Neurons are highly specialised cells with a large bioenergetic demand, and so require a healthy mitochondrial network to function effectively. This network is compromised in many neurological disorders, in which damaged mitochondria accumulate. Dysfunctional mitochondria can be removed via an organelle-specific autophagic pathway, a process known as mitophagy. The canonical mitophagy pathway is dependent on the actions of PINK1 (PTEN-induced putative kinase 1) and Parkin and has been well studied in immortalised cells and cultured neurons. However, evidence for a role of this mitophagy pathway in the brain is still limited, and studies suggest that there may be important differences in how neurons respond to mitochondrial damage in vitro and in vivo. Here, we first describe the evidence for a functional PINK1/Parkin mitophagy pathway in neurons, and review how this pathway is affected in disease models. We then critically evaluate the literature by comparing findings from in vitro models and more recent in vivo studies in flies and mice. The emerging picture implicates that alternative mitophagy pathways operate in neurons in vivo. New mouse models that employ fluorescent biosensors to monitor mitophagy in vivo will be instrumental to understand the relative role of the different clearance pathways in the brain under physiological and pathological conditions.