Evolving Minds

The mental maps we never draw—why how students visualize genes to evolution matters

Imagine facing a biology student who confidently explains natural selection, then casually mentions that "giraffes got long necks because they needed to reach higher leaves." This isn't a isolated case—research shows even students who can parrot back correct evolutionary terminology often harbor deep misconceptions about how evolution actually works. The challenge of teaching evolution effectively has plagued science education for decades, with studies consistently revealing that students struggle to connect genetic-level processes with population-level outcomes ¹ .

In recent years, an innovative approach has emerged from college classrooms: asking students to draw their understanding. Not just sketches of cells or chromosomes, but comprehensive conceptual models that illustrate the connections between genes, variation, selection, and evolutionary change. These drawings serve as windows into students' mental models, revealing both sophisticated understanding and persistent gaps. Educational researchers at institutions like Michigan State University have pioneered the use of these "gene-to-evolution" (GtE) models to analyze how students' understanding changes throughout introductory biology courses ² .

The results have been transformative. By examining these student-generated models, educators are discovering why certain evolutionary concepts remain persistently difficult to grasp and developing more effective strategies to help students build accurate mental frameworks. This research comes at a critical time—as genetic technologies advance rapidly and society faces evolving biological challenges from antibiotic resistance to climate change, citizens need a solid understanding of evolutionary principles more than ever before.

Modeling mind shifts: How conceptual models reveal learning pathways

What are conceptual models?

In the context of biology education, conceptual models are visual representations that show how different biological components relate to one another. Unlike traditional diagrams that might illustrate anatomical structures or biochemical pathways, these models require students to identify key elements and specify the relationships between them—often using boxes for concepts and arrows to show connections ² . For example, a student might create a model showing how a DNA mutation leads to protein variation, which results in differential survival, ultimately causing population-level evolution.

Why models matter in science education

Scientists regularly use models to construct and test hypotheses, evaluate evidence, warrant arguments, and identify system unknowns. Models are foundational to the practice and epistemology of biological science ² . When students engage in model-building, they're not just learning biological content—they're engaging in authentic scientific practices that mirror how biologists actually work.

Research in the learning sciences suggests that this approach is particularly important in large-enrollment introductory biology courses, where traditional assessments often emphasize factual recall over conceptual understanding and scientific reasoning ² .

The MSU experiment: Tracking conceptual change through model-building

Methodology and approach

At Michigan State University, researchers implemented an instructional approach that uses system models as a way to teach, learn, and assess students' understanding in introductory biology ² . Over one semester, students iteratively constructed, evaluated, and revised "Gene-to-Evolution" (GtE) models designed to promote understanding of the connections linking molecular-level processes with population-level outcomes.

The study analyzed change in students' models throughout the semester in two major content areas typical of introductory biology: principles of genetics and ecosystem ecology. A component of genetics instruction included student construction of conceptual models of gene expression, while ecology instruction included models of ecosystem dynamics ² .

Students created initial models at the beginning of the semester, received targeted feedback, and revised their models at multiple points throughout the course. The research team collected and analyzed these models using a mixed-methods approach, quantifying both the complexity and correctness of the models while also looking for qualitative patterns in how students represented key biological concepts.

Key findings: What changes in student models

The analysis revealed fascinating patterns in how students' models evolved throughout the semester:

• Model correctness increased significantly from initial to final models, showing that students were developing more accurate understandings of evolutionary processes.
• Model complexity peaked near mid-term then subsequently declined, suggesting that as students developed better understanding, they created more parsimonious models that eliminated irrelevant information ² .
• Biological language became more precise over time, with students using more appropriate and accurate terminology to explain relationships between concepts.

Perhaps most encouraging was the finding that lower-performing students (those who entered the course with lower mean GPA) showed the greatest relative gains in model correctness, effectively closing the achievement gap with the highest-performing students by the end of the semester ² . This suggests that model-based pedagogy may be particularly beneficial for students who struggle with traditional science instruction.

Why evolution is hard: Cognitive barriers to understanding evolutionary processes

The challenge of genetic variation

One of the most persistent findings in evolution education research is that students struggle to understand the origin and role of genetic variation in evolutionary processes. Despite instruction, many students continue to believe that evolutionary change happens because individuals "need" to adapt to environmental pressures, rather than understanding that selection acts on existing variation ^{1 3} .

This study found that even after instruction, approximately one-third of students still did not include mutation in their models of evolutionary change ² . This is significant because mutation is the ultimate source of genetic variation upon which evolutionary processes act. Without understanding this fundamental concept, students develop incomplete mental models of evolution that emphasize change in individuals rather than changes in population allele frequencies.

The neglect of non-adaptive processes

Another critical finding from related research is that students overwhelmingly favor adaptive explanations for evolutionary change while largely ignoring non-adaptive processes like genetic drift ¹ . Even after explicit instruction about genetic drift, students rarely incorporate it into their evolutionary explanations.

When researchers used clinical interviews and written assessments to investigate how students reason about evolutionary causation, they found non-adaptive factors to be "remarkably uncommon" in students' explanatory models ¹ . This aligns with what biologists call the "adaptationist paradigm"—the tendency to assume that all traits are adaptive products of natural selection, a bias that even professional biologists must guard against.

Cognitive construals: Deep-seated thinking patterns

Research into the cognitive psychology of learning evolution has identified several intuitive thinking patterns that interfere with accurate understanding:

• Teleology: The tendency to explain biological phenomena in terms of purposes or goals (e.g., "populations want to evolve to get better")
• Essentialism: Treating categories as having underlying essences and sharp boundaries (e.g., assuming all individuals of a species are identical)
• Anthropocentric thinking: Using humans as the reference point for all biological comparisons ³

These cognitive construals represent deep-seated thinking patterns that develop early in childhood and persist into adulthood, making them particularly resistant to change. They appear across multiple representations of student thinking, including interviews, written responses, and drawn models ³ .

Educational implications: How model-based pedagogy transforms biology education

The power of iterative model-building

The research from Michigan State and other institutions suggests that having students repeatedly build, evaluate, and revise models of biological processes leads to more sophisticated understanding. This iterative modeling approach allows students to test their mental models against biological reality, receive feedback, and make adjustments—a process that mirrors how scientists actually work ² .

This approach represents a significant departure from traditional biology instruction, which often emphasizes content coverage over conceptual depth. By spending extended time having students work with core concepts like the connection between genes and evolution, instructors can help students develop more robust mental frameworks that will support future learning.

Assessment that drives learning

Another critical implication of this research concerns assessment. Traditional multiple-choice exams often emphasize factual recall and can reinforce the idea that biology is about memorizing correct answers rather than constructing evidence-based explanations. As one researcher noted, "Our exams – not our learning goals – tell students how to study for our classes" ² .

Model-based assessments provide a powerful alternative that can reveal student thinking in more nuanced ways than multiple-choice questions while being more practical to grade than extensive essays ² . These assessments can be designed to specifically target difficult concepts like the origin of variation or the role of non-adaptive processes in evolution.

The importance of multiple representations

Research comparing different formats for assessing student understanding—including interviews, written responses, and drawn models—suggests that each format reveals different aspects of student thinking ³ . For example, students may express more sophisticated understanding in interviews than in written responses, perhaps because the interactive nature of an interview allows for clarification and probing.

This suggests that instructors should provide students with multiple ways to demonstrate their understanding rather than relying on a single assessment format. It also highlights the importance of helping students develop representational competence—the ability to create and interpret multiple representations of biological processes.

The future of evolution education: AI and digital modeling tools

Digital evolution and simulation tools

Emerging technologies offer promising new approaches for teaching evolution concepts. Digital evolution systems like Avida-ED allow students to observe evolutionary processes in action by creating populations of digital organisms that evolve based on the same principles that govern biological evolution ² .

Digital Evolution in Action

Simulation tools allow students to manipulate parameters and observe evolutionary outcomes in real-time.

Research using these tools has shown that they can help students develop more accurate understanding of evolutionary processes, particularly when combined with model-based pedagogy. However, certain evolutionary principles—especially those involving genetics concepts like the origin of variation—remain challenging even after instruction with these tools ² .

The promise of AI in biology education

Recent advances in artificial intelligence are beginning to impact biology education as well. Tools like Evo 2, an AI system trained on genomic data from over 100,000 species, can predict the effects of genetic mutations and even design novel genetic sequences ^{4 5} .

While these tools are primarily used in research settings currently, they have potential educational applications. Imagine students using AI systems to test hypotheses about evolutionary processes or to visualize how genetic changes might affect phenotypic outcomes across generations. These technologies could make abstract evolutionary concepts more tangible and accessible to students.

Virtual laboratories and interactive simulations

Virtual biology labs that simulate population genetics processes are already being used in some classrooms. These tools allow students to manipulate parameters like population size, selection pressure, mutation rate, and migration to observe how these factors affect evolutionary outcomes ⁶ .

For example, the PopGen Fish Pond simulation allows students to conduct virtual experiments that violate the assumptions of Hardy-Weinberg equilibrium, helping them understand how factors like small population size, selection, mutation, migration, and non-random mating affect evolutionary change ⁶ . Such tools provide valuable opportunities for inquiry-based learning, particularly in settings where traditional wet labs are impractical.

Conclusion: The evolving understanding of evolution education

Research on how students' gene-to-evolution models change throughout introductory biology courses represents more than an academic curiosity—it provides crucial insights into how learning actually happens and how we can better facilitate it. The findings reveal both the persistent challenges in evolution education and promising pathways forward.

What emerges most clearly from this research is the need to engage students in authentic scientific practices like model-building, rather than simply presenting them with scientific facts to memorize.

The work also highlights the importance of specifically addressing difficult concepts like the origin of variation and non-adaptive evolutionary processes. These concepts don't come naturally to students, and they won't develop accurate understanding unless instructors explicitly target these areas with appropriate pedagogical strategies.

As biology continues to advance at an astonishing pace—with breakthroughs in genomics, synthetic biology, and evolutionary medicine—the importance of evolution education only grows. By leveraging research on how students learn evolutionary concepts and using tools like model-based pedagogy, digital simulations, and perhaps someday AI assistants, educators can prepare students to understand and engage with the biological challenges and opportunities of the 21st century.

The journey to understand evolution is itself an evolutionary process—one that involves variation in teaching approaches, selection of effective strategies, and inheritance of successful practices across generations of educators. Through continued research and innovation in evolution education, we can hope to see increasingly sophisticated student models of evolutionary processes—both on paper and in their minds.