Wednesday, May 27, 2015

Does Mendel fit in contemporary high school curriculum?

Ken and I met Tuomas Aivelo in person last August when we were in Finland teaching Logical Reasoning in Human Genetics, though we'd known each other on Twitter for a while.  He's a PhD student in Ecology and Evolutionary Biology at the University of Helsinki, with a keen interest in how genetics is taught in secondary schools.  He frequently speaks and writes on this topic (e.g., here), and has been active in efforts to revise genetics textbooks in Finland and elsewhere.  He has a blog of his own, with a very large following (which I know because occasionally he links to our blog and it seems that most of Finland follows the link), but unfortunately for those of us who don't speak Finnish, it's in Finnish.  Google Translate has a long way to go to make sense of Finnish.  We're pleased today to have Tuomas's thoughts in this guest post on teaching genetics.

Does Mendel fit in contemporary high school curriculum?

By Tuomas Aivelo

It’s not too often one gets real eureka moments, but I had one of those last August. I participated on Logical Reasoning in Human Genetics course taught by Ken Weiss, Anne Buchanan and others in University of Helsinki. I promised to write this up for Mermaid's Tale (as Anne has been continuously complaining it's difficult to read my own blog Kaiken takana on loinen via Google Translate).

One of my self-set objectives for the course was to figure out what kind of curriculum high school (or as it's called in Nordic context – upper secondary school) biology should have. Finland is about to update its national core curriculum and I have a soft spot for genetic education so I have been involved in curriculum design.

The problems with genetics education are well-known: students have poor understanding of what genes actually are, how genes function and students are not able to perceive how 'gene' means different things in different fields of biology. Lack of understanding of scientific models is very pervasive in biology. While our peers in chemistry and physics education have done better work in explaining what scientific models are (i.e., “Bohr model”, “Standard model”), we have been overwhelmed by many different models for ‘gene’.

I remember when I for the first time understood that gene can actually mean multiple things. It wasn’t that long ago, probably in the beginning of my PhD studies. I wish somebody would have earlier told me this explicitly. Suddenly I understood most of the debates which transcend the fields of biology and involve genes are related to different meanings of genes in different fields rather than anything based on reality.

Scientific models are most of all helpful in making the reality more easily approachable for scientific inquiry. They are very powerful heuristic tools. The problem lies in genetic textbooks having several different scientific models of ‘gene’ which are not explicitly outlined as scientific models. A standard high school biology textbook can contain several, contradictory, gene models while not explaining why they are contradictory.

This all should cause worry as genetics education is actually very important. The most often discussed result of misconceptions in genetics is genetic determinism. Here I refer to the idea that traits are fixed by genes as genetic determinism. Genetic determinism excludes meaningful impact of environment out from the genotype-to-phenotype relation and it has never been scientifically sound theory.  (Genetic determinism can also refer to broader idea that genes have an effect on phenotype. This is obviously widely accepted idea. I’ve suggested the use of term “hard genetic determinism” to differentiate from “scientific genetic determinism."

School curriculum, teachers and textbooks are not the only problem, but also science journalism is rife with ”gene for” news. “A gene for a trait” is very deterministic view of gene function. The school should at least equip students with the tools to assess these news. The real problem lies in genetic determinism being implied to lead racists views and lack of intestest in personal behaviour in disease risk. In Logical Reasoning in Human Genetics course, Joseph Terwilliger also gave a good reason why every geneticist should be worried of genetic determinism. He attributed the people's worry of genetic privacy (for example, related to the insurance companies and job applications) to the overselling of genetic determinism: people feel unsafe sharing their genome because they think a meaningful understanding of them as humans can be simply read from DNA. Furthermore, this worry could lead to too strict laws which prohibit the use of samples and lack of consent for genetic studies.

So, the question is: what should we teach the students? At the moment, genetics education is dominated by Mendelian genetics. The canonical learning sequence in genetics starts with Mendel and his peas and works out pea pedigrees and segregation analysis. One of the central exercises is to figure out from pedigrees which kind of genotypes and phenotypes individuals have and which probabilities there are for a given phenotype to be expressed. This approach has been strongly attacked in last years. For example Michael Dougherty and Rosie Redfield (here and here) have attacked the Mendelian approach in genetics education in high school and introductory university courses, respectively.

There has also been defensive effort on the importance of the Mendelian approach. Mike Smith and Niklas Gericke argued that Mendel's work belongs to the general scientific literacy as it is a central part of our culture. Furthermore, they suggested that the chronological approach to genetics increases student motivation and understanding of the nature of the science. They also suggested using Mendel as an example that has heuristic value as a simple model of inheritance.

I'm not impressed with these arguments. If we have a culture of learning genetic determinism in school, purging Mendel from contents would only make it easier to change this culture. Historical or chronological approaches to different fields of biology are rare: normally genetics and evolution (e.g., Mendel and Lamarck-Darwin) are the only ones widely used. Furthemore, it is highly contentious if the Mendelian ”laws of inheritance” are actually a simple model of inheritance which could be used for a more detailed view of genetics. Simple, yes, but they work in very limited context. In fact, in contemporary teaching, while “the historical” aspect is often included, the contemporary part of the genetics is rather limited. In oursurvey of gene models used in Finnish high school textbooks, we did not find any modern models: all the models used were formulated before 1960s. We have an obvious problem: history is taking space from the contemporary understanding of genetics.

This is the point, when eureka comes to help: it’s not about Mendel or history. It’s not about the contents. It’s about what we are teaching at a more fundamental level.

Until now, Mendel and his peas have been used to answer how traits are inherited. In fact, the current Finnish curriculum notes that the course contents should address “the laws of inheritance”. The right question, as I see, is how the inheritance of the traits is studied. Let’s not focus on how inheritance happens but rather how we can study this phenomenon. Here we are at the heart of one of the central problems in science education in general. In school, we learn how things are, rather than how we know.

When we present Mendel as an example of how early inheritance patterns were studied, it brings all the contentious concepts into the right context. 'Dominance' and 'recessiveness' are highly useful concepts in the Mendelian context. This approach would bring a natural possibility to discuss why Mendel used peas for his studies and how useful these studies are to infer the inheritance patterns of other traits. This would highlight what differences there are in inheritance of traits in humans and peas.

Obviously, this approach would lead to less dominant place of Mendelian inheritance in biology classes as then Mendel would be only one of the many approaches to study the inheritance of traits. In any case, it would still make it possible to have a historical approach or other innovative approaches to the teaching. It allows for an easy approach to the models in genetics and what they mean without making Mendelian inheritance the model, but rather one of many models.

The current trend in science education is to put the emphasis on the Nature of Science (e.g., what science is and how it is done) and socioscientific issues (e.g., how knowledge of science can be used by active citizens). The question of how things are studied obviously suits the Nature of Science emphasis nicely and it shouldn't be too much of a stretch to expect that the understanding of how inheritance is studied makes it easier to have informed decisions on inheritance.

One of the real problems in biology education is that the teaching strategies are rarely studied in classroom context. Only empirical testing can answer if this emphasis on why more than what and how leads to better learning in genetics.


Holly Dunsworth said...

Wow! This is terrific! I hope it's widely read by anthropologists who teach and who write about genetics, theirs or others'.

Ken Weiss said...

Very fine, Tuomas! Finnish students will be very fortunate if your ideas are adopted!
Actually, in striving for a 'standard model' we may have made two serious mistakes. First, Mendel was about probabilistic inheritance of deterministic causal Elements. We confound that and assume that causation is inherited, rather than genetic elements. Those elements, once inherited, need not be causal in Mendelian ways. So talking of inheriting traits (in any way) confounds transmission (which is generally probabilistic) with causation which is generally not very deterministic.

We should not speak of Mendelian inheritance of traits, but only of Mendelian inheritance of genes....and even then we should be careful to note that Mendel studied single 'genes', but inheritance overall is of a mix of chromosomal segments, relative to what parents have, which is not so simple as just 1/2.

But for many practical and other reasons, it is easier to think of Mendelian inheritance of traits, and that leads us far away from having an adequate 'standard model'.

I think you're completely right about the importance of having students (and the general public) understand these things, but people seem to hunger for easy, simple it's a challenge! We're very glad you are taking it seriously and trying to do something about the problem.

tuomaz84 said...

Thanks Holly and Ken!

You raise important points. I looked for five historical gene models (as formulated by Niklas Gericke: Mendelian, Classical, Biochemical-Classical, Neoclassical and Moder) in the textbooks and Mendelian model was really abundant - 34% of all models. I think it's partially due to one of the properties of the Mendelian model being "there is no distinction between genotype and phenotype". This idea is rife in textbooks: it's not even "gene for" but even stronger: blood group gene, eye color gene, pea color gene and so on. In that context, Mendelian inheritance of genes is a bit anachronistic. Mendel was effectively studying traits and genes were just something which came along.

The later gene models have the idea that genotype and phenotype are fundamentally different. Some textbooks had peculiar cognitive dissonance where they first declared that traits are not inherited by only genes are, and then in the next page they talk about inheritance of certain traits. I've been trying to weed out this talk of inheritance of traits, but it's not easy. In English, I've seen "passing of traits" as a substitute for this phenomenon that parents and offspring seem to look alike, epi-inheritance or some sorts. We need a word for that phenomenon, as it's intuitive concept and a logical starting point for genetics: Why do you look like your parent? Why you differ from your parents? said...

I think modern genetics courses should talk in terms of 'gene regulatory network' instead of genes. I am putting together an article related to that and will mail to anyone interested.

Holly Dunsworth said...

Because I care about these issues, and because they're fundamental to understanding evolution and overcoming misconceptions, I try to teach all genetics/inheritance/gene expression in my Intro BioAnth (aka Human Origins) course from that exact point of inquiry: Why do you look like your parent? Why you differ from your parents?

And I keep hammering on that question/point as we go along.

This post is significant and important (and all the superlatives, to my mind!) not just for high school biology, but prior to that and also at the college and university level too. This is incredibly important and thank you again for writing about it, for working on it so hard (in the wrong country! haha) and for sharing your work with us here on our blog.