Some may argue that the data stories are created and structured in peoples’ minds and not in the visualization design itself, as the latter only supports them in constructing this mental model. A similar point has been raised by Kosara (2010) where he suggests that ‘stories don’t tell themselves’, and that visualization only sets a background or context, but that stories are told by people, including the users of the data.
Chen (2006) explains that different users may create different mental mod- els even when using the same system. Hullman et al. (2011) also explain that small changes in representing information visualization may provide different results or insights. In the same way, it can be argued that different storytelling strategies with information visualization may result in differences in the way people perceive, understand, and construct narratives from them. There are various factors that can be examined, and one of them is the story delivery model, which will be addressed in the experiments reported in Chapter 3.
Various modes of communication exist, regardless of what is being com- municated. These include scientific data in educational settings or stories in novels, films, or children’s books. Each of these communication modes has its own nature and characteristics. Each may also have a different impact on how people (whether they are readers, listeners, users, etc.) comprehend what is being communicated. Before detailing the procedures and findings of the conducted experiments/user study, it is beneficial to review some modes of communication and story delivery techniques from other fields to enrich the scope of the study in hand and explain the relevance of these techniques and how each is expected to influence stories and comperehension on the part of
the recipients.
Speech and writing are two communication modes that can be used in ed- ucation, in telling stories and in communicating many other forms of content (Schallert et al. 1977). Although some well-crafted information visualization studies have assessed differences in various techniques or features in user tasks such as the use of animation and animated transitions (for example, Heer & Robertson (2007)), no studies have been found that examine differences between communication modes. For example, there is little or no study on the relationship between writing and speech or between direct narrative by a speaker and interactive exploration of the data. The differences between written and spoken communication can be thought of in relation to different communication modes of data stories through information visualization. One of the main differences between the nature of speech and that of writing is that speech has some prosodic features that are missing in writing, such as rhythm and stress. The lack of this information in writing requires the reader to have enough skills to interpret the writer’s message (Schallert et al. 1977). It can be argued that, in the same way, watching a speaker who is using information visualization to tell a story (closer to speech mode) differs from interactive exploration of the data (closer to writing mode). Although both use infor- mation visualization to help deliver a data story, each of them has their own characteristics, advantages, disadvantages and effect on comprehension.
In films and documentaries, a technique called ‘direct narration’ or ‘voice- over narration’ is sometimes used to tell stories (Pramaggiore & Wallis 2011). Voice-over narration means that there is a narrator, who might be a character in the film, who narrates and directs the viewers through the scenes in the film. The role the narrator has in the story they describe may be made ex- plicit at the outset or revealed as part of the storytelling process itself. The
extent to which this technique is used varies from extensively narrated events to restricted narration, where the events in the story just unfold without di- rect narration. An interesting critique of the voice-over narration technique is that it replicates what the visual narrative already shows (Henderson 1983). Hence, it is suggested that this technique should be used to convey something different, and in some cases to highlight or emphasise some information (Hen- derson 1983). A similar delivery model to the voice-over narration technique in information visualization is employed when a speaker uses information visu- alization to tell a story. An example of this model of delivery is Rosling’s talks on various issues in global health and economics using an animated bubble chart (Rosling 2009, 2008). It is difficult to avoid narrating what is shown in the visualization in such talks, but it should provide something different and add depth to the data stories and insights already shown in the visualization. Another consideration relevant to this story delivery technique is the effect of the personal and subjective interpretation and/or recommendations provided by the narrator on the viewers’ engagement in the story, and on their accep- tance of its message (Pramaggiore & Wallis 2011).
The lack of empirical evidence as to what effect an authored narration has on the interpretation of data-rich visualization was the motivation for the work carried out as part of this research. In the experiments reported in Chap- ter 3, the effect of the story delivery and narration model on the narratives constructed by audiences will be explored.