Konstantinos N. FountoulakisRecently, an article by Hebert et al. () reported a rod retinal response anomaly in high-risk offspring of patients with schizophrenia or bipolar disorder. These authors reported significant results mainly concerning the rod b-wave amplitude but only a trend for the latency time. However, the reliability of amplitude measurement is problematic even if the authors tried to control it by means of gender and age. These results are in accordance with a previous study from our group that utilized electro-ophthalmogram (EOG) and electroretinogram (ERG) in depressed patients (). The study reported that depressed patients manifested a decrease of both EOG dark trough and light peak values in comparison with control subjects, while the Arden ratios were within normal range, a finding also reported previously by other authors (). The flash-ERG (we used only latency time and not amplitude) under photopic and scotopic conditions (the latter was not published) suggested that there are no differences between depressed patients and control subjects. The comparison between melancholic and atypical patients suggested no difference in EOG and ERG recordings. Another study form our group () used pattern reversal visual evoked potentials and revealed that there might be an underactivation of the anterior right hemisphere in melancholic depressives and a hyperactivation of the same region in atypical depressives. Neither our studies nor Hebert et al. () support a hypersensitivity to light hypothesis for depression.