Sunday, February 15, 2015

2/16 Dan - Mangling with Models

1. The main theme that I was able to discern from Pickering’s opening chapter is that science is not the static observations of isolated phenomena. It is a fluid, interactive process that continually involves that action and reactions between human and non-human agents. If we accept this as a fundamental truth, then it follows that science classrooms cannot be approached as a set of facts that need to be learned, but instead needs to reflect this relationship between people and the world around us. I think this relationship is what Pickering meant by the “mangle”, although it this is definitely a topic that I would like to clarify in class. I think that computational modeling is a good way to explore science in this manner because it does not limit science to just a set of facts or equations, but instead lets an individual interact with the environment. It also replicates an important part of the relationship between human and non-human agents that Pickering calls tuning. This is similar to the ideas the Nersessian expressed in her article, as she highlighted the importance of refining the model of the brain to more closely replicate the behavior that the researchers observed in the real samples of synapses.

2. Does modeling only work with organisms that don’t have agency, or at least have “less agency” than humans? Or can we expect human behavior to follow a few simple rules that would essentially ignore the idea of free will?

3. I could use the “Gravitation” model in a Physics class, to help with Newton’s Law of Universal Gravitation (HS-PS2-4). It models how bodies motion are influenced by the inverse-square law. I could also use the “Rope” model that shows how waves travel and how individual points simply oscillate. This would be helpful when studying waves and electromagnetic radiation (HS-PS4-1).

3 comments:

  1. I think Pickering's argument is that everything - from the material to the living - has agency in that it all acts in concert to produce scientific knowledge. I don't think it takes away from free will or humanhood to share in this agency, nor does it take away free will to be on the other side of the proverbial microscope. I feel certain that human behavior can be modeled by simple rules if you ask a specific enough question, which is true for any organism or system. Human behavior is no more special than any other complex system, we just happen to know more about the potential inputs (being human and all) so it seems infinitely more complex, but I think you'll find the same to be true for other systems as you become more entrenched in research. To that effect, I love the thought "Humans are not uniquely unique" from an old teacher. Food for thought from the biological perspective at least!

    ReplyDelete
  2. I think that the models we've been using to far in class have agency, and that most models we'll be using will have components of agency. Where we will find ourselves as instructors, is finding the perspective of agency in the model. For example, some models place the agent within the model, while for others the agent controls the model. Going back to what we've said in class, "All models are wrong, some are just useful," this is an important concept for our students to comprehend. Some models offer the affordances to engage in the interactions between agents. It is an important part of constructing science knowledge, perhaps in the revision process, that students can understand how models are not exact replications of a concept. Computational models use random code to show us how interactions between agents may play out, but certainly do not predict all possibilities of outcomes.

    ReplyDelete
  3. I think that good models will take into account the agency of the elements in the complex system. I feel like this ties into Laura's point from class last week about the goals of individual agents in relation to aggregate behaviors. In a model, this agency would be see in a part of code that randomizes motion or behavior to some degree. On the global level, we will likely see trends in behavior, but individuals are not bound to act in this way or another.

    To your point about human behavior, there are definitely people who believe humans behave according to a rational set of rules; they're called economists! ;)

    ReplyDelete