Spatial Thinking in 3D and 4D Geology Visualization

Thomas F. Shipley¹, Cathryn Manduca², Carol J. Ormand²³, and Basil Tikoff³
¹Department of Psychology, Temple University, Philadelphia, PA
²Science Education Resource Center, Carleton College, Northfield, MN
³Department of Geoscience, University of Wisconsin, Madison, WI

Abstract

Research on spatial thinking in the geosciences in the Spatial Intelligence and Learning Center has two basic aims. The first is to better understand how humans reason about complex spatial problems, and the second is to use what we know about how humans think about spatial problem to improve science education. In the context of this conference, we consider two long-term goals: 1) Understanding how spatial skills develop; and 2) Understanding how best to support spatial thinking. A key idea in considering long-term research investment is that spatial skills can be improved. We are beginning to understand spatial reasoning limitations, individual differences in spatial reasoning skills, how to measure differences in skills, and how skills can be improved early in geoscience education. There is every reason to believe that skills can continue to improve over a lifetime, so an important research agenda will be understanding commonalities and differences in learning as spatial skills develop in a disciplinary specialist. We also are beginning to understand how to support spatial reasoning using tools (such as gestures, sketches, and physical models) that reduce the cognitive load and free cognitive resources to work on complex problems. We have begun to understand how these tools work for novices; it is critical to understand how these tools can best be applied as expertise develops. It may be possible to further hone the geologist's most important tool, their mind, through continuing education in spatial thinking. In this talk I will focus on two broad classes of spatial thinking problems to consider what we know and key questions for future research to bridge the gap between novice and expert learning: 1) Inferring 3D spatial relations from sparse (or poorly constrained) data; and 2) Inferring events from spatial consequences.

Cognitive science research provides sophisticated models of how the visual system fills in missing data due to sensory limits or optical consequences of opaque objects. For example, we can recognize objects even when portions of the objects are partially visually blocked, or occluded, by nearer objects. Seeing a whole object despite occlusion occurs as a consequence of an active filling in process, and does not depend solely on recognizing familiar parts of familiar objects. Our work on understanding when geologists infer structures from sparse data has led to the development of a new research program that extends cognitive science object-completion research to include cases where objects are not visible because they are underground. Partial information, which may include proportions of an object that are visible at the surface, or from seismic reflections, can be used to infer complex 3D structures. We have worked to develop strategies to help students successfully infer 3D structures in these circumstances by freeing cognitive resources and off-loading some of the information needed for successful visualization, for example, by using hand gestures to help the student visualize the 3D properties of an object displayed on a map. Bridging the gap between supporting novice geological reasoning about 3D objects and analogous problems faced by working geologists (such as visualizing 3D spatial relations between permeable and impermeable layers in a fault system) requires a program of research that takes on a few key questions: 1) What errors do experts make and why do they make the errors, 2) Why are there individual differences in solving these problems, 3) Can these skills be trained, and 4) How can these skills be best supported?

Identifying the skills that form the foundation of spatial thinking is one of the more important cognitive science outcomes from the interdisciplinary research program. Although there is general agreement that spatial thinking is not one thing, there is little agreement about what the component skills might be. By working with a discipline that carefully describes spatial properties of the world (Geology), we are beginning to identify some distinct components of spatial thinking. Cognitive science research has historically focused on a limited class of mental simulation of events (visualizing objects from different perspectives, also known as mental rotation). Geological deformations include a wide range of material response (e.g., brittle versus ductile deformations), which may not be characterized solely by rigid-body rotations. We have found that geologists are particularly good, relative to scientists in other academic disciplines such as chemistry, at visualizing brittle (discrete) transformations suggesting that mental brittle deformation is a spatial skill that is distinct from mental rotation. Further, testing of novices suggests that two important dimensions of spatial thinking are reasoning about: 1) Rigid versus non-rigid changes; and 2) Changes that involve a single object versus changes that involve changing spatial relations among objects. Important questions for future research include: 1) How do experts visualize interactions between 3D forms and material or fluids moving through these forms, and 2) How do experts align mental models and observations to constrain reasoning about events?

AAPG Search and Discovery Article #120140© 2014 AAPG Hedberg Conference 3D Structural Geologic Interpretation: Earth, Mind and Machine, June 23-27, 2013, Reno, Nevada