Day-old chicks (Gallus domesticus) were trained on a one-trial passive avoidance task where the aversive stimulus was an unpleasant tasting substance, methyl anthranilate. Chicks were killed 6.5 h after training. The kinetic parameters of [3H] alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ([3H]AMPA) binding were determined using quantitative receptor autoradiography and Scatchard analyses in 15 discrete forebrain regions of trained and control (water-trained) chicks, revealing two components of binding in each. KD values showed some regional variation, but were 22.2 +/- 1.1 nmol l-1 for the high-affinity component and 685 +/- 25 nmol l-1 for the low-affinity component of binding to whole forebrain sections from control chicks. Analyses also revealed that Hill coefficients were significantly less than 1 in all regions measured. A significant decrease in KD for the low-affinity component occurred bilaterally in the intermediate and medial hyperstriatum ventrale (IMHV; left, 34.8%; right, 33.3%), a region that has previously been shown to be implicated in the processes of memory formation, following passive avoidance training. A significant decrease in KD for the high-affinity component occurred in the right palaeostriatum augmentatum (19.5%). Significant decreases in Bmax accompanied the KD alterations in both cases. Additionally, bilateral intracerebral injections (administered 4.5-5.5 h after training) into the IMHV of 500 nmol l-1 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), a selective antagonist of non-NMDA glutamate receptors (particularly AMPA receptors), resulted in amnesia for one-trial passive avoidance training in day-old chicks tested 6.5 h after training.(ABSTRACT TRUNCATED AT 250 WORDS)