""Anxiety disorders"" are the commonest mental disorders in New Zealand, but their diagnosis is still based on clinical symptom check lists with no biological markers to diagnose specific causes. Over recent decades, we have developed: (1) a detailed theory of the brain systems controlling threat-avoidance and threat-approach, where, in specific brain structures, activity generates specific normal behaviours, hyperactivity generates abnormal behaviours, and hyper-reactivity (hypersensitivity to input) generates specific clinical syndromes; and (2) a specific non-invasive measure of activation of the threat-approach system in humans. We will test if our new biological marker divides untreated ""anxiety"" patients with superficially similar clusters of symptoms into distinct high scoring and low scoring groups with different treatment-responses. If positive, the results will validate the first human biological marker for a theoretically-defined ""anxiety"" syndrome. This marker should predict treatment efficacy better than current symptom-based diagnoses, greatly improving treatment outcomes and cost-effectiveness.