Recent studies have demonstrated that speech perception with cochlear implants can be significantly affected by electrode configuration. Contrary to expectations, broader configurations (monopolar or broad bipolar) produced equal or better speech recognition compared with narrower configurations (narrow bipolar or common ground). One hypothesis that would account for these results is that broader configurations excite larger populations of neurons providing a more robust representation of information on each channel of the prosthesis. It is known that the number of neurons excited by an electrical stimulus increases considerably as the stimulus level increases. Furthermore, many types of discrimination improve as a function of stimulus level. If the discrimination improvements seen with increasing stimulus level are due to increasing the size of the neural population carrying the signal, and if broadening the electrode configuration also increases the size of the activated neural population, then one would expect level and electrode configuration to affect discrimination in similar ways. To test this hypothesis, we studied several types of discrimination as a function of level and electrode configuration in four nonhuman primates with cochlear implants. We tested electrode configurations that produced current fields ranging from very restricted (tripolar) to broad (parallel monopolar). For each configuration, pulse-rate discrimination, amplitude-modulation-frequency discrimination, and level discrimination were tested at current levels spanning much of the psychophysical dynamic range. Results showed large effects of current level on discrimination in many cases. However, effects of electrode configuration at comparable levels within the dynamic range were smaller or absent. Furthermore, the effect of level on discrimination was independent of electrode configuration in most cases even though the rate of spread of neural activation with level is expected to depend on electrode configuration. Possible interpretations of these results are that (1) the current level adjustments necessary to achieve comparable loudness for the various configurations significantly countered any effects of electrode configuration on the size of the activated neural population, or (2) the effects of level on discrimination do not result from its effects on the spatial extent of neural activation.