One-day-old chicks (Gallus domesticus) were trained on a one-trial passive avoidance task in which the aversive stimulus was an unpleasant lasting substance, methylanthranilate (MeA). Control birds were presented with a water (W) coated bead. Five minutes after training a group of McA-trained chicks were given a brief sub-convulsive trans-cranial electric shock, which rendered half amnesic whilst the remainder were able to show recall for the aversive stimulus. Thirty minutes after training birds were killed and quantitative receptor autoradiography was used to determine NMDA sensitive [3H]L-glutamate binding in specific regions of the forebrain of: (i) MeA-trained chicks; (ii) water-control chicks; (iii) MeA-trained electroshocked chicks showing recall; and (iv) MeA-electroshocked chicks amnesic for the aversive stimulus. Increases (805 in the left lobus parolfactorius and 67% in the left intermediate medial hyperstriatum ventrale) in NMDA sensitive [3H]L-glutamate binding occurred in electro-shocked chicks which showed recall of the aversive experience but were absent in MeA-trained chicks rendered amnesic by electro-shock. The increased binding in electroshocked MeA-trained birds which showed recall was similar to that observed previously in MeA-trained birds (without electroshock), compared to water control birds, whereas binding levels in McA-trained electroshocked amnesic birds were not different from those of water control birds. These data argue strongly in MeA-trained electroshocked amnesic birds were not different from those of water control birds. These data argue strongly that alterations in binding to glutamate receptor sub-types are specific to memory formation for the passive avoidance task.