Skip to content
Carmelics
Topics
Thinkers
Changes
Contributors
Loading account…
Statements
321,452
Perspectives
108,905
Topics
42
Home
/
Original
/
inverse
See Original
Inverse View
It is not the case that Expected utility calculations assign probabilities to outcomes based on the agent's epistemic state, not the act's causal structure.
?
Set your confidence on the premises below to see your aggregate.
Reasons For
1 perspective
Reason for
?
1.
An act's causal structure determines which outcomes actually follow from it; rational agents should track actual consequences, not mere beliefs.
?
How convincing is this?
Think about whether this reason is strong or weak
2.
Two agents with identical beliefs but different causal structures face genuinely different decision problems with different objectively optimal solutions.
?
How convincing is this?
Think about whether this reason is strong or weak
3.
Expected utility theory should correct agents toward accurate causal models, not merely codify whatever probabilities they happen to assign.
?
How convincing is this?
Think about whether this reason is strong or weak
Reasons Against
1 perspective
Reason against
?
1.
Agents only have access to information in their epistemic state; causal facts unknown to them cannot rationally guide their decisions.
?
How convincing is this?
Think about whether this reason is strong or weak
2.
Two agents with identical beliefs but different objective causal structures should make identical rational choices given identical options.
?
How convincing is this?
Think about whether this reason is strong or weak
3.
Probability assignments in decision theory reflect degrees of belief, which are inherently features of the agent's mind, not of the world.
?
How convincing is this?
Think about whether this reason is strong or weak
Next step
Based on where you are in your exploration
Strongest counterpoint
Explore the most compelling reason on the other side.