Can you reverse engineer the human brain? When will we have computers that replicates or ventures beyond human intelligence? Will computers ever become conscious beings? What is “intelligence”? What is “consciousness”? Are we humans installed with a free will, and if this is so, how does this apply to computers?

These are indeed deep questions that are tough to answer. But in his book “How to Create a Mind: The Secret of Human Thought Revealed”, Ray Kurzweil comes up with intriguingly simple and intuitive answers though admittedly through complex analysis of areas such physics, chemistry, neuro- and computerscience where many – me included – will struggle to assess whether his is right or wrong.

What strikes me as truly different about Kurzweil’s analysis and conclusions are that they are made by a person who has extensive technical and indeed commercial experience with creating artificial intelligence (AI). As opposed to most other treaties on the subject that are made by 100% theoreticians that find their empiri in the vacuum of a laboratory. Kurzweil has succesfully applied his research and work to create voice recognition software based on AI. His inventions partly form the basis of a lot of widely used speech recognition applications such as Apple’s Siri.

Kurzweil’s core thesis is that the type of intelligence that distinguishes humans from other mammals is to be found in the workings of the neocortex. The neocortex is the outer layer of the cerebral hemispheres. The human neocortex is vastly larger than that of other mammals. All other animals than mammals do simply not have a neocortex. The neocortex holds neurons. A neuron is an electrically excitable cell that processes and transmits information through electrical and chemical signals. Kurzweil argues that these neurons serve as pattern recognizing modules linked together through dendrites and are exchanging information through electrical and chemical signals (“firing”).

The pattern recognizers are organized hierarchically. Lower level pattern recognizing modules hold information about “simple” things such a curves and edges and other shapes. The higher the level of hierarchy the more abstract the patterns. The many different pattern recognizers are organized in a hierarchy – explains Kurzweil – by algorithms based on Hidden Markov models (a mathematical concept that I do not fathom). When we perceive and try to make sense, our brains process information by these pattern hierarchies through a process of massive parallel computing. However, Kurzweil’s comparison to computer science and architecture does not end there. He also draw on the concept of redundancy and Shannon’s theorem on information, the works of Turing and von Neumann and much more.

As a layman I cannot assess whether Kurzweil has solved one of the great mystery of life: How does the brain work? Intuitively, however, I find his ideas to be an important step in the right direction. His very pragmatic approach makes his case even more plausible. As a believer in science, an optimistic rationalist and a big fan of Darwin, this book has had me subscribe to Kurzweil’s Pattern Recognition Theory of the Mind (PRTM).

If you have read other Kurzweil books and writing on the Singularity and his Law of Accelerating Returns, you will not find “How to Create a Mind: The Secret of Human Thought Revealed” a difficult read. For non-technical newbies it might be too complicated. That being said, I have no training with the natural science, including mathematics, and to me the book was a very enjoyable reading experience. I highly recommend “How to Create a Mind: The Secret of Human Thought Revealed” to everybody interested in our brains.