e-Literate

Present is Prologue

Tag: informational-cascades

  • Full Context: Democratic Primaries as an Informational Cascade

    I have been digging more into the subject of informational cascades as I work on editing my submission to eLearn on “emergent learning” (which I’m going to refer to as social learning instead, given the current lack of agreement about what “emergent learning” means). The more I dig, the more excited I get. There’s some deep and important stuff here.

    Many of the best insights are buried in heavily mathematical academic papers on efficient market theory; but every once and a while you run across a gem that’s written to a broader audience. As William Dulker points out in his Full Context weblog, Slate’s William Saletan provides a wonderful, easy-to-understand example of an informational cascade in the Democratic primaries.

  • Book Recommendation: The Selfish Gene

    It may seem odd, given the focus of this blog, to recommend a book on evolutionary biology. But Richard Dawkins’ book The Selfish Gene lays a solid foundation for helping to understand developments in the aggregation sciences.

    Dawkins’ main thesis is that evolution is driven by the survival of the fittest genes, not organisms and not species. If it seems odd to think of dumb genes “competing” or acting “selfishly,” as the title suggests, then you’re catching on to what’s provocative about this book. If you have trouble wrapping your head around the ideas of complex adaptive systems and emergence, then take a step back and start by trying to understand selfish genes. If you can make the latter conceptual leap then the former becomes much easier to get. Dawkins helps with the learning process by giving many, many concrete examples and carefully reasoned, easy-to-follow arguments.

    If you borrow the book rather than buying it, make sure that you get the 1989 edition or later; the newer edition has two new chapters that are essential. In particular, Chapter 12, “Nice guys finish first,” is about the evolution of cooperation and has quite a bit about game theory in it. Again, if you understand game theory and its impact on a population, then other ideas in aggregation science will come a lot easier.

    One last point of interest is that Dawkins’ book is the origin of the term “meme”. If you’ve heard the term “idea virus,” then you have a sense of what memes are. (And by the way, thinking about memes is one good way to start getting a sense of why informational cascades are a potential problem in the kind of networked conversational learning that happens, for example, in a blogging community. You could argue that an “idea virus” is either a cause or a manifestation of an informational cascade.) To see an example of somebody using the “meme”…er…meme productively, see this oldie but goodie post from the Truth Laid Bear weblog.

  • The Problem with "Emergent Learning:" Informational Cascades

    I just submitted an article to eLearn Magazine that starts to get at one of the reasons why I am skeptical about emergent learning as a panacea. I’m not going to give away too many of the surprises here (unless eLearn decides not to publish the article, in which case I will post it here), but let’s just say that you can find the roots of the argument in the article Conversation, Learning, and Informational Cascades [PDF]. From the abstract:

    We offer a model to explain why groups of people sometimes converge upon poor decisions and are prone to fads, even though they can discuss the outcomes of their choices. Models of informational herding or cascades have examined how rational individuals learn by observing predecessors’ actions, and show that when individuals stop using their own private signals, improvements in decision quality cease. A literature on word-of-mouth learning shows how observation of outcomes as well as actions can cause convergence upon correct decisions. However, the assumptions of these models differ considerably from those of the cascades/herding literature. In a setting which adds ‘conversational’ learning about both the payoff outcomes of predecessors to a basic cascades model, we describe conditions under which (1) cascades/herding occurs with probability one; (2) once started there is a positive probability (generally less than one) that a cascade lasts forever; (3) cascades aggregate information inefficiently and are fragile; (4) the ability to observe past payoffs can reduce average decision accuracy and welfare; and (5) delay in observation of payoffs can improve average accuracy and welfare.