Definitions, Benefits, & Underpinnings
Animation – a dynamic visualization that can display the components of a system and their operations as a sequence of events occurring in time and space. An animation either play at a constant speed and for a set amount of time that cannot be manipulated by the learner, or the learner can speed up, slow down, stop, view and review sections at their own pace.
Cognitive load – the effort associated with a particular task assuming that the human brain has limited capacity to attend to information. Sweller (1998) proposed the cognitive load theory which suggests a learner can process only a limited amount of information at any given time; instruction must therefore limit extraneous or unneeded material to avoid overloading the learner and reducing student learning.
Constructivism – a theory of learning and teaching that suggests knowledge is not passively received, but must be constructed by the learner through active engagement in sense-making and relating new information to existing knowledge and experiences.
Debrief – a process that can follow learning activities (e.g., modeling) that allows participants to reflect upon performance and to assess acquisition of knowledge, skills, attitudes, or behaviors.
Deliberate practice- refers to purposeful and systematic practice requiring focused attention with a goal of improving a specific skill like modeling.
Epistemology – refers to a theory of knowledge, particularly how we know things. Epistemology considers the origin, nature, and validity of knowledge, and is thereby distinguished from opinions or unjustified beliefs.
Model – a model is a simplified representation of a natural or social phenomenon, which, depending on its purpose, emphasizes specific aspects of the system (e.g., components, processes, or system function). Often, models are used to represent elements of a system that are abstract, invisible, or exist on a scale that is too small or too large to be perceived. Models serve to explain and/or predict phenomena and are thereby distinguished from other visualizations that merely represent or recreate.
Model-based instruction (MBI) – instruction that engages students in creating, using, or evaluating models as a means to learn content. Students may create models of well-known biological phenomena, evaluate existing models, use models to generate and test hypotheses, or build new models based on evidence. Proponents of MBI argue that modeling is a core scientific practice and should thus be a core instructional approach.MBI has alternatively called modeling instruction, modeling-based instruction, or modeling pedagogy.
Model-based learning – the construction of mental models of phenomena occurring in the learner’s mind in response to particular tasks (Gobert and Buckley, 2000). Mental models are personal, internal representations of a system or phenomenon, which develop in the learner’s mind and may be externalized in multiple ways, including verbal and graphic depictions. External models, produced by learners as representations of their mental models, can be shared and used for social construction of knowledge. Model-based learning is an iterative process that involves the creation of mental models, and their subsequent evaluation and revision as the learner acquires new knowledge that extends and/or modifies a mental model.
Scaffolding – instructional techniques that move students progressively towards stronger understanding. Scaffolding can come in two forms, either by (1) providing assistance that is progressively removed as students no longer need it, or (2) progressively increasing complexity, difficulty, or sophistication of an activity over time. Both of these scaffolding techniques provide support to students attempting difficult tasks and are intended to decrease or prevent students’ frustration, intimidation, and discouragement during the learning process.
Scientific practices – as defined by Next Generation Science Standards, any behavior or activity a scientist uses, such as modeling, to build knowledge about the natural world.
Simulation – An imitation of a real-world process or system. Simulations allow users/participants to manipulate variables or views to observe changes and make/test predictions of phenomena. The phenomena normally explored using simulations can be difficult to directly observe because of inaccessibility due to temporal or spatial scales.
Structure-Behavior-Function (SBF) – A theory put forth by Goel & Stroulia (1996) for describing functional aspects of engineered devices based on causal relationships among system components. The framework was subsequently adapted for modeling complex systems in biology. SBF models represent system components as structures in boxes; structures are connected by labeled arrows (behaviors) that reflect a mechanism or relationship. Taken together, structures and behaviors explain a function or output of a system.
System – a system (natural or engineered) is a set of interconnected parts that, together, form an integrated whole. The function of a system is dependent on (and can be explained by) the structure, actions, and interactions of its parts. Examples of biological systems exist at all scales of organization (cells, organisms, and ecosystems are systems).
Systems thinking – a way of thinking which focuses on identifying and describing the interconnections between and among the parts of a system in order to understand the whole of the system.
- Modeling is widely regarded as foundational to the practice of science. Models may be abstract or concrete and seek to represent reality. In contrast, a theory is a generalized statement supported by extensive and rigorous testing. Models can be used to test a theory.
- Practicing scientists perceive models differently than students. These perceptions vary with respect to both the utility and value of models, as well as the origin of data and ideas that comprise the model. This disparity in model perceptions can be exacerbated by instructors’ selection of models for use in instruction, the language he/she applies to modeling, and by the types of model-based activities students engage in.
- Students tend to value models that are simple and easy to understand, and frequently perceive provided models (e.g., from textbooks) as correct versions of scientific explanations. As a result, students tend to devalue model complexity and disregard the credibility of self-constructed models or models that generate outcomes that are ambiguous or unexpected, and rarely associate the practice of modeling with scientific epistemology.
- There are a range of opinions and strategies regarding model classification and naming schemes. It is not uncommon for a particular label or model name to be used to represent fundamentally different types of model-based representations. When using, teaching, or reading about models/modeling, it is important to have a clear understanding of the rules and conventions associated with different model types, as they may be relevant for construction and/or interpretation.
- Instructors must be thoughtful in their selection of models and model types for use in their instruction. Model selection should be purposeful and guided by specific objectives for learning.
- Practical models may make teaching easier, but compromising complexity might impact students’ perceptions about role/utility of models
Gilbert, S. W. (1991). Model building and a definition of science. J Research Sci Teaching 28: 73-79. In this foundational paper, Steven Gilbert puts forth a definition of science as “a process of constructing predictive conceptual models”. His proposed definition serves the purpose of uniting the process of science with the products of science, and identifies model building as a superordinate skill. In this view, the purpose of research is “to produce models which represent consistent, predictive relationships.” Gilbert’s proposal for a model-based definition of science derives from his findings from a large-scale, survey-based study about students’ perceptions about both models and science. He suggests that defining science as a process of model building could promote science literacy by making explicit the concept that knowledge is constructed – consistent with constructivist philosophy that knowledge must be built. Further, a model-based definition of science relies on a broad conceptualization of “model” that is capable of transcending disciplinary bounds, thereby promoting a perception of unity among disciplines about the essence of what is considered science. Gilbert suggests that such a definition could have potential to counter stereotypes held by many undergraduates that science is rigid in its adherence to logic, singular in its methodological approach, conducted in isolation, and immune to influences from society and social contexts.
Justi, R. S., & Gilbert, J. K. (2000). History and philosophy of science through models: Some challenges in the case of ‘the atom.’ Intl J Sci Ed 22(9): 993–1009. In this paper, the authors propose that models and modeling can serve as a vehicle by which history and philosophy of science can be represented within science curricula, and consequently develop students’ understanding about the role of models and modeling in the epistemology of science. Models are viewed as “between scientific theory and the world-as-experienced”, and therefore play a pivotal role in bridging understanding between these realms. The authors make a case for curricula that enables a progressive and historical examination of models as a way to foster a more model-based view of science. The authors use the specific case of atomic models and curricular approaches to teaching atomic structure as a way to develop and illustrate key components of their argument. Perhaps one of the most significant contributions of this paper is in connecting prior findings by Grosslight et al. with their observations about modeling perceptions. In doing so, they are laying the groundwork for subsequent research that more rigorously examines perceptions held by teachers and students about models and modeling in the context of science.
Louca, L. T., & Zacharia C. Z. (2012). Modeling-based learning in science education: Cognitive, metacognitive, social, material and epistemological contributions. Ed Rev 64(4): 471–492. In this paper, the authors provide an overview of research on Modeling-based Learning in science education. Although one can find descriptions of multiple types of model-based teaching and learning approaches in the literature, this paper focuses specifically on Model-based Learning, defined as an approach that utilizes students’ construction of models to help leaners build an understanding of a physical phenomenon’s mechanism. This is in contrast to approaches that rely on models as visual aids and simplified representations of complex phenomena, but do not engage students directly in the practice of building, evaluating, or revising their own models. The authors associate 5 key aspects with Model-based Learning: (1) cognitive – the contribution of modeling to students’ conceptual understanding and cognitive skill development; (2) metacognitive – students’ abilities to monitor and reflect on their own work and understanding; (3) social – the interactions and discourse that take place among students and between teachers and students; (4) material – the role of specific tools and/or curricula in providing students opportunities to engage in Model-based Learning; and, (5) epistemological – for example, students’ awareness about developing and using models for representing phenomena and/or defending or rejecting claims. The authors discuss literature to define the modeling process involving 4 phases: (a) making systematic observations about a phenomenon, (b) constructing a model based on those observations and experiences, (c) evaluating the model for its potential to explain or predict, and (d) revising and applying the model in new contexts. For instructors, as well as researchers, thinking about the range of ways learner can and should engage with models will promote a more holistic way of thinking about modeling practice well beyond one-and-done construction activities.
Harrison, A.G., & Treagust, D.F. (2000). A typology of school science models. Intl J Sci Ed, 22(9):1011-26. This paper provides an in-depth analysis of analogical models – the type most frequently used in the context of science teaching. The authors argue that models and modeling are foundational components of science instruction, but instructors’ choice of models and model-based analogies should be thoughtful and motivated by explicit purposes driving their selection. For example, one must consider the appropriateness and familiarity of the analogical features of selected models from the perspective of students – at both surface and deep levels. The authors propose a classification framework that is grounded in literature on model-based learning, as well as their own empirical observations of students and teachers in model-based learning contexts. Broadly, model types are grouped into 4 categories: scientific and teaching models, pedagogical models that build conceptual knowledge, models depicting concepts and/or processes, and personal models. Within each of these larger categories, model types are further subdivided into lower, more descriptive levels in the hierarchy. The authors provide rich descriptions and a theoretical rationale for their proposed groupings, alongside specific recommendations for teaching and learning allied with each model type.
- Science uses models to explain observations, describe theories, explore complex systems, make predictions, and communicate ideas in general. Modeling is a core practice of science and as a result, should be a component of undergraduate biology education.
- Biologists regularly use a wide range of models, but it is not yet clear which models are essential to undergraduate biology instruction. There is a need for biology educators to identify core biological models. Doing so would support modeling activities across the biology curriculum.
- Model-based instruction (MBI) is a theoretically-grounded pedagogical approach that engages students in constructing and evaluating models of biological phenomena, analyzing and revising models, and using models to argue scientifically.
- The phases of MBI can include eliciting student prior ideas, collaborative model generation, feedback, and iterative modeling.
- Research on the implementation of MBI in undergraduate biology courses is currently limited. Further research is needed to identify both the best practices of implementation and how MBI promotes science process skill development.
- Instructors should explicitly discuss the models they regularly use in the classroom.
- Students’ perceptions of models are diverse, from viewing models as realistic representations to recognizing models as abstract representations that are tentative and testable. Instructors can help students expand their understanding of scientific models.
- Instructors may also help students recognize that all models have limitations (all models are wrong, some are useful).
- Explicit instruction can also help students see models as more than explanations of phenomena. For example, models can be used to build and test predictions.
Baze, CL, & Gray, R. (2018). Modeling Tiktaalik: Using a model-based inquiry approach to engage community college students in the practices of science during an evolution unit. Journal of College Science Teaching, 47(4), 12-20. In this article, the authors describe a unit of instruction in a non-majors biology course that centered on the appearance of tetrapod-like features in Tiktaalik. Using model-based instruction (MBI), students worked in small groups to build an initial model and written explanation of “what caused Tiktaalik to develop tetrapod-like features”, drawing on their own experiences and biological knowledge. Students worked to build and refine their ideas through seven days of instructional activity and including lab exercises. After each activity, students contributed to a summary table that was a public record of ideas (e.g., what did we learn about evolution? what new terms did we learn? how does this relate to the case study?), jointly constructed by all students in the class. The table then served as a reference tool for students as they revised their models. Instruction culminated with student groups generating a final model and explanation. Coding of pre- and post-instruction models using the Learning Progression Rubric developed by Schwartz et al (2009) showed that all student groups improved in their modeling. Most pre-instruction models were literal representations of the phenomenon (e.g., a drawing of Tiktaalik), while post-instruction models included causal explanations and predictions. Although these results are limited (just a single, small course (n=18)), they are promising and underscore the need for additional research on MBI at the postsecondary level.
Bierema, A. M.-K., Schwarz, C. V., & Stoltzfus, J. R. (2017). Engaging undergraduate biology students in scientific modeling: Analysis of group Interactions, sense-making, and justification. CBE—Life Sciences Education, 16(4) ar68. Model-based instruction (MBI) can engage students with models through different pathways, including through model construction or model use. This paper focused on model construction, as a means for students to work collaboratively, practice sense-making and justification, and develop model-based reasoning skills. First, the researchers developed a novel MBI approach that would enable them to investigate how students engage sense-making when modeling. Collecting data from two introductory biology courses, the researchers recorded student groups as they constructed models. Selecting four groups for analysis (n=12 students), the authors coded student dialogue in several ways, including identifying evidence of sense-making and model justification, two science process skills. Students interacted extensively when building models, and all groups engaged in sense-making and justification actions. These actions included clarifying information, agreeing on model components, comparing ideas, and referring to information sources, all of which reflect the process of science. Thus, this paper concludes that by engaging in the process of modeling, students also engage in the process of science.
Manthey, S., & Brewe, E. (2013). Toward university modeling instruction-biology: Adapting curricular frameworks from physics to biology. CBE—Life Sciences Education, 12(2), 206–14. While there is little consensus among practitioners about instructional practices that help students develop modeling skills, there is also little consensus on the core models of introductory biology. The goals of this paper were twofold: (1) to introduce a pedagogy from physics, university modeling instruction (UMI), and describe its adaptation to biology, and (2) describe the core models used in an introductory biology course at a selected institution. UMI, as described in this paper, uses the modeling theory of science to argue that science is a process of modeling, whereby we move through phases of model construction, validation, use, and revision. Thus, instructional practices must engage students in these same practices. Adapting UMI from physics to biology, the authors argue that we must first identify the core models of biology. Using textbook analysis paired with two biology expert interviews, the authors identified a suite of essential models typically taught in an introductory biology course focused on cellular biology and genetics; these models included the cell, evolution, mitosis/meiosis, and the central dogma. Although the results are limited by the case-study approach (i.e., one university and one course), they do underscore a need for our community to identify essential models in biology that align with the core concepts of Vision and Change.
Treagust, D. F., Chittleborough, G., and Mamiala, T. L. (2002). Students’ understanding of the role of scientific models in learning science. International Journal of Science Education, 24(4), 357–68. Modeling as an authentic scientific practice is not often an explicit component of science classrooms. This paper sought to validate an instrument that could capture students’ perceptions of the role of models in science. Surveying 200+ middle school students using the Students’ Understanding of Models in Science (SUMS), the authors found five themes related to student understanding of models: (1) models as multiple representations, (2) models as exact replicas, (3) models as explanatory tools, (4) uses of scientific models, and (5) the changing nature of models. Many students had a fairly robust understanding of what models are and how they can be used in science; however, while students are aware of the various roles models can play, they overwhelmingly ascribe to models as descriptive. The authors conclude that focused instruction is needed to help students discern the differences between abstract and realistic models and call for instructors to use models in multiple ways in their classrooms. While focused on K12 students, these results inform higher education as we reflect on the experiences and ideas students bring to our introductory classrooms. Further, this study prompts college instructors to explicitly discuss the models we use in our teaching and to use them for more than explanatory purposes.
- Deep conceptual understanding of complex biological systems requires developing systems thinking skills. While the field of Biology Education Research (BER) acknowledges the importance of systems thinking skills, there is little research in this area and, as a result, no consensus on what comprises systems thinking or how to assess it in the undergraduate classroom.
- Systems thinking is characterized by an ability to (1) identify and describe a system of interest, and (2) reason about the system and system processes.
- Many studies use model-based instruction as a means to teach some aspects of systems thinking, particularly those skills that involve identifying, describing, and organizing system components and processes.
- Students struggle to build connections within and across biological systems and to reason about system dynamics and emergence. Modeling, in particular, may promote development of these competencies.
- There is a distinct need for additional research that explores how models and MBI can better promote systems thinking. Assessments are also needed that specifically target systems thinking skills.
Hogan, K. (2000). Assessing students’ systems reasoning in ecology. Journal of Biological Education, 35(1), 22-28. Ecology provides a medium to explore the interface between modeling and students’ system thinking abilities. Carbon movement, for example, can be quite complex and requires non-linear, dynamic reasoning approaches. In this paper, Hogan uses food webs to explore the system thinking abilities of sixth graders and the impact of instruction on the development of such skills. Classroom instruction used eco-columns, manipulative and interactive models of complex food webs. Using two distinct modeling tasks (analyze a model, construct a model), Hogan captured and characterized students’ pre- and post-instruction system thinking skills. In the first task, students (n=52) were given a food web and asked to trace the effects of a perturbation (e.g., a loss, increase, or decrease of one population) on the size of other populations in the food web. Through analysis of students’ written responses, Hogan found that pre- and post-instruction, most students used one-way linear reasoning. Most surprisingly, the type of reasoning used depended, to some extent, on the trophic level the perturbation acted on. One-way linear reasoning was most common when the perturbation acted on primary producers while two-way linear reasoning was most common when the perturbation acted on herbivores. Further, cyclic reasoning was observed most frequently when the perturbation acted on top-level carnivores. In a second task, students (n=16) were interviewed following instruction. These students were asked to first create a diagram of a food web and then describe how a pollutant would move through that ecosystem. The majority of students described only the direct impacts of the pollutant, either on a single population or as moving linearly through the ecosystem. Few students (37%) described indirect effects like eutrophication. This research highlights the utility of different assessment tasks in revealing students’ system thinking skills and makes it apparent that current instruction is insufficient in developing students’ system thinking skills. There is a need for instruction that is more explicitly focused on systems and system properties.
Liu, L., & Hmelo-Silver. C. E. (2009). Promoting complex systems learning through the use of conceptual representations in hypermedia. Journal of Research in Science Teaching, 46(9), 1023-40. Liu and colleagues use the structure-function-behavior (SBF) framework (adapted from the AI community, see Goel et al. 1996 and Hmelo-Silver et al. 2007) as a means to frame systems thinking and instruction. As described in this paper, SBF decomposes systems into their structures (the elements of the system), behaviors (mechanisms or how of the system), and functions (the role or output of the system). Because SBF provides clear language to describe systems, it supports instruction that can focus the learner on behaviors and functions over the structures. Further, SBF supports the hierarchical nature of systems thinking (see Assaraf and Orion 2005, 2010; Sommer and Lücken 2010), by recognizing the role of structures in developing an understanding of system function. This paper tests the effectiveness of two distinct hypermedia learning environments, one focused on system structures and the other focused on system function, in promoting systems understanding. Both hypermedia environments included models of the respiratory system and all research was conducted in a psychology research laboratory. Working with two different student populations (pre-service teachers and seventh graders), the research found that students in both groups could describe the macro-level features of the system; however, students in the function-focused treatment were better able to describe micro-level features of the system and to describe system behaviors. This research provides evidence that models can support student development of systems thinking skills.
Reinagel, A., & Bray Speth, E. (2016). Beyond the central Dogma: Model-based learning of how genes determine phenotypes. CBE—Life Sciences Education, 15(1), ar4. The challenges of learning molecular genetics are multiple and to a great extent, reflect the challenges of learning about complex biological systems. For example, in molecular genetics, students must reason across biological levels – from molecule to organisms and entire populations, and across temporal scales – from within an individual to across generations. In addition, molecular genetics involves distinct ontological levels, including information storage and gene expression. In response to these challenges, this paper used a suite of model-building tasks to develop and elicit students’ system thinking skills and their content knowledge of molecular genetics. Working in an introductory biology class for majors, the authors implemented scaffolded, model-based instruction. Initial activities focused on helping students build biologically meaningful relationships between pairs of molecular genetics concepts (e.g., DNA and mRNA or gene and protein) before asking them to integrate their understanding into a general model that explained gene expression. Finally, students contextualized their generalized model to specific case studies. Data, including pair-wise relationships and models, were collected across the semester, including prior to instruction, on midterms, and the final exam. Students’ accuracy in articulating pair-wise relationships improved quickly and persisted through the semester. Interestingly, students did not consistently transfer these pair-wise relationships to their models; in fact, the accuracy of these same relationships as articulated in student models was lower and significantly so in several cases. While students quickly excelled at articulating simple pair-wise relationships, integrating these ideas to explain gene expression added a layer of difficulty. Finally, through analysis of student-generated models, researchers identified two content challenges for students, the origin of variation and phenotypic expression. This research exemplifies how modeling and a model-based pedagogy can help instructors teach system thinking skills while also revealing important information about student understanding.
Sommer, C., & Lücken, M. (2010) System competence – are elementary students able to deal with a biological system. Nordic Studies in Science Education, 6(2), 125-43. Building from systems theory and research on systems thinking in science education, Sommer and Lücken propose a hierarchical framework of systems thinking skills (which they term system competence). At the base of the framework is system organization, which translates to the skill of modeling the system. Specifically, at this level, system thinking is the ability to identify and organize system elements and identify system boundaries. The second level in the framework is system properties, a broad category that includes the ability to identify dynamic relationships, predict the consequences of change, and describe emergence. The hierarchical nature of the framework reflects the findings of others, notably Assaraf and Orion (2005, 2010). Sommer and Lücken then used their framework to identify whether elementary-aged students (n=350) can develop and show evidence of system thinking skills. To assess the competency of system organization, the researchers analyzed student-generated concept models (termed maps) that were collected before, during, and after instruction. Student models were characterized according to (a) the level of interconnectedness of elements within the model, and (b) the over-all structure of the model (linear, cyclic, networked, etc). They found that interconnectedness increased significantly following instruction. Model architecture also changed significantly, moving from mostly linear to networked models. To assess the competency of system properties, the researchers asked a series of follow-up questions that asked students to reflect on given system. The found that students struggled to identify instances of emergence or dynamic relationships, or to predict feedback loops. In general, this research demonstrates that students can acquire some basic system thinking skills and that modeling may be an important tool in the development of system thinking more generally.
Verhoeff, R. P., Waarlo, A. J., & Boersma, K. T. (2008). Systems modelling and the development of coherent understanding of cell biology. International Journal of Science Education, 30(4), 543-68. A major learning challenge of cell biology is developing a conceptual understanding of the cell as a complex biological system. Learners struggle to build connections between cellular processes and phenomena observed at the level of the organism (e.g., cellular respiration and breathing). Verhoeff and colleagues propose that instruction based on systems modeling and systems thinking can support students’ development of a coherent and integrated understanding of cell biology. However, translating system modeling and thinking into practice is not intuitive. Using a design-based research approach which is grounded in theories of learning and research on teaching systems thinking and modeling, the authors developed an instructional strategy to focus the learner on interactions and functions within a system and connections across biological levels. Their instructional strategy begins with basic content acquisition, as a means to motivate student interest and drive the learning process. Subsequently, students engage in a series of modeling tasks, including generating 2- and 3D models and running simulations. Early modeling tasks focus on representing a specific cell and creating concrete models. Later modeling tasks move towards generalizations. Overall, the modeling pedagogy promotes horizontal (within a biological level) and vertical integration (across biological levels) of cell biology concepts. Through classroom testing, the authors note that students have little difficulty building initial models; however, subsequent models that prompt the learner to build connections within and across biological systems required instructor guidance.
Cite this guide: Wilson KJ, Long TM, Momsen JL, Bray Speth E. (2019) Evidence Based Teaching Guide: Modeling in Classroom. CBE Life Science Education. Retrieved from https://lse.ascb.org/evidence-based-teaching-guides/modeling-in-the-classroom/