Studies in the neurobiology of social phobia have used neuroendocrine, naturalistic and chemical challenges, pharmacological probes, neurotransmitter system measures, peripheral receptor binding and magnetic resonance measures. Studies of the hypothalamic-pituitary-adrenal and hypothalamic-pituitary-thyroid axes have been largely unrevealing; adrenaline, carbon dioxide, caffeine and yohimbine tests have provided mixed results; probe studies using L-dopa, clonidine and fenfluramine have provided some evidence of post-synaptic serotonergic abnormality; studies on platelet and lymphocyte binding have failed to distinguish social phobia from other groups; magnetic resonance imaging and magnetic resonance spectroscopy studies suggest possible differences between patients with social phobia and healthy controls in respect of dopamine, serotonin and second-messenger function. In aggregate, these studies have provided some neurobiological basis for separating social phobia from panic disorder and non-psychiatric healthy controls.