Selective breeding against feather pecking in laying hens depends on identification of individual birds with the lowest feather pecking activity. If certain behavioural traits are phenotypically and genetically associated with, or predictive of, feather pecking activity then tests for those traits may offer a quicker method of identifying suitable parent birds. In a previous study, pairs of pullets that pecked most frequently at a feather bundle also avoided a novel object in a separate test. In the present study, 319 Brown pullets were tested with a novel object at 7-9 weeks of age to determine whether response to novelty predicted either pecking at feather bundles presented in two different ways (loose or fixed) at both 11-13 and 25-27 weeks of age, or feather pecking in the home pen between 15 and 33 weeks of age. Response to novelty did not predict pecking at feather bundles. There were no associations between the amount of pecking directed towards different types of feather bundle, or between pecking at feather bundles and pecking at the feathers of live birds. Response to novelty also failed to predict tendency to feather peck, although recorded levels of feather pecking, especially severe feather pecking, were relatively low. However, when birds with a varied range of responses to a novel object were housed together, more of the birds feather pecked (P < 0.05), they feather pecked more consistently (P < 0.05) and showed more pecking at the environment in a test than birds with similar responses to novelty.