omputational modeling interpretation underlying elevated risk-taking propensity in non-labor income

strong>Abstract：</strong>Individuals have been observed to show higher propensity to make risk investments using non-labor income compared to labor income, although the underlying mechanisms behind this phenomenon remain unclear. In this study, we proposed that non-labor income leads to a higher prior expectation of risky investment and a reduced sensitivity towards losses. To quantitatively test this hypothesis, we employed computational modeling. A total 103 participants were recruited and completed the Balloon Analogue Risk Task (BART) with an equal monetary endowment, either as a token for completion of survey questionnaires (labor income) or as a prize from a lucky draw game (non-labor income). We found that individuals endowed with non-labor income made more risky investments in the BART compared to those with labor income. To formally compare the differences in the dynamic risk investment process between individuals with different source of income, we built four candidate computational models (Bayesian Sequential Risk-taking Model, Target Model, Scaled Target Learning Model and Scaled Target Learning with Decay Model (STL-D)). Through computational modeling, we found that within STL-D, the optimal model, the non-labor income group preset a higher targeted number of pumps at the beginning, showed a lower learning rate towards loss trials where the balloon exploded, and had lower behavioral consistency. Our study suggests that the increased tendency for risky investments with non-labor income can be attributed to an increase in prior expectations on risk-taking and a diminished sensitivity towards loss. These findings provide potential intervention targets to mitigate irrational investments associated with non-labor income.<br />