Abstract In order to verify the Feynman-alpha formula with stationary multiple emission sources [Pázsit, I., Yamane, Y., 1998. Nuclear Instruments and Methods A403, 431–441], a series of experiments were performed by using three different neutron sources, i.e., an Am–Be, a 252Cf, and a randomly triggered D–T pulsed neutron generator. As a result, it was found that the prompt neutron decay constant evaluated from the measured Y-function was hardly affected by the difference of neutron sources that have respective source multiplicities. On the other hand, it was demonstrated that the amplitude of Y-function, i.e., the correlation amplitude, depended on both the subcriticality and the neutron source multiplicity. These experimental facts agreed with the theoretical prospects that were deduced from the Feynman-alpha formula with multiple emission sources. Therefore, it was concluded that the validity of this formula was experimentally shown.