Neurophysiological investigations of a recognition memory system for imprinting in the domestic chick

The responsiveness of neurons in a region of the chick brain involved in the learning process of imprinting, the right intermediate and medial hyperstriatum ventrale (right IMHV), has been investigated in unanaesthetized, trained and untrained chicks. The results demonstrate that neuronal responsiveness in this region reflects a variety of behavioural consequences of imprinting and is markedly altered as a result of the learning process. Groups of chicks (nine in each group) were either dark-reared or trained (imprinted) by exposure to a rotating red box or a rotating blue cylinder. Recordings of single or small groups of neurons were subsequently made from 156 sites in the right IMHV while the 2-day-old chicks were free to move in a running wheel. There was a highly significant increase in the proportion of sites responsive to the stimulus used to train the birds compared to the proportion responsive to that stimulus in dark-reared birds (30 and 9% respectively). These changes were found when either the red box or the blue cylinder was used to train the bird, the changes being similar for both stimuli. There was also a significant increase in the mean magnitude of the change in neuronal activity on stimulus presentation for the training stimulus compared to the same stimulus when not used in the bird's training. No significant effects of the training experience of the chicks were found upon either the magnitude of evoked activity or the proportion of sites responsive to a rotating stuffed jungle fowl or the sound of the maternal call. The presence of the training stimulus was selectively signalled by the response at certain sites. At other sites there was response generalization across stimulus shape or colour. A comparison with results for the left IMHV demonstrates both similarities and differences in neuronal responsiveness between the two regions. In both regions imprinting selectively enhances neuronal responsiveness to the training stimulus. However, for trained birds the mean proportion of sites responding to whichever of the red box or the blue cylinder was not used in the bird's training was significantly lower in the right than the left IMHV. These results are discussed in relationship to previously reported asymmetries in the response of the right and left IMHV regions to imprinting. A model is introduced to explain the physiological findings. The effects of training on right IMHV neuronal function are consistent with a long-term role for this region in the recognition memory of imprinting.