Cognitive impacts of computer vision-based decision support for international nuclear safeguards-relevant visual analysis tasks .
Abstract not provided.
Engineering topics
Publications and source records attributed to Divis, Kristin.
Abstract not provided.
Abstract not provided.
With machine learning (ML) technologies rapidly expanding to new applications and domains, users are collaborating with artificial intelligence-assisted diagnostic tools to a larger and larger extent. But what impact does ML aid have on cognitive performance, especially when the ML output is not always accurate? Here, we examined the cognitive effects of the presence of simulated ML assistance—including both accurate and inaccurate output—on two tasks (a domain-specific nuclear safeguards task and domain-general visual search task). Patterns of performance varied across the two tasks for both the presence of ML aid as well as the category of ML feedback (e.g., false alarm). These results indicate that differences such as domain could influence users’ performance with ML aid, and suggest the need to test the effects of ML output (and associated errors) in the specific context of use, especially when the stimuli of interest are vague or ill-defined
Due to their recent increases in performance, machine learning and deep learning models are being increasingly adopted across many domains for visual processing tasks. One such domain is international nuclear safeguards, which seeks to verify the peaceful use of commercial nuclear energy across the globe. Despite recent impressive performance results from machine learning and deep learning algorithms, there is always at least some small level of error. Given the significant consequences of international nuclear safeguards conclusions, we sought to characterize how incorrect responses from a machine or deep learning-assisted visual search task would cognitively impact users. We found that not only do some types of model errors have larger negative impacts on human performance than other errors, the scale of those impacts change depending on the accuracy of the model with which they are presented and they persist in scenarios of evenly distributed errors and single-error presentations. Further, we found that experiments conducted using a common visual search dataset from the psychology community has similar implications to a safeguards- relevant dataset of images containing hyperboloid cooling towers when the cooling tower images are presented to expert participants. While novice performance was considerably different (and worse) on the cooling tower task, we saw increased novice reliance on the most challenging cooling tower images compared to experts. These findings are relevant not just to the cognitive science community, but also for developers of machine and deep learning that will be implemented in multiple domains. For safeguards, this research provides key insights into how machine and deep learning projects should be implemented considering their special requirements that information not be missed.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Abstract not provided.
Recently, an approach for determining the value of a visualization was proposed, one moving beyond simple measurements of task accuracy and speed. The value equation contains components for the time savings a visualization provides, the insights and insightful questions it spurs, the overall essence of the data it conveys, and the confidence about the data and its domain it inspires. This articulation of value is purely descriptive, however, providing no actionable method of assessing a visualization's value. In this work, we create a heuristic-based evaluation methodology to accompany the value equation for assessing interactive visualizations. We refer to the methodology colloquially as ICE-T, based on an anagram of the four value components. Our approach breaks the four components down into guidelines, each of which is made up of a small set of low-level heuristics. Evaluators who have knowledge of visualization design principles then assess the visualization with respect to the heuristics. We conducted an initial trial of the methodology on three interactive visualizations of the same data set, each evaluated by 15 visualization experts. As a result, we found that the methodology showed promise, obtaining consistent ratings across the three visualizations and mirroring judgments of the utility of the visualizations by instructors of the course in which they were developed.
Exposure to extreme environments is both mentally and physically taxing, leading to suboptimal performance and even life-threatening emergencies. Physiological and cognitive monitoring could provide the earliest indicator of performance decline and inform appropriate therapeutic intervention, yet little research has explored the relationship between these markers in strenuous settings. The Rim-to-Rim Wearables at the Canyon for Health (R2RWATCH) study is a research project at Sandia National Laboratories funded by the Defense Threat Reduction Agency to identify which physiological and cognitive phenomena collected by non-invasive wearable devices are the most related to performance in extreme environments. In a pilot study, data were collected from civilians and military warfighters hiking the Rim-to-Rim trail at the Grand Canyon. Each participant wore a set of devices collecting physiological, cognitive, and environmental data such as heart rate, memory, ambient temperature, etc. Promising preliminary results found correlates between physiological markers recorded by the wearable devices and decline in cognitive abilities, although further work is required to refine those measurements. Planned follow-up studies will validate these findings and further explore outstanding questions.
Evaluating the effectiveness of data visualizations is a challenging undertaking and often relies on one-off human subject studies that test a visualization in the context of one specific task. Researchers across the fields of data science, visualization, and human-computer interaction are calling for foundational tools and principles that could be applied to assessing the effectiveness of data visualizations in a more rapid and generalizable manner. The visual saliency model for data visualizations described here addresses this need for evaluation tools. Visual saliency models assess the visual features (e.g. color, contrast, edges) of an image to predict which areas of that image will draw a viewer’s attention. Providing predictions of where viewers will look in a visualization based on the saliency of the various regions and elements in that image could help designers to assess whether or not their design will draw the viewer’s attention as intended. The Data Visualization Saliency model, a saliency model for data visualizations, consists of two parts: a text saliency map and a modified version of the Itti, Koch and Niebur model [1]. The final saliency map is a linear combination of the two. The text saliency map is created from an algorithm that computes the likelihood of belonging to a text region for each pixel of an input visualization image. The modified Itti model transforms the original input image to a representation in CIE LAB color space, which provides an accurate representation of perceptual color opponency. For a given input data visualization image, the Data Visualization Saliency model computes and returns the visualization saliency map highlighting areas of the data visualization that will draw a viewer’s attention. It also returns the lower level saliency maps for text and the modified Itti model.