Responses of one example unit to temporally isolated call-echo elements and to the natural echolocation sequence.
(a) Frequency tuning curve calculated from responses to different intensity-frequency combinations of a pure tone. The tuning curve demonstrates that the unit is sensitive to high frequencies. (b) Delay tuning curves calculated from neuronal responses to artificial frequency modulations that simulated the second harmonic of the bats’ calls (left) or to natural call-echo pairs (right). Different combinations of call-echo delays (2–22 ms in steps of 2 ms) and echo sound levels (80, 70 and 60 dB SPL) were presented to the animal. The interstimulus time interval was set to 400 ms. The delay tuning curves show that the unit responded strongly to short delays at around 4 ms. Responses to the echo alone [E] without preceding call is given in the last vertical bins of the data field. (c) Top: Oscillogram of the echolocation sequence. Vertical red lines indicate time borders between different call-echo elements that were used for segmenting the sequence. Bottom: Raster plots calculated from the response to the echolocation sequence (black dots) and to the elements (colored dots), where the call-echo elements were randomly presented with a 400 ms interstimulus time interval. To visualize which spikes were elicited in responses to which call-echo element, the responses to consecutive call-echo elements are indicated through alternating colors. Red arrow points to a characteristic initial response. Horizontal gray lines separate raster plots obtained at three sound levels. (d) Post-stimulus time histograms (PSTHs; binsize = 5 ms) for the three sound levels and for the sequence (blue) and element (orange) situation. Note that the response to the sequence is suppressed resulting in temporally sharper responses than those in the element situation. Horizontal dashed orange and blue lines indicate thresholds (50% from maximum bin). See also Supplementary Fig. 1.