Heart failure (HF) is associated with activation of various neurohumoral factors including the renin–angiotensin–aldosterone and sympathetic nervous systems. In HF patients, elevated sympathetic nerve activity (SNA) is reflected by increased whole body, as well as regional (e.g. cardiac and renal) noradrenaline spillover (Kaye et al. 1994; Petersson et al. 2005), and increased sympathetic nerve discharge (Leimbach et al. 1986). Importantly, the degree of sympathoexcitation predicts mortality in HF patients (Petersson et al. 2005). Consequently, an immense amount of research effort has been directed towards the identification of mechanisms and circuitry underlying elevated SNA in HF. Neurons of the hypothalamic paraventricular nucleus (PVH) regulate SNA in a variety of physiological conditions and pathophysiological states through mono- and polysynaptic projections to the medulla and spinal cord (Guyenet, 2006). Evidence from studies, largely conducted in anaesthetized rodents, suggests HF activates PVH neurons through a reduction in GABAergic function (Patel, 2000). A critical question is the extent to which PVH neurons contribute to elevated SNA in HF. In this issue of The Journal of Physiology, Ramchandra and colleagues (2013) provide novel data utilizing chronic SNA recordings in conscious sheep to address this important question. Interruption of PVH neurotransmission by radiofrequency ablation or neuronal inhibition by microinjection of glycine or the GABAA agonist muscimol failed to lower cardiac and/or renal SNA in both control and HF sheep. Furthermore, blockade of PVH GABAA receptors with injection of bicuculline increased cardiac SNA, mean arterial blood pressure and heart rate in control but not HF sheep. The absence of sympathetic-cardiovascular responses to GABAA receptor activation or blockade in HF versus control sheep confirms a reduced GABAergic function in the PVH. More importantly, these findings challenge the general notion that PVH neurons contribute to sympathoexcitation in HF. Several issues may explain the discrepancies between the present and past studies: (1) differences between anaesthetized versus conscious animals, (2) differences in the central nervous system circuitry/mechanisms between HF models (i.e. coronary ligation versus ventricular pacing), and (3) the possibility that PVH neurons contribute to the development, but not maintenance, of elevated SNA in HF. A second intriguing finding is the differential contribution of PVH neurons to cardiac and renal sympathoinhibitory responses to volume expansion. In conscious sheep, PVH inhibition with injection of the GABAA agonist muscimol reversed the renal, but not the cardiac, sympathoinhibition during intravenous infusion of gelofusine. At least two critical questions arise from these findings. First, PVH neurons abundantly express the vesicular glutamate transporter but are not GABAergic (e.g. GAD67 positive) (Stocker et al. 2006). If volume expansion reduces the discharge of spinally projecting PVH neurons (Lovick & Coote, 1988), why does further inhibition using muscimol reverse the sympathoinhibition? Second, how do PVH neurons differentially regulate SNA to various end organs? The notion that PVH neurons can selectively increase SNA to one organ but not to another has been reported previously (Deering & Coote, 2000; Ward et al. 2011). Therefore, are there distinct populations of PVH neurons that regulate renal versus cardiac versus muscle SNA? Do these neurons receive different anatomical and neurochemical inputs? Previous studies have utilized transneuronal viruses to map the neuronal circuitry associated with a specific end-organ, but limited data exist to demonstrate whether PVH and other central nuclei contain overlapping or distinct populations of neurons to permit differential regulation of SNA. The findings of Ramchandra and colleagues (2013) provide evidence against a role of PVH neurons in elevating SNA in HF but emphasize the ability of PVH neurons to differentially control sympathetic outflow to distinct end-organs. The present findings highlight our limited understanding of how many CNS structures are anatomically and neurochemically connected to coordinate sympathetic outflow to a variety of organs.