Which of the following is not a strength of scientific models?

Question 2 options:

Question 3 (1 point)

Saved

Which of the following is not a weakness of scientific models?

Question 3 options:

Question 4 

Researchers are interested in the spread of the waterborne pathogen that is responsible for Typhoid fever. Even when only a fraction of a population visits a contaminated water source, individuals infected with Typhoid fever can transmit it to others whom they interact with. The researchers believe that even simple rules for interaction at an individual level can lead to a total epidemic at the group level.

The researchers are curious how aspects of the physical environment (like narrow passageways) might influence the spread of Typhoid. They plan to build a model where people move around a space, following simple rules of behavior, and when they interact with an infected person they have a chance of catching Typhoid. For an experiment, the researchers want to change the landscape that the simulated people move around in. In one run there might be open space to move around in all directions, and in another run there might be a narrow passageway between areas.

Which type of modeling would be most appropriate for investigating this phenomenon?

Question 4 options:

Question 5 (1 point)

After setting up the scientific model on Typhoid Fever, researchers start testing it. They change the values for things like the rate of infectivity, the number of people, etc. They notice that even when the rate of infectivity is set to zero, simultaions still result in a Typhoid epidemic. The researchers are confused why this is happening, since no one should be catching Typhoid when they set the rate of infection to zero. They suspect there is an error in the simulation code.

Which issue are they having a difficulty with right now?

Question 5 options:

Question 6 (1 point)

Alleles are specific forms of a gene. A gene with one allele might give you dark colored hair, whereas a different allele might give you light colored hair. There is a mix of different alleles in the world and if animals reproduced at random the proportion of each allele should stay the same over time.

Hardy and Weinberg test for randomness of reproduction with a mathematical equation, called the Hardy-Weinberg Equilibrium Model. It is often written as (p2) + (2pq) + (q2) = 1. The authors argue that proportions of each allele will add up to one and will stay the same over time if reproduction is truly random. If data suggest that allele proportions do not stay the same over time, the Hardy-Weinberg model will conclude that something is influencing population genetics, like mate selection or genetic mutations.

When Hardy and Weinberg test genetic data for randomness, are they doing scientific modeling?

Question 6 options:

Question 7 

Heisenberg and Schrodinger independently developed scientific models based on the theories of quantum mechanics. Their models have been shown to be mathematically equivalent, suggesting that these two researchers have represented quantum mechanics in the same way.

Does this indicate good verification or good validation of their models? (Hint: Does it mean that their views of the system are similar? Or that their models are accurate?)

Question 7 options:

Question 8 (1 point)

Overfishing in Lake Victoria has led to a dramatic decrease in the number of Nile Perch fish over recent years. Researchers want to make a model of overfishing so that they can better understand overfishing dynamics and make policy recommendations to keep the fish population healthy.

Based on a collection of existing studies, the researchers believe that the number of fish in the lake is probably part of a complex feedback loop: More fish leads to more fishermen, which leads to less fish and then less fishermen. These opposing forces should balance out and reach an equilibrium, with an appropriate number of fish and fisherman. Unfortunately, there seems to be a delay between a decrease in the fish population and a decrease in fisherman, so people keep overfishing before they realize the fish are endangered.

The researchers want to experiment with different fishing policies to prevent extinction fo the Nile Perch fish. In researching the policies, they also want to learn how different fishing policies might change the feedback loop and the delays before fishermen reduce fishing.

Which type of modeling would be most appropriate for them?

Question 8 options:

Question 9 (1 point)

In building their scientific model of overfishing, the researchers are informed by existing research on many aspects of the overfishing problem. For example, they know the number of fish exported each year since 1990, and they have estimates of the number of fishermen each year since 1995. However, the researchers have only a rough estimate of the exact fish population at any point of time. They think their rough estiamtes are probably pretty accurate, because the model seems to be working correctly: When they run a simulation, all other aspects of the model (like exporting and fishing levels) seem to match up with real data.

However, the researchers are still cautious about the fish population, because they don’t know if their estimates are accurate. They try running the model with a bunch of different values for a fish population to see whether “it matters” if they are off by a bit. Unfortunately, the model results change dramatically when the fish population is adjusted even slightly, so the model appears to be very sensitive to this parameter.

Which issue are they having difficulty with right now?

Question 9 options:

Question 10 (1 point)

Of the following research questions, which would be most appropriate for investigating with our scientific model of overfishing?

Question 10 options:

Question 11 (1 point)

Neoclassical Economics is the study of how prices, outputs and incomes are distributed in a market. This field uses a lot of complex mathematical equatiosn, most of which have simplifying assumptions in them, like perfect information and perfect rationality. (When buying tomatoes at the store, I am assumed to know how much all other tomatoes cost in the world, and I am assumed to consistently make the most rational choice possible in my tomato purchase.)

Today, many of these economic models allow you to plug in information for different parameters (i.e., numbers or terms in an equation). You can set the value of a product and the demand for that product. When you change these parameters you are basically experimenting on the model, and the results of your experiment can generate more knowledge about underlying processes, like how supply and demand may influence each other for a certain type of product. These models are not very good at making accurate predictions (like the 2007 recession), but they can help us learn about forces in the economy.

Does this type of research fit within our course definition of scientific modeling?

Question 11 options:

Question 12 (1 point)

The simulated results from an economic model can be compared with historical data to determine the model’s value. For example, you might see whether your scientific model can reproduce an economic collapse, based on information about how our our economy was functioning right before 2007. You could “tune” your model to replicate the data leading up to 2007, while hiding away your data from 2007 onward. You could then run your simulation past 2007, and you could see if your model was able to recreate the recession trends. If your model performs well for years after 2007, even when you didn’t use the post-2007 data during phases of model tuning, it would be a good sign.

Does this type of model testing indicate verification or validation?

Question 12 options:

Question 13 (1 point)

Of the following research questions, which would be most appropriate for investigation with the economic model described earlier?

Question 13 options:

Question 14 (1 point)

Immunosuppressant drugs are important for patients who are getting organ transplants. With immunosuppressants, a patient’s immune system is less likely to attack the new organ.

Let’s say researchers have developed a new immunosuppressant drug, and they are trying to learn how and why it works before doing clinical trials with humans. The FDA has approved the drug for veterinary surgeries, so the researchers are able to consider dogs (who need surgeries) as scientific models. As expected, they find that that the drug helps dogs accept organ transplants successfully. But an interesting side effect is also noticed: the drug reduces symptoms of autoimmune diseases in the dogs, like arthritis and multiple sclerosis. After more testing, the researchers find that dogs taking the drug have significantly higher levels of the hormone cortisol.

Researchers start to believe the new drug works mainly by increasing cortisol levels, which trigger suppression of the immune system. But before doing their clinical trial with human organ transplants, they first plan to give small doses of the drug to humans with arthritis. If the humans show higher levels of cortisol and lower symptoms of arthritis, the researchers will be more confident in the findings from their animal model. (Hint: They will have some concrete evidence that the drug works similarly in humans as it does in dogs.) Researchers will then do a clinical trial with humans who are undergoing organ transplants.

What type of model testing are the researchers employing?

Question 14 options: