Laboratory components are often integral parts of both K-12 and college science courses. I certainly had a lot over the course of my science education; 5 courses with labs in high school, 8 in college. But for the overwhelming majority of them, I was essentially following a recipe and doing by rote things which had already been done and where the answers were already known. It was only in science-fair-style projects that I typically had any control over the questions I was asking, or how I would go about trying to answer them. But science education doesn’t have to be like that. Inquiry-based science practice is a growing part of the recommendations for science education1 2. Thankfully, computational tools are making these practices more accessible.
Avida-ED provides a great platform for students to do actual scientific research within the settings of a course on general, population, or organismal biology. Students have the ability to explore questions which they devise themselves, in part because they do not have the cost associated with consumable materials that most natural science experiments do. Further, because computational data can be generated so much more rapidly than physical experiments, students can perform exploratory work to generate hypotheses, and then gather sufficient data to test them. We have found inquiry projects work well when students work in small teams.
Classroom assessment studies work from the Avida-ED group have shown a benefit to students conducting their own active research projects using this software. Students have to grapple not only with defining questions and designing experiments, but also with figuring out what data to collect, how to analyze this data, and how to present findings to others.
How does one go about using Avida-ED for classroom research projects? Great question; that’s what this post is about to lay out.
Within Avida-ED, there are several parameters that are easy for the user to change, including:
- World Size (which influences population size)
- Mutation Rate
- Which tasks are rewarded
- Which organism(s) is/are used as the ancestor
Likewise, there are multiple different responses that users can track. These range from some very simple ones, such as:
- Fitness, Merit, or Gestation Time at a given time point
- Whether any organisms in the environment perform a given function
- How many organisms perform any given function
To more complex ones such as:
- Maximum Fitness, Merit, or Minimum Gestation Time in the population
- Number and complexity of tasks performed by the most fit organism
- Relative abundance of descendants from different ancestors
To ones that require real-time monitoring of populations like:
- Time at which a function was first observed
- Whether new tasks appear in a background already performing many other functions
- Whether gain of a new function involves loss of a previous task
Part of the scientific process involves first being able to show a correlation between an outcome and a set of conditions. Therefore, I highly encourage students to pick a single parameter to manipulate at a time, and a single measured response to use for a particular experiment. Once a correlation is established, researchers can then gather additional evidence to test not only correlation but also the causation that is implied – changing a parameter leads to a different outcome, so therefore the parameter likely contributes to the outcome. The fewer parameters we change, the more confident we can be that the difference in the results flows from a specific change we made. It is especially instructive for students to be able to run experiments of this sort to see for themselves how evolutionary processes can produce adaptations. Instructors may find value in a simple demonstration for a class where the instructor changes many parameters at once, gets a different result, and then asks the class which of the parameters they think contributed to the difference and how they would be able to tell.
Overall, programs like Avida-ED offer a way for students to pose real questions (and deal with all of the challenges in figuring out what data to collect, how to interpret it, etc.) just as researchers regularly face in their work. This takes a class experience from being essentially an exercise in recipe-following – a “lab experience” in name only — to an activity where they engage in authentic science practice. In a later post, I’ll outline some of the basic types of data analysis likely to be relevant to a wide range of introductory research questions.