Successful antidepressant treatment has been associated with concomitant changes in brain function, consolidated as long as treatment is continued and remission is preserved. The present study aimed at assessing the impact of prior antidepressant treatment on brain function in currently depressed but unmedicated individuals by investigating for any differences between antidepressant-naïve vs. antidepressant-experienced subjects. Fifty right-handed patients (22 medication-naïve vs. 28 medication-experienced), suffering from major depression participated in the study. They all underwent a standardised clinical interview and psychometric assessment combined with neurobiological tests (brain SPECT, Dexamethasone Suppression Test, Dexfenfluramine Challenge Test, electro-oculogram, flash-electroretinogram and flash-visual evoked potentials and pattern-reversal visual evoked potentials). No significant differences between medication-naïve and medication-experienced depressed subjects were found in terms of the neurobiological markers assessed, after controlling for age, sex, age at onset, number of depressive episodes, depression subtype (melancholic, atypical or undifferentiated) and severity of current episode. Unmedicated currently depressed patients, no matter their previous exposure to antidepressants, show similar changes in brain function. This does not necessarily mean that antidepressants do not have a long term effect on brain physiology, since not all patients relapse. However, it seems that those patients who relapse after stopping medication, seem to 'regress' to an 'as if never medicated' state, with regard to brain function. These findings might suggest that continuous maintenance treatment with antidepressants is essential for patients at high risk to relapse. Alternatively, they might suggest that our methodology assesses only a shallow and mainly state part of the pathophysiology of depression.