Changes in DCN circuit connectivity parameters from the model of Hancock and Voigt ( 1999 )
| Input population → target | Frequency offset CAB (octave) | Integration bandwidth BWAB (octave) | Number of inputs × synaptic weight NAB × gAB |
|---|---|---|---|
| AN → T2 | 0.0 | 0.2 | 23.1 (26.1) |
| AN → WBI | 0.0 | 2.0ab (3.0) | 10.8 (8.4) |
| AN → T4 | 0.0 | 0.24 | 2.75 (4.8) |
| WBI → T2 | 0.3a (0.0) | 0.05 | 7 (21) |
| WBI → T4 | 0.05 | 0.1 | 0 - 19.8a (9) |
| T2 → T4 | −0.1a (0.0) | 0.2 | 47.25 |
| NSA → T4 | N/A | N/A | 0.7 |
Original values are shown in parentheses when they have been changed. Input population parameters: CAB is the frequency offset of the input population from the target cell; BWAB is the integration bandwidth of the input population; NAB is the number of cells in the input population bandwidth [unlike by Hancock and Voigt (1999), NAB is determined by the bandwidth as one input unit per 1/100 octave]; gAB is the synaptic strength of the input population relative to the resting conductance of the cell. Because the product NAB × gAB, and not gAB alone, determines the net effect of each input population, only NAB × gAB is shown for comparison with original values. N/A, Not applicable; T2, type II; T4, type IV; NSA, nonspecific afferent.
↵ a These changes had the largest effect on model behavior.
↵ b Bandwidth is effectively 2.0 because of Gaussian-distributed weighting of inputs > 3.0 octaves, with the largest weights concentrated over a 2.0 octave bandwidth.