Genetic algorithms (GAs) have a wide variety of applications in control. However, GAs may suffer from slow convergence rates, and require the user to make difficult choices of ranking and scaling schemes and subpopulations that may lead to complexities in implementation. A new computationally inexpensive alternative to GAs, the continuous adaptive culture model (CACM), is proposed in this paper. This new optimization algorithm is inspired by sociological models of culture dissemination and uses operators that act directly on vectors of real numbers to avoid the computation associated with binary encoding and decoding in GAs. The new algorithm does not use global information sharing which makes it amenable to parallel implementation since computational bottlenecks are avoided. The De Jong test suite of optimization problems is used to test the new optimization algorithm. Effects of various parameters on the performance of the algorithm are investigated through simulations. ©2005 AACC.