Using Machine Learning to Generate Novel Hypotheses: Increasing Optimism
about Covid-19 Makes People Less Willing to Justify Unethical Behaviors
Abstract
How can we nudge people to not engage in unethical behaviors, such as
hoarding and violating social distancing, during Covid-19? As past
research on antecedents of unethical behavior did not provide a clear
answer, we turned to machine learning to generate novel hypotheses. We
trained a deep learning model to predict whether or not World Values
Survey respondents perceived unethical behaviors as justifiable, based
on their responses to 708 other questions. The model identified optimism
about the future of humanity as one of the top predictors of
unethicality. A pre-registered correlational study (N=218 US-residents)
conceptually replicated this finding. A preregistered experiment (N=294
US-residents) provided causal support: participants who read a scenario
conveying optimism about the Covid-19 pandemic were less willing to
justify hoarding and violating social distancing guidelines. The
findings suggest that optimism can help reduce unethicality, and
document the utility of machine learning methods for generating novel
hypotheses.
Accepted for COVID-19 fast-track publication in Psychological
Science.