Differences

This shows you the differences between two versions of the page.

--- public:t-713-mers:mers-25:concepts_terms [2025/08/19 12:18] – [Theories of the Scientific Method (Philosophy of Science)] thorisson
+++ public:t-713-mers:mers-25:concepts_terms [2025/08/19 12:27] (current) – [Correlation, Knowledge, Causation] thorisson
@@ Line 24: / Line 24: @@
 ====Correlation, Knowledge, Causation====
-| Correlation  | Some factors/variables co-vary when changes in one variable are related with changes in the other, negative or positive.  |
+|  Correlation  | Some factors/variables co-vary when changes in one variable are related with changes in the other, negative or positive.  |
-| Correlation: Powerful source of information  | Any variables in the world can be measured for correlation (i.e. to see if they are correlated). Only two variables are needed (independent and dependent) for doing correlation studies.   |
+|  Correlation: Powerful source of information  | Any variables in the world can be measured for correlation (i.e. to see if they are correlated). Only two variables are needed (independent and dependent) for doing correlation studies.   |
 | Main operating principle behind correlation  | There is no causation without correlation. \\ BUT: It is not guaranteed to be measurable.  |
-| Correlation: Pitfall   | Correlation does not imply causation between the variables measured! \\ BUT: ALL correlation that is NOT a coincidence has a cause.    |
+|  Correlation: Pitfall   | Correlation does not imply causation between the variables measured! \\ BUT: ALL correlation that is NOT a coincidence has a cause.    |
+|  Modeling Correlation  | A model that makes use of observed correlation between A and B is only good for prediction, not for manipulation; to be useful for manipulation (goal-directed behavior and planning) it must include the causal direction of the relationship between A and B.   |
+|  Causal Relations are Invisible  | Causal relations do not jump out at us when we observe something unfamiliar because the cause and effect are "spread out" over time -- cause-effect happens over time, as the cause must happen before the effect.   |
+|  Modeling Cause-Effect  | Intelligent agents can model cause and effect with many methods; two of them are invention (coming up with a wild idea for the relationship betweeen A and B, e.g. that spirits make people sick) and discovery (through observation and experimentation with A and B, e.g. that unclean surgical knives can bring disease-carrying material between people).   |
 \\