Figure 4.
The consequences of deblurring and varying source diameter. (A) Peak of derived release current versus “deblurring length,” the FWHMz of the PSF used for deblurring, which ranges from 0 (no correction) to 1.4 μm (full correction). FWHMxy was equal to FWHMz/3.5. Filled symbols: simulations for a source as in Fig. 2 with (♦) or without noise at 10% of F 0 (same as in Fig. 3). Open circles: results for an average of 157 experimental sparks illustrated in Fig. 5 and Fig. 6. Note that the dependence of the current estimate is monotonic and smooth for simulations without noise (•), but goes through a peak or an abrupt change in slope for the experimental average and simulations with noise, at or about a deblurring length of 0.9 μm. (B) Peak of derived release current and (C) FWHM versus source diameter, for simulations of sparks with a 20 pA/5 ms current from a source of variable diameter. Filled symbols: simulations with noise added at 10% of F 0. The analysis was carried out as it is done with spark averages (see text). Open symbols: simulations with noise added at 30% of F0. The analysis was as done with individual sparks (see text). Represented are averages ± SEM for six realizations of pseudorandom noise. Note a good agreement between the low noise estimates and the simulation current (dashed line).