Discriminatory social norms drive high levels of gender inequality in India. However, there is a paucity of evidence on how gender discrimination manifests in economic decision-making with real payoffs in a rural, underdeveloped setting. In this paper, I present a lab-in-the-field experiment using incentivized trust and dictator games to distinguish between statistical and taste-based gender discrimination. Negative stereotypes that manifest as lower trust is interpreted as statistical discrimination. Prejudice that manifests as lower trust and social preferences is interpreted as taste-based discrimination. Next, I evaluate whether a behavioral nudge can influence discriminating individuals’ preferences over gender versus previous trustworthiness. The evaluation nudge tests whether moving from separate (single-choice) to joint (multiple-choice) evaluation setting triggers a shift from gender-biased to payoff maximizing decision-making. Results indicate that participants demonstrate statistical discrimination. Signaling higher trustworthiness leads to gender unbiased decision-making under joint evaluation, but not under separate evaluation.