belief determined by the support for the hypothesis in one’s mind
•Unpacked events may remind people of overlooked possibilities or enhance their salience
•Reliance on availability heuristic contributes to the effect (unpacking enhances the accessibility of particular causes and their apparent likelihood)
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding