Visual elements & visual paradigms

This post provides a summary of my research to date, working toward my PhD thesis, Visual Elements and Visual Paradigms: an investigation into scientific conceptual figures. Key references are listed at the bottom of this post.

The research project is investigating visual communication within scientific peer groups, specifically the conceptual figures that scientists use to explain processes, ideas and theories. These figures are published in papers in scientific journals and used in printed posters and PowerPoint presentations at scientific conferences. They are constructed from visual elements that utilise a visual language specific to that field of science. This language requires a great deal of background knowledge, on the part of both the originator of the figure and their audience, to be used and understood. In the case of the earth and life sciences, these visual languages have developed over several centuries and are often rooted in a strong tradition of observational drawing.

Philosopher of science Thomas Kuhn stated that scientific knowledge advances in a series of paradigms (Kuhn, 1996), and the longer a scientific paradigm persists, the more specialised its scientific language, both visual and verbal, becomes. It can be argued that each paradigm also has an associated image that represents that particular scientific concept. Kuhn uses Copernicus’ theory of a heliocentric solar system as a prime example of a scientific paradigm. Copernicus also produced a diagram that perfectly encapsulated his theory, and this has been used ever since, in a wide variety of forms, as the visual representation of the heliocentric solar system; a visual paradigm.

In his 1935 book, The Genesis and Development of a Scientific Fact (Fleck, 1979), Ludwik Fleck describes how a scientific theory emanates from a small ‘inner core’ of scientists, before disseminating through the broader scientific community, on to students and school children and, finally, to the general public. It can be argued that the visual representation of a theory, the visual paradigm, is created and disseminated in the same way. It is this ‘inner core’ of scientists that is the concern of this research project; those conceptual figures produced by scientists themselves, using highly specialised visual languages, to communicate with each other.

The aims of the research are three-fold:

  • To analyse the historical development of visual languages in the life and earth sciences, highlighting visual paradigms.
  • To evaluate the appropriateness of existing methods of visual analysis when applied to conceptual scientific figures. To develop a range of tools for visual analysis applied to these figures, some of which will be practical design tools for analysis and communication.
  • To construct a group of case studies based on collaboration with working scientists, while learning their visual language and how it is employed in a professional setting.

To achieve these aims, there are three strands to the research: historical development; visual analysis; professional application.

Historical Development

To understand the significance of the visual elements within a scientific conceptual figure, and to identify those elements that can be considered visual paradigms, it is necessary to trace their origin. Daston and Galison, in their book Objectivity (Daston, 2007), describe the historical development of the visualisation of science in three stages that follow on from, but do not necessarily replace, each other. The first stage is that of direct observation, as typified by anatomical and botanical drawings. The advent of photography raised the prospect of a more objective observer than the human eye, and marked the second stage of development, that of mechanical observation. However, scientists quickly realised the limitations of a medium that captured absolutely everything. Better to create a visual representation that highlighted what was important and, crucial for the acceptance of a scientific idea, made a convincing argument. This is the third stage in Daston and Galison’s development, where trained judgment is required to both produce and understand these visual representations.

Chronologically-ordered collections of figures that depict the same scientific concepts, such as the blood circulation system, demonstrate the birth of conventions for visual representation, and also confirm their longevity and persistence. Visual motifs can be seen in 21st century conceptual figures that first appeared in the 18th or 19th centuries. To an informed viewer, these familiar representations, particularly those that can be regarded as visual paradigms, instantly establish context and meaning, without the need for additional explanation.

This blog post provides a link to an journal article that describes the development of scientific visual languages in more detail.

Visual Analysis

Given the enormous amount of information that is contained within a scientific conceptual figure, and which is only apparent to an informed viewer, how would one perform a visual analysis of such a figure? The second strand of the research investigates this issue.

A conceptual figure can be analysed as a piece of visual communication, and described in terms of elements such as line, colour and text. Similarly, content analysis can be performed, but neither of these approaches would address either the science or the external influences that were applied during the figure’s creation. The art historian James Elkins has written extensively on what he terms ‘nonart’ images (Elkins 1999), stating that artists cannot analyse such images due to their lack of scientific knowledge. However, he also asserts that, equally, they cannot be analysed by scientists, who do not possess the necessary expertise.

Semiotician Francois Bastide acknowledges the limitation of semiotic analysis without scientific knowledge, when applied to conceptual scientific figures (Lynch, 1990). Linguist Elizabeth Rowley-Jolivet has written extensively on the role of images in scientific presentations (Rowley-Jolivet, 2000), particularly the importance of a knowledgeable viewer, but she does not attempt to analyse the conceptual figures that form a key part of those presentations.  Regarding the figures as multi-modal ensembles, as defined by Frank Serafini (2013) is the most appropriate way to acknowledge their complexity but, again, that does not directly address their scientific content.

Given that none of the visual analysis methods discussed above have been designed to specifically address the scientific meaning of conceptual figures, the research puts forward a method of analysis that will accomplish this. The use of graphic design practice as a research method, in both the analysis phase and in communicating the results of the analysis, is a key aspect of the research. This practice is described as a set of graphic tools that can be used as a methodological approach to research. The analysis also highlights those external factors that impact upon the production of the figures and the issues faced by the scientist who creates them. If a graphic designer did understand the scientific visual language, could they usefully get involved in the creation of figures such as these, and help mitigate some of these issues?

Professional Application

This question brings us to the third strand of the research project, where the research is applied in a professional setting. The Centre for Neuroimaging Sciences at King’s College London proposed a collaboration, as they were looking for an easier way of producing conceptual figures, with a consistent visual style that could be associated with the CNS. The conceptual figures for three upcoming journal papers were put forward as case studies. By conducting semi-structured interviews with the scientists involved, and by analysing existing conceptual figures from the appropriate fields, it is possible to build up a good understanding of the visual language being employed and deduce which visual elements, and visual paradigms, are required for the figures.

By concentrating on the individual visual elements, rather than on a completed figure, it is possible to create a digital library of such elements that can then be used to construct conceptual figures. Each visual element is created in Adobe Illustrator software and, by making use of the ‘Layer’ functionality within the software, the elements are made to be easily edited and adapted for use in different figures or in different media. The success of this approach is dependent on the scientists themselves, given appropriate instruction, being able to use and edit the visual elements and combine them into a complete conceptual figure.

The collaboration is documented in detail on the Neuroscience & Graphic Design website, which also contains the image gallery of visual elements.


Key references
Daston, L. & Galison, P. (2007) Objectivity. New York: Zone Books.
Elkins, J. (1999) The Domain of Images. Ithaca: Cornell University Press.
Fleck, L. (1979) Genesis and Development of a Scientific Fact. Translated from the German by Fred Bradley & Thaddeus J. Trenn. Chicago: University of Chicago Press.
Kuhn, T.S. (1996) The Structure of Scientific Revolutions. 3rd ed. Chicago: The University of Chicago Press.
Lynch, M. & Woolgar, S. eds. (1990) Representations in Scientific Practice. Massachusetts: The MIT Press.
Rowley-Jolivet, E. (2000) Image as Text. Aspects of the shared visual language of scientific conference participants. Groupe d’Étude et de Recherche en Anglais de Spécialité. vol. 27-30, pp 133-154. [Internet]. Available from: <; [Accessed 24 May 2014]
Serafini, F. (2013) Reading the Visual: An Introduction to Teaching Multimodal Literacy. New York: Teachers College Press.