Associative learning shapes mixture representation and improves perception of relevant odors.

MARACHLIAN EMILIANO; LOCATELLI FERNANDO

Huerta Grande, Cordoba

Congreso; Congreso SAN 2013; 2013

Sociedad Argentina de Investigaciones en Neurociencias

Associative learning shapes mixture representation and improves perception of relevant odors.   Emiliano Marachlian  and Fernando Locatelli   Laboratorio de Neurobiología de la Memoria, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires. IFIByNE- CONICET. Argentina   Odors in nature are complex mixture, in which irrelevant components may hide the presence of relevant ones. In the present work we ask if olfactory experience increases the ability of animals to detect the relevant odors. The olfactory system of insects provides a good model for this study. The antennal lobe is the first processing center for olfactory information in the insect brain. The local network conformed by excitatory and inhibitory local neurons transforms the olfactory information before it leaves the antennal lobe to other brain areas. We train bees to pure odors and perform calcium imaging in projection neurons of the antennal lobes to measure neural activity patterns elicited by pure odors and mixtures that contain these odors. On the basis of patterns obtained for naïve animals we assayed different algorithms that allow accurate prediction of the pattern elicited by the mixture. The prediction algorithms were later applied to animals that have been trained on appetitive conditioning using as conditioned stimulus one of the components of the mixture. We found that the representation of the mixture in trained animals deviates from the mixture predicted for naïve animals. This deviation is in favor of the representation of the rewarded odor.  In addition we found a general decrease in the activity elicited by odors in the trained animals.