This book was released in 2004, however the important thing to remember is "There are keys in it to understand consciousness."
As Jeff Hawkins says, [association to prediction] process is based on [input to memory]. And his hypothesis claims his brain neocortex model feasible. Actually I also feel that the input to be well-known if the input is already in the memory. If the input is not in the memory, the input would be sent to upper state, and understood if it is in the upper state memory.
This process does not require an observer in the brain. Each neuron fires and only associates with next neuron. Consciousness arises at each neuron decentrally. And each portion understands the situation, (not central controlled.) At this point his hypothesis's thought [of generalization and supplemental if the input is incompletion] works in the background.
In "On Intelligence" there were a lot of excellent thoughts. However some "wet" portions were not shown mainly. At this point I consider the "wet" portions: consciousness, metacognition, free will, instinct, and emotion are very important. My program: BLUE considers (also) the "wet" portions as the key for behavior as human including consciousness.
Keyword: neuroscience, consciousness