Learning Mental Models Project
Complex systems are an important part of the world that we live in and, as such, are recognized as a key idea in national science standards. Developing an accurate mental model of complex systems is an important part of learning in science and engineering. Mental models, with their explanatory and predictive power, are critical for explanation, analysis, prediction, monitoring, diagnosis, and design of complex systems, and subsequent acquisition of a more sophisticated understanding of them. The goal of this proposed project is to design and assess the effectiveness of a modeling methodology and technology that will foster learning about (i) the nature of complex systems, (ii) the concepts in modeling a complex system (e.g., causality, process, outcomes, aggregation, abstraction), and (iii) the role of models in scientific inquiry and experimentation.
In this project, we have selected fish-aquarium systems as the context in which middle-school students will learn about the nature of complex systems and the proposed modeling framework is called SBF, which grew out of our lab.
Learning about complex systems through aquaria
The complex, yet learnable, nature of aquaria appeals to middle school teachers seeking to incorporate hands-on investigation of broad concepts, constructed knowledge, and interrelated themes to promote scientific problem-solving skills. Aquaria, because of their size and relative low cost, have often been used for classroom study. Middle-school teachers as well as the students are generally familiar with aquaria and their associated organisms. This familiarity, often paired with the need to know, provides a motivating context for students to learn.
The ACT learning environment
We propose to design, develop and evaluate an interactive learning environment called ACT (Aquarium Construction Toolkit). ACT will support middle-school students learn various aspects related to establishing and maintaining classroom aquaria. The ACT environment will have the following four main components built into it.
Modeling will be an important aspect of the envisioned learning process. We are proposing to use the SBF modeling framework for this purpose. Artificial Intelligence (AI) theories of model-based analogies in conceptual engineering design have led to a methodology and a language for building functional and causal models of physical systems called Structure-Behavior-Function (SBF) models. An SBF model of a system explicitly represents its structure [S] (i.e., its configuration of components and connections), its functions [F] (i.e., its output behaviors), and its behaviors [B] (i.e. its internal causal processes that compose the functions of the components into the functions of the system). The SBF language provides a vocabulary for expressing and organizing knowledge about complex systems, which captures functionality and causality at multiple levels of abstraction and aggregation. With the help of this framework we are planning to build an interactive learning environment that helps students learn deeply about complex systems.
|ACT will include an SBF modeling tool called SBF Author. SBF Author provides a visual annotation for the SBF language and enable easy, visual construction and manipulation of SBF models of aquaria. It can make important behavioral and functional aspects of a system salient and open for learners’ consideration and discussion. Originating in AI, the SBF language was only available in machine readable form. SBF Author tries to bridge the gap between AI theory and the needs of the Learning Sciences.|
Aquaria simulation tool
| ||SBF Model ||Simulation |
|Unlike computational models, SBF models are static (not "runnable"), conceptual models that provide more explanatory power than predictive power. Predictive power is very useful in the context of learning for: (1) validation of constructed models and (2) hypothesis generation and experimental verification. Therefore, it is necessary to for SBF models to have the "runnable" feature so that the students can run the SBF models that they create. We are developing a theory and implementation of "computational model generation" from SBF models based on the notion of "behavior patterns." The target simulation platform on which these compitational models will compile and run is NetLogo. This will allow the learners to run simulations of their SBF models on NetLogo platform.|| || |
The Inquiry notepad helps students organize their thoughts and ideas, record hypotheses, design experiments, collect data from observations and link models and simulations in a coherent manner.
Interface to the physical aquarium
Finally, ACT will provide a "one-stop" interface to various sensors attached to the physical aquarium. Students can get current and historical readings, which can then be treated as data and graphed, processed etc.
This research is supported by a grant from the US - National Science Foundation.
Last modified 23 March 2007 at 2:43 am by svattam