August 20, 2009

The test-and-learn model of change

Herminia Ibarra explains in her book Working Identity: Unconventional Strategies for Reinventing Your Career that the conventional way of thinking about career change corresponds with the plan-and-implement model. This model says that you first have to analyse and reflect in order to be able to develop a clear picture of what you want to achieve and only then you can take steps to realize this picture.

However, Ibarra’s research shows that effective career change follows a different pattern, one which is described by the so-called test-and-learn model. This model is based on the idea that learning is a circular and iterative process. “We take actions, one step at a time, and respond to the consequences of those actions such that an intelligible pattern eventually starts to form”. On page 34 of her book she explains this by the following table:


  1. Interesting.
    I would say there is one thing missing in the test-and-learn model: you need to have a criteria, or a general outcome in mind (not necessarily a goal, just "how things will be different when..."). In other words, you need to have something against which you can compare the results of your tests ("it works better" compared to what?).
    See also the "prediction neurons" in the ACC, i.e. we activate our attention when, in navigating the world, we find our expectations are violated, and that is a key mechanism in learning - our neurons adjust accordingly.
    Having said that, in the debate between "thinking then doing" vs. "doing then thinking", I have always been a great fan of the latter - as Tim WIlson says, first you act and then the rest will follow. One paper about complex decision making that I read at SFI also oriented me towards this kind of view, adaptive and based on exploration.

    The table is very interesting, and I am fascinated by the "lineair" (a new airline? a line in the sky?) vs, the "curculair" distinction (I guess definitely an exotic airline...) :)))

  2. Interesting read...
    For some reason I keep thinking about Rationality / Irrationality dichotomy from Socionics.

  3. Paolo makes some great points. One of my favorite quotes keeps coming back as a heuristic in my daily life: "plans are useless, but planning is indispensible."

    Working without goals is a really bad idea, because we typically end up in places pretty far from where we wanted to be, and not in a good sense, then we rationalize that we're in a better place. Without some navigational reference to assist feedback, the human mind tends to magnify small errors into big deviations. That's why some people wander in circles when lost without any guiding signal at all.

    On the other hand, we don't need very precise or detailed goals in many cases, rough criteria will often do and dynamic specific goals. If we choose the criteria wisely. The more specific the goal, the more effort it takes to revise it when it isn't working out. But the process of planning for it is extremely useful and can be applied to the next change in goal. The planning process is dynamic, the goals change, but we do keep moving toward a desired end state rather than wandering. That's like shooting randomly with an arrow and then drawing a circle around wherever it sticks and calling it the target. That's the nature of feedback without external navigational landmarks.

    Also agree that taking action is often the best way to set the process in motion. Particularly if you have at least a plausible rough guess at your first action.


Enter your email address:

Delivered by FeedBurner