Breast screening, chicken sexing and the search for oil: challenges for visual cognition

Interpretation of images of the Earth's subsurface is a process whereby humans perceive and categorize visual features derived from seismic data. The seismic data are presented in the form of vertical slices showing points of change in some variable being measured (e.g. acoustic impedance) and horizontal slices showing surfaces interpolated between values at a particular time or horizon across multiple vertical slices. These images are usually highly complex and their nature has been determined largely by the technical capabilities of the hardware and software of the imaging technology. Because of these constraints, we argue, images do not convey information as readily as they could. We believe that these images could be more informative if they were constructed and tailored with known properties of the human visual system. Furthermore, little or no consideration has been given to the training and selection for image interpretation vis-A-vis the fundamental psychological skills that distinguish good from poor interpreters. In this paper we argue that tailoring images to the human visual system and developing working practices that eliminate biases will improve the detection of subtle features related to hydrocarbon traps. Furthermore, establishing training procedures that enhance the visual system's ability to detect and encode hydrocarbon traps, and creating selection procedures that select individuals with excellent visual imagery skills will also facilitate performance.