Related Experiment Video
Updated: May 22, 2026

Manipulation of Color Patterns in Jumping Spiders for Use in Behavioral Experiments
Published on: May 21, 2019
Does natural selection favour the Rescorla-Wagner rule?
Pete C Trimmer1, John M McNamara, Alasdair I Houston
1School of Biological Sciences, Woodland Road, Bristol BS8 1UG, UK. pete.trimmer@gmail.com
The Rescorla-Wagner learning rule, used in animal behavior, may be favored by natural selection because it is robust to parameter changes, even if not always optimal. This evolutionary advantage explains its prevalence in associative learning.
Area of Science:
- Animal behavior
- Computational neuroscience
- Evolutionary biology
Background:
- Animals learn associations between stimuli and rewards, a process often modeled by the Rescorla-Wagner learning rule.
- The Rescorla-Wagner rule is a widely applicable method for updating associative strength but is not always optimal.
Purpose of the Study:
- To investigate the evolutionary viability of learning rules with properties similar to the Rescorla-Wagner rule.
- To understand why a potentially suboptimal learning rule might be favored by natural selection.
Main Methods:
- Modeling the evolution of learning rules in a simplified environmental context.
- Comparing the Rescorla-Wagner rule with an optimal rule of similar complexity.
Main Results:
- The Rescorla-Wagner rule demonstrates greater robustness to parameter variations compared to the optimal rule.
- A wider range of parameter values supports the initial viability of the Rescorla-Wagner rule structure.
Conclusions:
- The Rescorla-Wagner rule's evolutionary advantage lies in its parameter sensitivity, not necessarily its accuracy.
- Natural selection may favor the Rescorla-Wagner rule due to its robustness, leading to its prevalence in animal associative learning.
Related Concept Videos
What is Natural Selection?
Natural Selection and Mating Preferences
Females, due to their biological roles in conception, pregnancy, and nursing, inherently...
Types of Selection
Limits to Natural Selection
Frequency-dependent Selection
Hardy-Weinberg Principle

