The objective of this study was to examine relationships between behaviors in three different standard tests of temperament in pigs. Sus scrofa. For the backtest, pigs (n = 575) were placed in the supine position for 60 s. Each pig was tested twice. Amount of time struggling and number of struggle attempts during each test was recorded and combined to give total time struggling and total struggle attempts. Increased total time struggling was weakly associated with low preweaning average daily gain (r = -0.24; P < 0.01) and low adjusted 21-day weight (r = -0.19; P < 0.05). Pigs were ranked, within replicate, based on total time struggling. Piglets in the top (longest total time struggling) and bottom (shortest total time struggling) 10% of their respective replicates were selected for additional coping style tests (n = 120). These tests were the resident intruder test, to measure aggression toward other pigs, and novel object test, to measure behavioral and physiological responses to a novel object. Correlations between total time struggling during the backtest and time to contact, time between contact and attack, and time to attack during resident intruder test (r = 0.01, r = -0.03, and r = -0.03, respectively) did not differ from 0. The backtest performance was also not correlated with latency to explore ((r = 0.13), time exploring (r = -0.05), or cardiac response to the novel object during novel object tests (P > 0.1). Pigs grouped based on total time struggling did not differ consistently during resident intruder test or novel object test trials. In conclusion, results of the present study do not support the theory that backtest performance is indicative of behavior in a test of aggression (resident intruder test) or a test of fear/curiosity in response to novelty (novel object test). (c) 2012 Elsevier B.V. All rights reserved.