How Do NASA Engineers Perceive MMOD Risks? A Fuzzy Trace Approach

Abstract

In this study, we aim to address two overarching questions: 1) How do expert engineers perceive risks associated with their design choices; and 2) how do these decision-making processes differ from those of laypeople. To address these research questions, we draw upon Fuzzy Trace Theory (FTT), a leading account of decision under risk. According to FTT, individuals form several parallel yet independent mental representations of risk information, known as verbatim and gist. While the former encodes the precise details of a stimulus, the latter encodes the basic meaning of the stimulus. As expertise increases, individuals rely more on gist, rather than verbatim, representations. Here, we test FTT’s predictions on a sample of engineering experts. Specifically, we measure how NASA experts perceive the risks of Micro Meteoroid and Orbital Debris (MMOD) impacts within the context of a hypothetical mission design scenario. Our sample consists of two groups: 1) 42 NASA employees recruited via email and 2) 234 laypeople recruited through an online crowdsourcing platform. Our survey instruments measure subjects’ assessments of the risk of mission failure due to MMOD strike, the benefits of adding additional shielding to mitigate that risk, and endorsement of general gist principles. Consistent with FTT’s predictions, results of an exploratory factor analysis suggest that three gist representations capture the most variance. The first factor captures schedule risks with NASA respondents more likely to agree that these risks were non-negligible. The second factor reflects perceptions of cost/mass, with NASA respondents more likely to agree that costs of MMOD shielding were non-negligible. The third factor captures perceptions of the risk of mission failure due to MMOD strikes with NASA respondents more likely to indicate that these risks were essentially nil. These results support FTT’s predictions regarding the categorical nature of risk perception by expert engineers. A fourth factor reflects respondents’ verbatim assessments of the risks of MMOD strikes. NASA respondents’ risk estimates were lower (22% risk of penetration and 17% loss of mission) than those of laypeople (29% probability of penetration and 29% probability of loss of mission) with both groups’ average risk estimates were broadly consistent with historical data for International Space Station missions (29% over the 2007-2016-time period). This indicates that quantitative risk estimates likely did not drive subjects’ responses, consistent with FTT’s predictions; however, NASA respondents’ estimates are more consistent with mission requirements (probability of penetration <24%) than with actual historical data. Implications for systems design are discussed.