Anorexia nervosa (AN) is a significant mental health issue, yet little is known regarding the underlying factors that contribute to illness onset and maintenance. AN has the highest death rate of any mental illness and a recovery rate of less than 50% in surviving patients. Thus gaining an understanding of the neurobiological underpinnings of this condition is of great importance for tailoring appropriate treatments for this illness. To date, the scientific study of AN is considerably variable with few empirical findings being consistently reported. This has been particularly true in relation to the illness' underlying neurobiology. Unlike many other psychiatric conditions, the brain regions and neurotransmitter systems involved in the illness are unknown. A key review published by our group in ANZJP (Phillipou et al., 2014a) highlighted the lack of consistency of the neurobiological literature in AN, and emphasised the need to utilise tasks known to employ specific brain networks to uncover the neurobiological underpinnings of this illness. This was the specific aim of our recent research programme in which we compared patients with AN to healthy individuals on a range of saccadic eye movement tasks. Saccadic eye movement tasks are well-characterised research tools frequently utilised in psychiatric populations to understand cognitive deficits while drawing inferences about compromised neural pathways. Our recent research programme was the first detailed analysis of saccadic eye movements in AN.Eye movements have been thoroughly assessed in aetiological investigations of other psychiatric and neurological conditions, providing valuable information on brain regions and neurotransmitter involvement in these illnesses. The neural circuitry of saccadic eye movements is well understood, and deficits in different saccadic parameters (e.g. error rate, latency, gain and peak velocity) are indicative of specific neurobiological abnormalities. Collectively, our recent research has identified four saccadic eye movement abnormalities in AN, all of which are consistent with saccadic disinhibition and dysfunction of gamma-aminobutyric acid (GABA) in the superior colliculus (SC). These are as follows: (1) hyperscanning (i.e. increased fixations of shorter duration) during two different tasks of face and body viewing, respectively (Phillipou et al., 2015a(Phillipou et al., , 2015b; (2) shorter prosaccade latencies (Phillipou et al., 2016); (3) increased inhibitory errors on a memory-guided saccade (oculomotor delayed response) task (Phillipou et al., 2016); and (4) increased rate of saccadic intrusions, called square wave jerks, during a fixation task, which in combination with anxiety levels were identified as a potential biomarker of AN (Phillipou et al., 2014b).These findings have encouraged the development of a hypothesis for a novel neurobiological underpinning to AN, and have implicated a brain region and neurotransmitter system in the aetiology of the illness that have been previously overlooked. Although further researc...