<p>This study uses an eye-tracking experiment to investigate the effects of accountability on favoritism in subjective performance evaluations. We develop and test theory suggesting that the effectiveness of accountability to alleviate favoritism depends on managers’ cognitive processing of performance information. We argue that accountability will induce managers to examine performance information more thoroughly, which may entail having to scrutinize information that reflects unfavorable on preferred subordinates, thus causing cognitive dissonance. To resolve this dissonance, managers can engage in motivated reasoning by cognitively discounting such information. The more managers reduce cognitive dissonance by discounting unfavorable information, the lesser is the favoritism-mitigating effect of accountability. Consistent with this prediction, our results show that the higher the proportion of cognitive processing effort managers devote to information that reflects unfavorable on a&#xa0;preferred subordinate, the lower the efficacy of accountability in mitigating favoritism. We contribute by enhancing our understanding of the effects of accountability in subjective performance evaluations.</p>