On the spring scale, the extra downward force will register as an increase the effective weight of whatever pan and support structure there is and cause the scale to show a higher weight. When the acceleration stops the spring will spring back and overshoot the zero mark and then come back with the oscillations dying out due to friction.

It's assumed that the electronic scale works on the piezo-electric compression forces on crystals. Presumably these would work much the same way as the spring scale, but on a smaller scale. However, even the small scale oscillation would presumably be interpreted as weight changes as large as those on the spring scale. But also, however, there might be some intelligence built into the device to lock the initial reading that was taken during acceleration to prevent the maximum weight from disappearing until the oscillations had died down.