Seeing Is Believing: Using Eye-Tracking Devices in Environmental Research.

OBJECTIVES: This article aims to provide methodological guidance for research that uses eye-tracking devices (ETDs) to study environment and behavior relationships. BACKGROUND: Vision is an important human sense through which people acquire a large amount of environmental information. ETDs are tools for detecting eye/gaze behaviors, facilitating better understanding about how people collect visual information and how such information is related to emotions and psychological states. However, there is a lack of guidance for the application of ETDs to environment and behavior studies. METHODS: A literature review was conducted on articles reporting empirical studies that used ETDs. The data were extracted and compiled, including information such as research questions, research design, types of ETDs, variables measured, types of physical environment (or visual stimuli), stimuli durations, data analysis methods, and so on. RESULTS: Fifty articles were identified. The main research topics were related to urban and landscape environments, and architecture and interior spaces. Most of the research designs were experimental or quasi-experimental designs, with a few cross-sectional studies. The majority types of ETDs were screen-based ETDs, followed by mobile ETDs (glasses). Main variables were gaze fixations, fixation durations, and scan paths. Typical types of stimuli included images, videos, virtual reality, and real environments and/or objects. CONCLUSIONS: Guidance for eye-tracking research on environment and behavior was developed based on the literature review results, to provide direction for determining research questions, selecting appropriate research designs, establishing participant inclusion and/or excluding criteria, collecting and analyzing data, and interpreting research results.

Factors affecting the COVID-19 risk in the US counties: an innovative approach by combining unsupervised and supervised learning.

The COVID-19 disease spreads swiftly, and nearly three months after the first positive case was confirmed in China, Coronavirus started to spread all over the United States. Some states and counties reported high number of positive cases and deaths, while some reported lower COVID-19 related cases and death. In this paper, the factors that could affect the risk of COVID-19 infection and death were analyzed in county level. An innovative method by using K-means clustering and several classification models is utilized to determine the most critical factors. Results showed that longitudinal coordinate and population density, latitudinal coordinate, percentage of non-white people, percentage of uninsured people, percent of people below poverty, percentage of Elderly people, number of ICU beds per 10,000 people, percentage of smokers were the most significant attributes.