When learning about novel concepts, we often use metaphors to frame unfamiliar concepts in terms of ones with which we are more familiar. Doing so often involves the creative generation of novel metaphors, but it also requires critical thinking to avoid potential misconceptions. As part of my dissertation work, I conducted a study using computational metaphor identification to improve students’ awareness of their own use of metaphors, thus fostering both critical thinking about, and creative generation of, metaphors.

Working with teachers from local middle schools, we developed a computer based module (using WISE) around cellular biology, specifically incorporating the metaphor A CELL IS A CITY. The module included several written questions asking them to describe in their own words functions of the cell and various organelles. I then used CMI to analyze students’ writing for potential metaphors. The identified metaphors were then integrated back into the instructional module to determine how they might impact various aspects of learning. The results indicated that the computationally identified metaphors were effective at fostering both critical thinking and creativity (see publications below for more details).

On the one hand, this work carries important implications on the use of metaphor in instruction and learning, both in science and in other domains. On the other hand, this work suggests an alternative role for the use of computational analysis in teaching. Rather than facilitating automated writing assessment, I suggest that it may be beneficial to incorporate the results of such computational analyses back into the learning process, specifically with a focus on supporting critical thinking, creativity, and other modes of thought that enable people to see familiar situations from novel, different perspectives.