Lecture 2 (5/20/09): Getting ready for the quiz (Part II)

Don't forget, the Lecture 2 Blog has been divided into two posts: this one, and yesterday's Blog ("Getting ready for the quiz (Part I)"). Yesterday, I talked mostly about how/why we can think of genes as principles. This was a change from the previous Blog, in which I talked about genes as concepts. Today, the main topic is models. We're going to explore what Professor S means when he talks about "scientific models." The hope is that this discussion helps get you ready for Friday's quiz, on which you can probably expect a few questions about models. Lets get started...


Developing a concept of scientific models...

There are lots of ways that we use the term "model" in everyday language: people can be models in magazines or on billboards; kids can play with model trains and airplanes; sometimes people who do good deeds are called "model citizens." But we need to start developing a finely-tuned idea of scientific models, which I think Professor S was doing when he showed you the following slide (below left).

Just like I did in yesterday's blog with genes, I also tracked the statements that Professor S made when talking about models. Using the same form that I did with genes, take a look at the statements that I cataloged about models and the other 3 terms--theories, hypothesis, experiments--that were used in the slide:

• A theory (is a broad) model.
• A model (is a specific) theory.
• Hypotheses (inform/guide) experiments.
• It takes many hypotheses (to make a) theory.
• Hypotheses (generate) theories.
• Experiments (can inform) a model.

Practice: Connectivity work...just connect the 'dots'

With only the bulleted statements listed above, you can begin to do some of your own connectivity work. For example, the last bullet says, "Experiments (can inform) a model." The first bullet says, "A theory (is a broad) model." If you connect the ideas in these two statements, on your own you can create a new statement: Experiments (can inform) a theory (because a theory is a broad model). Congratulations...you've just practiced using formal logic.

There's nothing like an example...

If you're like me, I crave concrete examples. Let me offer you an example of the use of the four terms discussed above. Let's start with an observation...

• I planted seeds of a flowering plant in my front yard. The seeds were for the same species of plant, but they were seeds that were supposed to yield two different colors of flowers: red and white. The first year, plants with either red OR white flowers appeared. After a couple of years, I noticed that there were also plants with pink flowers.

As soon as I start asking questions like, "How did this happen?" or "Why are pink flowers appearing when I only planted red and white ones?" I am moving in the general direction of seeking a explanation. In other words, I am taking a stab at making some statements that could explain what I observed. Here's me imagining a few possible explanations for the appearance of the pink flowers...

• Explanation #1, aka. The Night Neighbor: My neighbor is sneaking into my yard at night and planting seeds of pink flowers in my flower beds.

• Explanation #2, aka. The Pooper: A non-human creature, say, a bird or a mouse, is depositing poop in my flower beds which has the seeds of pink flowers in it.
  

• Explanation #3, aka. Flower Power: The red and white flowers are somehow making pink flowers. I'm not sure how, but they're responsible for the appearance of the pink flowers.
  

Believe it or not, each of these explanations could be considered models. Why? Well, each of the 3 explanations suggest some type of explanatory "mechanism" (a human plants, a bird poops, flowers have 'flower sex') of an observed pattern (i.e., pink flowers appear when only red and white flowers are planted). Can each of these models explain the appearance of the pattern that we're observing in our flower beds? Absolutely. Can each of these models be used to predict the results of certain types of tests or experiments that we could design? Absolutely. Let me do these two things for Explanation #1:

• The Night Neighbor Model explains the appearance of the pink flowers (the pattern) by the actions of my stealthy, night-gardening neighbor (the cause).

• Can I generate a hypothesis from this model? Sure, and it might go something like this: IF I can figure out some way to keep my neighbor (the cause) out of my yard, THEN I should never observe any pink flowers in my flower beds (the pattern).

• Can I use this hypothesis to inform/guide the design of an experiment? Sure, here's one possible experiment: Send the suspected neighbor on a 3-year vacation (prevent the cause from acting!). Do you see any pink flowers (the pattern) while he's gone?
  

What if you DO see pink flowers while he's gone? Well, then maybe you have to consider other neighbors, or maybe you have to start considering The Flower Power Model or The Poop Model. I'm sure you can imagine some kind of IF/THEN hypothesis generated from The Poop Model ('If I can just figure out a way to keep the birds and mice from pooping in my flower bed, then...'). I'm also pretty sure that by now you can imagine how to use that hypothesis to design an experiment (I'm thinking about something cool like a diaper for birds and mice...) and make predictions for an experiment involving animals and diapers (e.g., The flower beds being attended by the diapered animals will not grow pink flowers, but the flower beds being attended by the diaper-less animals will grow pink flowers).

Moving towards a theory...

The big question is, after all these models have been tested, hypotheses generated, experiments designed, and predictions made...have we generated any kind of theory? In this short example, the answer is, "Well, sort of..."

Each time we test these different models we're getting closer to a 'proper' scientific theory of how to explain the pattern (the appearance of the pink flowers). However, in order to achieve scientific theory-like status the model MUST be able to explain MORE than just the appearance of pink flowers in MY flower beds...(and here's the important point)...

In other words, it must be able to explain the general appearance of the 'mixed' flower color phenomenon BEYOND the confines of my front yard! It must explain the mixed flower color phenomenon in places where there are flowers, but maybe no people; it must explain the mixed flower color phenomenon in places like greenhouses, where there are likely no birds or mice. See what I'm getting at here? Remember our statements from earlier: Theories are broad models; models are more specific theories. At this point, we're finally ready to summarize models with a few basic statements that may have more meaning to you at this point.

Scientific models....

• Explain patterns identified in data/observations/phenomena.
• Correctly predict the results of new experiments or observations.
• Are consistent with other ideas (e.g., other scientific models, beliefs, etc.)
  

Wrapping things up...

Interestingly enough, a big part of BS110 this semester will be loosely related to my Flower Power Model mentioned above. In fact, Professor S even started telling you about my Flower Power Model when he started talking about "meiosis" in Lecture 2 (see slide at left). Now, you can be sure that he probably won't use phrases like "flowers having flower sex," but nevertheless you'll be talking about certain models and theories that explain how pink flowers can result from, say, red and white colored ones.

As you talk more about meiosis in Lecture 3 and beyond, however, you could do yourself a huge favor by thinking of "meiosis" as The Meiosis Model (which is part of a broader Theory of Inheritance). Most scientists talk about concepts like meiosis SO often that they usually drop the "Model" or "Theory of" parts of their speech. A great example of this is "evolution." Yes, you still sometimes hear people say the "Theory or Evolution" or maybe even the "Theory of Evolution by means of Natural Selection," but for the most part scientists nowadays usually just say, "evolution." When scientists do this, you know that it's a concept/idea that has been or is currently pushing for consideration of being called a principle or scientific 'truth.'

I suspect we'll talk more about all of these things, and more, in future Blogs...