The aim of this study was to investigate the effects of music introduced and removed during a 10-km cycling time trial with reference to Rejeski's parallel processing theory and Karageorghis, Terry and Lane's conceptual framework for the prediction of responses to asynchronous music during sub-maximal exercise. A range of performance variables, ratings of perceived exertion, positive affect, negative affect, and blood lactate were assessed. Eleven males (mean age=24.9, s=6.1 years) completed a 10-km time trial under three conditions; no music, music played initially then removed between 5-10 km, and music played between 5-10 km only. Variables of time, power, cadence, speed, RPE, blood lactate, positive and negative affect were analysed using a ConditionxDistance ANOVA. There was no significant main effect for music conditions for the performance variables, perceived exertion, blood lactate, and affect (p>0.05). Nevertheless, a significant interaction effect for ConditionxDistance was found for cycling speed, with participants cycling 1-1.25 km/h faster at the start of the music introduced time trial than in both the music removed and no music time trials (p<0.05). The results indicate that performance and affect during a 10 km time trial are influenced by the introduction and/or removal of music during exercise and this finding can be used to extend current theory as it does not specifically address the periodic use music. The fact that participants exercised harder when they expected music to be introduced at a later stage illustrates the behavioural influences that music can engender during self-paced exercise.