A sketch of a theory and modelling notation for elucidating the structure of representations Diagrams 2020 Representational Interpretive Structure: Theory and Notation Diagrams 2022

Representational Interpretive Structure Theory and Notation (RIST and RISN, respectively) provide a means to model the interpretation of a representation by a particular reader. It does this by allowing you to compose four schemas with different kinds of links.

This website provides two brief examples of RISN being used to model interpretations of Playfair’s Line Graph, and the PowerPoint toolbar. We also host a tool for you to edit RISN models, found using the button at the bottom of this page.

Playfair’s Line Graph

Playfair’s line graph is the earliest example of a line graph. It details the trade balance, in pounds (£), between the United Kingdom and Denmark & Norway between 1700 and 1780.

In all the following examples, the graph is being interpreted in the context of general understanding. There is no specific information that the analysts were trying to extract.

Click to enlarge. Download model.

To the right we have Analyst 1’s RISN model for their interpretation of Playfair’s line graph. We would characterise this model as “broad”, notably making use of the implicit coordinate system idiom. The model also uses many anchoring connections, suggesting that much of the content of their interpretation is implicit in the graph.

Click to enlarge. Download model.

In contrast, Analyst 2’s model (left) is “deep” – their interpretation tends to “drill down” into details, rather than have them all available at the top level. Analyst 2 also only uses one anchoring connection, but there are many schemas below it, suggesting there was one key relationship that unlocked their interpretation.

Analyst 2 placed the trade balance at the top of their model, emphasising its importance to their interpretation. Almost all of the rest of their interpretation flowed from this.

Powerpoint Toolbar

A different kind of representation is the PowerPoint toolbar – rarely thought of as a representation, but it is! The toolbar above conveys two kinds of information: actions available, and the current state. For example, it shows a button for making text bold, and the button is not active so we know the current text is not bold (or the selection is not text). For this example, the analysts again had no direct goal, but were directed to focus on the text-editing tools.

Click to enlarge. Download model.

As before, Analyst 1 has created a “broad” model (left). Upon recognising the coordinating R-Scheme, they immediately access all the sub-schema and dimensions available to them. This fan-out occurs again when they focus on the text-editing tools, able immediately to access most of the features without further coordination. Analyst 1 identifies a complex sub-representation within the font picker, as it is not just a simple list but a collection of many names rendered in their respective fonts. Similarly, colours are not just a simple dropdown, but a complex colour-picking widget that the Analyst has indicated is its own representation.

Click to enlarge. Download model.

Analyst 2 has, again, gone for a “deep” model (right), using layers of coordinating R-Schemes to drill deeper and access the features of the toolbar. Unlike Analyst 1, Analyst 2 modelled all the different sections of the toolbar shown, not just the text-editing tools. But, focusing on the text-editing tools, we see a higher-level breakdown: attributes versus layout. At the next layer, the breakdown is similar, although the font and colour pickers are not considered to be new sub-representations.