After working a while with animal behavior and neuro-research, I tend to ask myself if there is proof an animal can't do something rather than if they can. They usually learn slower than humans, depending on animal and task, but they can learn complex tasks. What they fail at usually tends to be due to physical limitations such as visual acuity. I don't think we have found the limits of what many animals can do but we are raising the bar slowly. It takes a lot of thought and work to design an experiment with a complex task correctly.

Of Note: Pigeons can classify a Picasso from a Monet at an expert level and peck their answer within 300ms: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1334394/. They were tested on assembly lines to pick out defective parts. They did better than humans but were not used due to decreased moral of other humans on the line. This was done in the early 60's I believe. Here is an article on it in the New Scientist (1962) http://books.google.com/books?id=HxU-9UeDCI0C&pg=PA498&#...