Being There: Putting Brain, Body, and World Together Again Paperback – January 9, 1998

I'm ever grateful to Professor Hubert Dreyfus (UC Berkeley) for having given me this book to read, I believe, in hard back in 1997. It took me all the examples to really grasp what professor Andy Clark was driving at -- and even then, not entirely. I've recently bought a Kindle version of this book for a revisit. It is also the book I recommend to anyone who wants to know anything related to cognition, machine learning, artificial intelligence, robotics, equipment design, system or organizational design. This is a comprehensible (and an always fresh) start for grasping Andy Clark's later works.

This book gets my 5-star rating because it made an excellent addition to my understanding of human life and cognition. Andy Clark is remarkably careful and balanced in his approach, while being totally readable and clear. The book is very well-documented, and his claims are firmly grounded in an argument that relies on few assumptions. He's tackling fundamental questions that are similar to those discussed by Daniel Dennet, but Clark's presentation is more careful and complete, while using an attitude that is far less strident.

I recommend this book to anyone who wants to understand better the nature of the perceived boundary between individual person, cognition, language, and interaction, and the way these coevolve.

My highest rating.

This is a terrific book. One of the best I've read in a long time.

I am reading Clark's book to learn how the understanding of mind, brain, and action has changed since I got a doctorate in cog. psych many decades ago. I am learning a lot from it. Being There clarifies several complex issues by, among other things, providing vivid examples.

My only criticisms are minor:
- About halfway through the book, it started to get a bit repetitive. I didn't need to be told again and again that perception and cognition are active and use the world as its own memory.
- There are a few errors. E.g., page 29 puts the area of high visual resolution (i.e., the fovea) at "0.01 percent" of the retina. It's 1 percent (0.01) of the retina, not 0.01 percent.

Great philosophical book!

I found this book very interesting and exciting. It presents a very beautiful and unique insight into the nature of cognition and its role in various social emergences.

excellent!

Besides this being a very creative and unique home-work assignment, the point of this review is to help you decide if this book is what you want to read for your neuroscience assignment in the upcoming semesters. Plenty of authors have tried to give rational explanations for the ideas behind recent research. Clark is no different. He lays down the foundation of the phenomena in the field of "Artificial Intelligence." At first a person may wonder how robotics ties into neuroscience, but it becomes clear as we reason and look at the way this field of research was carried out that it has everything to do with neuroscience. In fact the whole purpose of "artificial intelligence"(AI) is to get the product, whichever form the robot takes, to perform, react, and control itself just like a human being. But just how easy is that to do? Well first, it's important to understand how the human brain performs, reacts and controls itself in various scenarios. And then understand how it responds to new scenarios it has never encountered before. And not to forget to understand how the brain responds when a particular reaction is not working out so well and when it is working well. All this would need to be the basis for building a machine that could think like a human. So Clark delves into the complex human brain and then does a fantastic job of linking it to the research field. He starts by underlying the pattern of cognition that the human brain exhibits. He describes the brain as "controllers for embodied activity". In context with this book it means that the brain is very flexible in its controlling the many activities or scenarios that come its way. And he cites some great research examples in which controlling the embodiment of activity was implemented.
Personally I feel that Clark is a great writer. His style is very conversational, which is always a plus for an easily bored reader like me. As you read Clarks' book you ask yourself questions and he eventually answers them in upcoming paragraphs. However he does like to go into too much detail on a matter. This may not be an issue for readers who enjoy knowing the fine details of certain mechanisms and thoughts. However, keep in mind that this is just my personal opinion, not every bored reader is in opposition to great detail.

A great job is done in mixing exciting research into this book. But I particularly enjoyed Clark's analysis of how the mind works when planning and problem solving. When faced with a problem Clark describes the old fashioned AI method to planning as a "very disembodied phenomenon", meaning that generally when researchers want to make a machine that plans and solves, they have the mindset that it takes information, processes and reacts. This is of course accurate. But according to Clark, the human brain does so much more than that when faced with a problem to solve. When the desired result to solving a problem is not met, the human brain works on re-thinking another plan to solve the issue. "In many cases, however, the plan turns out to be something much more partial, and much more intimately dependent on properties of the local environment." In other words, depending on what we want to accomplish, our brains will re-work another plan, but based off of clues in the environment or setting we are in. Using these clues makes re-working the problem less difficult than it was to begin with. Clark elaborates on this point for the reason that AI machines need to be programmed to be able to re-plan, "re-think" and reorganize themselves in order to achieve a problem solving status similar to that of humans. So sensing and working with the environment is so crucial to humans and should be also to these types of machines. Another chapter I really enjoyed was "collective wisdom, slime-mode style". He introduces the Dictyostelium discoideum a species of slime mold. This is a rather unique organism that starts its life as an amoeba type of cell. Until the food runs out in its environment and is forced to change shape to a ball of slime. The ball of slime (in response to cues from the environment) chases food and light so it can reach and develop into the next phase. When the right amount of nutrients is found in its environment the slime ball stops moving, sets up and grows into a stem to reproduce another generation of cells. How is this knowledge useful to understanding the brain and ultimately making machines modeled after the brain? Well the human brain acts in a slime-mold fashion. When we have a purpose or agenda (like the slime mold trying to reach the reproduction phase) we look for changes in the environment to help us. I love this analogy Clark uses: "you have to remember to buy a case of beer for a party. To jog your memory, you place an empty beer can on your front doormat. When next you leave the house, you trip over the can and recall your mission." Environmental activities are used to shape our problem solving skills or reach our goal. Similar to the slime-mold using environmental stimuli, like light and food, to reach its goal.

So the application to robots and their design. Out of the many postulates he lists for the design of these human like robots a few are; real-world, real-time focus where a machine takes data and creates output based on time frames; extended vision of cognition and computation is the machines act of using other objects around itself to create output options; constraint satisfaction networks in which an artificial machine would take pieces of info from its environment and deduce what is around by eliminating what it knows is not around. To do this requires flexible sensory mechanisms. Andy Clark shows those in the field of AI research just how they should be designing their machines by looking at the brain in the right way, as a flexible cognitive controller, that can still function when the environment changes.