Use this URL to cite or link to this record in EThOS:
Title: Thinking with data visualisations : cognitive processing and spatial inferences when communicating climate change
Author: Harold, Jordan
ISNI:       0000 0004 7426 9280
Awarding Body: University of East Anglia
Current Institution: University of East Anglia
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
Data visualisations can be effective for communicating scientific data, but only if they are understood. Such visualisations (i.e. scientific figures) are used within assessment reports produced by the Intergovernmental Panel on Climate Change (IPCC). However, IPCC figures have been criticised for being inaccessible to non-experts. This thesis presents a thematic analysis of interviews with IPCC authors, finding that a requirement to uphold scientific accuracy results in complex figures that are difficult for non-experts to comprehend, and which therefore require expert explanation. Evidence is subsequently presented showing that figures with greater visual complexity are associated with greater perceived comprehension difficulty among non-experts. Comprehension of complex data visualisations may require readers to make spatial inferences. When interpreting a time-series graph of climate data, it was found that non-experts did not always readily identify the long-term trend. Two experiments then show that linguistic information in the form of warnings can support spatial representations for trends in memory by directing visual attention during encoding (measured using eyetracking). This thesis also considers spatial inferences when forming expectations about future data, finding that expectations were sensitive to patterns in past data. Further, features that act on bottom-up perceptual processes were largely ineffective in supporting spatial inferences. Conversely, replacing spatial inferences by explicitly representing information moderated future expectations. However, replacing spatial inferences might not always be desirable in real-world contexts. The evidence indicates that when information is not explicitly represented in a data visualisation, providing top-down knowledge may be more effective in supporting spatial inferences than providing visual cues acting on bottom-up perceptual processes. This thesis further provides evidence-based guidelines drawn from the cognitive and psychological sciences to support climate change researchers in enhancing the ease of comprehension of their data visualisations, and so enable future IPCC outputs to be more accessible.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available