4
N. Andreakos et al.
Axo-axonic cell
Basket cell
Bistratified cell
Time (ms)
OLM cell
0
2 0 0
-50
0
50
V (mV)
0
2 0 0
-50
0
50
V (mV)
0
2 0 0
-50
0
50
V (mV)
0
2 0 0
-50
0
50
V (mV)
0
200
50
-50
0
50
V (mV)
Pattern pyramidal cell
THETA
0 o
180 o
270 o
90 o
Start of
Theta
End of
Theta
360 o
Duration (250 ms) of one Theta cycle
Storage
Retrieval
Fig. 2. Voltage traces of model cells with respect to a single theta cycle.
2.3 Network Training and Testing
To test the recall performance of the model the methodology described in [1] was adopted.
Briefly, a memory pattern was stored by generating weight matrices based on a clipped
Hebbian learning rule; these weight matrices were used to pre-specify the CA3 to CA1
PC connection weights. Without loss of generality, the input (CA3) and output (CA1)
patterns were assumed to be the same, with each pattern consisting of N (N = 5 or 10
or 20) randomly chosen PCs (active cells per pattern) out of the population of 100. The
100 by 100 dimensional weight matrix was created by setting matrix entry (i, j), w ij = 1
if input PC i and output PC j are both active in the same pattern pair; otherwise weights