People who are more likely to die care less about the future: Life insurance risk ratings predict personality

Adaptationist models predict that individuals at higher risk of death will be calibrated to prioritize immediate over future benefits. However, operationalizing individual mortality risk in empirical studies has proven challenging. We introduce and explore a novel method of operationalizing individual mortality risk: Using the risk ratings assigned by actuaries to purchasers of individual life insurance policies. Participants, who had recently gone through underwriting as part of the insurance application process, completed self-report instruments to assess personality traits related to present-future tradeoffs and a putative fast-slow continuum of life history strategy. Study 1 (n = 270) found that insurance-based mortality risk associated negatively with a measure of slow life strategy and positively with a measure of short-term mating orientation. Study 2 (n = 402), which was preregistered, found that insurance-based mortality risk associated positively with impulsivity and negativel..., , # Mortality risk estimates from life insurance policies predict individual differences in human behavioral traits

---

Data were collected from two U.S. online participant samples (N = 270 and N = 402), screened to include only individuals who had purchased individual life insurance policies within the past five years. In both data sets, participants were asked for the risk ratings they had been assigned by the insurance company, and to complete self-report instruments measuring constructs relevant to psychometric life history (especially the present-future trade-off). In the second data set, participants were also asked to indicate their self-estimated lifespan, and were asked to complete three instruments measuring recalled childhood environmental harshness. R code used to analyze the data is also provided.

## Description of the Data and file structure

#### Lukaszewski-Manson-Study-1-R.csv

Key to column headings

female: 1 = yes

age_bin: (1 = younger than 25 years, 2 = 25-29, 3 =..., We have received explicit consent from our participants to publish the de-identified data in the public domain.
We have de-identified the data by removing all individually identifying information (IP addresses, and for the six in-person participants in Study 2, their names) from data files before uploading them.