Previous music–driving research has largely failed to isolate musical characteristics from participants’ sociocultural and age-related music preferences. Accordingly, the proposed work followed a rigorous music-selection procedure that incorporated several objective components to establish the content of music conditions. We investigated the potentially distracting effects caused by the processing of lyrics by exposing drivers to the same piece of music with/without lyrics and at different intensities (60 dBA and 75 dBA).Using a counterbalanced, within-subjects design, we recruited 34 volunteers (Mage = 22.2, SD = 2.01 years). We compared six simulator conditions that comprised low-intensity music with/without lyrics, high-intensity music with/without lyrics plus two controls – ambient in-car noise and spoken lyrics. There was an omnibus main effect of condition for the NASA Task Load Index with step-down analyses indicating that, for Temporal Demand, a low-intensity with lyrics condition imposed lower demand than ambient in-car noise. Moreover, there were lower scores for Frustration in the music conditions vs. spoken lyrics. For affective valence, the four music conditions yielded higher scores than control conditions. For affective arousal, the music conditions yield lower scores than in-car noise. HRV data did not yield a main effect of condition, albeit a sex difference emerged for SDNN (men > women). Although there were some differences in subjective outcomes, these were not paralleled by HRV data, used as an objective index of emotionality. The most effective music condition in terms of mental state during simulated urban driving was soft, non-lyrical music.