The causal meaning of Fisher's average effect

Area of Science:

• Evolutionary Biology
• Genetics
• Biostatistics

Background:

• Fisher's Fundamental Theorem of Natural Selection relies on average excess and average effect of gene substitution.
• These concepts have been perceived as opaque, leading to reformulations using statistical terms like covariance and regression.

Purpose of the Study:

• To clarify Fisher's original intent regarding average effect and average excess.
• To reconcile statistical and causal interpretations of these genetic concepts.
• To highlight the importance of the correlation vs. causation distinction in evolutionary genetics.

Main Methods:

• Analysis of Fisher's original formulation of the average effect.
• Reconciliation of statistical and causal conceptions of average effect.
• Examination of theoretical implications for gene-trait mapping and evolutionary theory.

Main Results:

• Fisher's averages were intended to distinguish correlation from causation.
• The statistical and causal views of average effect can be reconciled under specific conditions.
• Fisher's causal formulation offers theoretical advantages, such as deriving constraints rather than imposing them.

Conclusions:

• Fisher's original causal framework for average effect is robust and theoretically advantageous.
• Understanding the distinction between correlation and causation is crucial for gene-trait mapping and evolutionary biology.
• This work supports Fisher's foundational concepts in natural selection theory.