I have read a lot of how-the-brain-works books. They fall into one of three categories depending on which cognitive science discipline the author most identifies with. First, Computer Science, philosophy, and computational linguistics authors portray the human brain as a computer. Next, Cognitive Neuroscience authors focus on reductionism (connectomics) and functionalism (divining the purpose of regions of the brain). Finally, Psychology authors describe behavior using statistical models…which may be predictive but explain very little. Each of these viewpoints is much too narrow to do justice to the complex and multidisciplinary nature of intelligence
So I was pleased to read a how-the-brain-works book that takes a different and—I believe—a more holistic tack. A Brief History of Intelligence by Max Bennett describes the incredibly complex human brain as a progression of simpler, evolving brains. Bennett bootstraps our understanding of the human brain, language, and culture by tracing our evolution from three billion years ago. He simplifies what could be a potentially lengthy and complex story by focusing on five key milestones, or what he calls “breakthroughs”, in the evolutionary path from a single-celled organism to human intelligence.
This is a big-idea book. Like Yuval Noah Harari's Sapiens, Jared Diamond's Guns, Germs, and Steel, or Malcolm Gladwell's Blink, A Brief History of Intelligence is multi-disciplinary, broad in scope, yet approachable. Fun facts fill the book, weaving into a neat whole. Like those other big-idea books, some academic specialists may challenge parts of the narrative proposed by Bennett. However, any highly cited academic journal paper can face the same criticism. As Thomas Kuhn points out, all paradigm shifts require multidisciplinary leaps of imagination that provide a better understanding and explanation of phenomena than the previous paradigm.
I direct my biggest criticism toward the publisher. When I read journal papers or popular science books, I like to check (and often read) the references. The bibliography asserts “To save paper, the full bibliography can be found at [a] briefhistoryofintelligence.com”. I suspect that this feature of the book was likely encouraged by the publisher in order to save money. However, this decision required me to spend extra time in a cumbersome process to check Bennett’s references. It begs the question: how does leaving a bibliography out of an eBook save paper?
The cognitive sciences have been stuck in a rut for a long time. We have pretty pictures of fly brain connectomes, but our understanding of brains has progressed little in the last twenty years. One problem is that cognitive science never invited evolutionary biologists and ethologists to the party. This is changing and I hope this book heralds a new trend.
Special thanks to Dr. William Q. Judge for recommending this book and for suggestions on this review.
A common problem with stories about the evolution of the nervous system from the simplest variants to the human level is the presence of an analysis of geometry/morphology in the absence of an analysis of the functions of the system components (at best, vague hypotheses exist). This makes the analysis of little use to developers of AI systems.
The best text about AI (actually about everything) created by R.Feynman: "If I can't build it, I don't understand it."