What are the differences of Autoencoder and Autoassociative neural networks?


I am pretty new to Machine Learning and I am a little bit confused about these two networks. They both try to find some latent features in the input, both seem to have the same architecture and are trained in the same way using back propagation. So what makes them different?