The present work investigates the cardiovascular activity linked to the cognitive workload induced by the realisation of a supplemenary mental task to driving in a simulator. Specifically, heart rate (HR) and heart rate variability (HRV) are computed to establish the physiological modulations under different types of cognitive demands in order to identify indicators to monitor driver state. Given that driving is a complex task where the executive functions requiring memory, visual attention, decision-making and flexibility are required, the study of concurrent tasks sharing the same cognitive resources is crucial to assess driver's mental workload and performance. For instance, to determine the influence of driver's distraction on driving performance, different additional tasks such as conversation and mobile handling in a real-setting have been employed in previous research works. On the other side, mental arithmetic tasks have been largely employed to increase cognitive workload to analyze distinct aspects of driver behavior. To disentangle how different cognitive workload categories and psychological processes impact driver physiology, this study, based on the Baddeley's model of the working memory (WM), explores the differentiated effects of two cognitive mechanims: listening associated or not with mental processing. According to this model, there are four subdivisions composing WM: the phonological loop which is involved in our ability to maintain and process verbal information, the visuospatial sketchpad responsible for mental image manipulation, the episodic buffer which relays between WM and long-term memory and the central executive for the administration of the three other subdivisions. Moreover, the concept of sensory memory is considered as a primary information processing stage. Specifically, the echoic memory, which is a short-term storage of auditory information, is relevant for auditory distraction. Besides the behavioral data, other objective markers are desirable to monitor driver's cognitive workload and to foresee potential risky situations. One of the most sensitive and cost effective measure to this aim is the electrocardiogram (ECG). HR increments with increasing task demand, manipulated by means of an 'N-back' task, while driving. These results are corroborated by several studies on the topic. Furthermore, the temporal HRV parameters are suggested to be suitable for the identification of instantaneous stress indicators. Indeed, HRV is pointed as an accurate marker of emotion-regulatory ability. Following this line, the present study aims at complementing the results reported in [9], where the authors were interested especially in evoked cardiac response, by analyzing more deeply the links between WM components, mental arithmetic and cardiac signals within longer time periods.