Baldwin effect

The Baldwin effect, named after the American psychologist James Mark Baldwin, is the process by which a learned or behaviorally acquired trait becomes genetically assimilated over evolutionary time. The effect describes a feedback loop: organisms that learn a beneficial behavior survive and reproduce at higher rates, and over generations, selection favors genetic variants that either facilitate the learning of that behavior or replace it entirely with an innate equivalent. The Baldwin effect is not Lamarckian — it does not claim that acquired characteristics are directly inherited. Rather, it claims that selection can favor genetic backgrounds that make beneficial learning easier or unnecessary, thereby shifting the population's phenotypic distribution through the interaction of learning and natural selection.

The Baldwin effect has been invoked to explain the evolution of language (genetic assimilation of linguistic competence), the evolution of immune responses (genetic capture of learned antigen recognition), and the evolution of social behaviors (innate versions of behaviors initially transmitted through social learning). It is particularly relevant to the extended inheritance framework because it provides a mechanism by which information initially transmitted through behavioral or epigenetic channels can eventually be 'canalized' into the genetic channel. The effect is also of interest in evolutionary computation and artificial life, where researchers have demonstrated Baldwinian evolution in silico: populations of learning agents evolve toward genetic backgrounds that reduce their dependence on learning, or that make the learning process itself more efficient.