Connectionist production systems are neural network realizations of pr
oduction rule-based systems. The connections are adjusted to a given s
et of rules to allow the system to perform reasoning. Adaptable connec
tionist production systems are introduced in this paper. They allow ad
aptation of the already pre-calculated connections to new data. The pr
oduction rules are used to initialize the connection weights after whi
ch training with data occurs. At any time of the neural network operat
ion, a set of updated rules can be extracted as a current knowledge ba
se accumulated by the network. Using a set of rules for initializing a
connectionist architecture before training may result in: (1) increas
e in the speed of training; (2) increase in the robustness of the neur
al network against the 'catastrophic forgetting' phenomenon; (3) bette
r explanation of the learned by the network knowledge from data. In ge
neral, the proposed method facilitates building flexible and adaptable
neuro-fuzzy production systems. This is demonstrated on a case proble
m of chaotic time series prediction.