As part of the MIT Media Lab's Project Headlight (1985-1990), led by Seymour Papert at the Hennigan School in Boston, Ricki Goldman Segall constructed multimedia portraits of three children, with three different preferred thinking styles: empirical/causal, narrative, and relational.

The first child, Josh, had a causal and empirical approach to making sense of the world; his approach matched the traditional way of learning about scientific issues. He broke down wholes into parts and searched for the causal link between events. The computer programs that he built in Logo were organized in complex procedures which segmented various functions into "logical" sequences.

He once told me that it's not possible to invent something unless you first see it in the real world. To build something new, he told me, you have to put together things that already exist in the world in some new way.

The two more narrative and relational thinking children which I focused on in the final dissertation seemed to be at a disadvantage in programming in Logo. These children would start building out from one sub-procedure and then link that part to other separate parts. Sometimes, they couldn't find the thread that brought the whole into a working structure; however, sometimes they invented new structures.

With multimedia tools, I hypothesized, children who think in relational and story-telling webs could have more opportunities for thinking about science in a style that matched their thinking. This led me to initiate the Gender, Science and Multimedia study at the Bayside a few years later.