Figure 3: Desynchronization reduces response intensity without altering stimulus identity.

(a) Evolution of odour-evoked ensemble responses over time is shown after dimensionality reduction using principal component analysis. The number of neurons (n) used for this analysis and the percentage of variance captured in the first three dimensions are shown in each plot. Numbers near response trajectories indicate time in seconds since odour onset. Red and blue trajectories are solitary introductions of two different odourants. Black trajectories reveal the ensemble responses following the introduction of the second odourant in the sequence. (b) Blue and red traces indicate pattern match between ensemble activities generated following the introduction of the second odourant with the response templates obtained for the component odourants (see Methods). (c1) Representative raster plots revealing responses of two different Kenyon cells in the insect mushroom body to odourants that do not produce coherent PN response during odour overlaps. (c2) Kenyon cell PSTHs (mean±s.d.; n=10 trials) are shown as a function of time for solitary (red) and overlapping (black) introductions of six different odourants (70 Kenyon cells in total; *P<0.01, paired t-test for peak firing rate comparison). Note that the six odourants used are those that evoked antennal lobe ensemble responses that were less synchronous when presented following another stimulus. (d1, d2) Similar plots as c1, c2, but analysing all Kenyon cell responses to those odourants that revealed coherent projection neuron responses no matter how they were introduced.