This paper studies the effect of providing relative performance feedback information on individual performance and on individual affective response, when agents are rewarded according to their absolute performance. In a laboratory set-up, agents perform a real effort task and when receiving feedback, they are asked to rate their happiness, arousal and feeling of dominance. Control subjects learn only their absolute performance, while the treated subjects additionally learn the average performance in the session. Performance is 17 percent higher when relative performance feedback is provided. Furthermore, although feedback increases the performance independent of the content (i.e., performing above or below the average), the content is determinant for the affective response. When subjects are treated, the inequality in the happiness and the feeling of dominance between those subjects performing above and below the average increases by 8 and 6 percentage points, respectively.