The Dely line Hypothesis (mainly Reiss et al.)
The Delay Line Hypothesis (mainly Reiss et. al.) The Delay Line Hypothesis
describes a possible neuron or set of neurons that acts to delay the first syllable
by a set time (like 35 ms for Gryllids), and then compare the output of the
delay line with the currently incoming signal. If they coincide, the signal
is at the right frequnecy.
According to the dalay line hypothesis we created a neuronal model consisting of 3 cells. The AN1 cell is getting input from
from a pulse generator and a random noise generator. The IN neuron is an excitatory interneuron firing after a delay of Delta-t
of 30 ms. This delay attenuates the spikes representing syllables for exactly that time, when the next input from the AN1 is
anded at the dendrite of the rec-neuron. The tuning the synaptic wheigth at the dendrite of the rec-cell prevents the neuron
from spiking upon input of the AN1 solely. The same holds true for the input comming from the Interneuron. Temporal
summation of both inputs lead to a discharge of the rec-cell.
The above graphics shows the response (membrane potential) of the recognising cell (rec-cell) for a SRI of 34 ms under noisy conditions (40/sec).
The above graphics represents the response magnitude of the rec-cell for different
SRI's under noisy conditions (40/sec). This result fitts the result of the behaviour
response very close. By reducing the average noise rate to 20/sec, the red
line (standing for the noisy condition) is lowered to appr. 30% response magnitude.
Back to Introduction
Other computer simulations:
Common element simulation
Rebound effect simulation
Anded filter hypothesis
