Bursts – Review

by: Barabási, Albert-László

Publisher: Dutton, Penguin Group USA Inc

Location: New York

Copyright: 2010

Cover: Richard Hasselberger, Terry Chan/Shutterstock

Type: Hardcover


reviewed by: Lynard Barnes 8/3/2010


Summary: The predictability of human behavior? or the nodes of changing circumstance? A statistical basis for examining human behavior.


This is not a must read book. However, there are snippets and gems scattered throughout and a read would definitely not be a waste of time.

It is almost an intuitive realization that people are predictable. BURST is a refinement of this intuitive, gut feeling. Indeed, Professor of Physics Albert-László Barabási and his team have put a number on that predictability. Ninety-three percent of human behavior is predictable. For those who participated in the B. F. Skinner mechanical behaviorism (operant conditioning) versus, for lack of a better term, human spiritualism wars of the 1980s, either vicariously or in word-combat, the 93% rating of human behavior may be a wave of the let’s-fight-this flag. Just as we intuitively recognize that human behavior is predictable, we have a reflex attitude that, no, human behavior is not predictable.

To his credit, Barabási bases his 93% predictability rating on sound, scientific observation. Cell phones-their use and abuse (which, in our opinion is mostly abuse). Who can possibly argue with science? Rhetorical question. Especially if you start with science as a sub-set of reality. But if you don’t want to expend the energy to delve into quantum physics and tick-off the requisite indicators of “reality”, accepting the 93% predictability rating for human behavior is the smart thing to do. Still . . .

BURSTS proceeds from an assumption that what you do today is based on what you or someone else did yesterday and before. Consequently, “you are far more predictable that you are willing to admit.” Into this nice little box of scientific precept is Hasan Elahi who the author describes as a thirty-year-old media installation artist. More significantly, the box includes György Székely’s (or György Dózsa) and the failed Hungarian peasant uprising of 1514. Barabási makes good use of both Hasan and Székely to bolster his contention that our behavior is 93% predictable.

Major chapters of the book are devoted to quantifying the extremes of randomness, least some think that human behavior is random and that our broader world is comprised of random events. What is the travel-universe of a dollar bill? What keeps pollen suspended in dew drops in constant, jittery and irregular motion? Does the seemingly haphazard roaming of wild animals in search of food follow any type of search pattern? Can we ever reach a point in sociology in which we can describe human behavior as accurately as we describe the physical world? Barabási addresses these questions and more. He provides an enjoyable foray into the exploration of predictability and behavior patterns. Still . . .

Barabási’s book loosely defines “bursts”, which “permeate human activity”, as an apparently sudden intensity of activity that was dormant for a time. He pursues this definition as it relates to activity lists: how tasks are prioritized and accomplished; the dependency of an item list upon some other item on the list, and so forth. It is a dauntingly complex definition that, in essence is relatively simple. People perform behaviors based on past behavior and, in most instances, perform behaviors which are dependent upon previous behaviors. What’s new here?

Did the resultant failed Hungarian uprising headed by György Székely have all the elements of a predictable failure? Did the constant travel lifestyle of Hasan Elahi lead inordinately to questioning by the Federal Bureau of Investigation? These questions are answered-sort of-in the author’s book. But you really have to dig for answers. Or, after reading about the circumstances, you can simply rest with your intuitive answers. This, believe it or not, brings us back to cell phones.

Barabási and his team at Northeastern University, the team being really a network of scientists, studied the mobility patterns of 100,000 anonymous cell-phone users who were randomly selected from six million users (see the [email protected] for the story). Results of the study indicated that people follow a “simple pattern regardless of time and distance, and they have a strong tendency to return to locations they visited before.” Is there anything surprising here? The real question is however is whether it is possible to predict the behavior of an individual?

The behavior of individuals is probably in the neighborhood of 99% predictability. FBI profilers have been trudging this road since the mid-1980s.Barabási’s “bursts” definitely adds a brick or two to the required protocol for analyzing individual behavior. Read the book. But do not expect any light-bulb moments.