This study aims to compare the distinct effects of a steady-state (SSN) and a fluctuating (ICRA) masker on speech reception performance. SNR, reverberation and masker type were combined as to create several acoustic scenarios; matrixed-word listening tests in the Italian language were proposed to a panel of young adults with normal hearing, collecting data on intelligibility scores (IS) and response time (RT). The listening conditions were objectively qualified with the short-term metric STIr, defined as the average of the STI values calculated over short time-windows, whose duration reflects the typical phoneme length. The results showed that for a given STIr, both maskers yield the same IS, being the fluctuation benefit already accounted for by the objective metric. The slope of the STIr-IS function only depends on the speech material. Anyway, the fluctuating masker calls for an increased amount of cognitive resources to be deployed in the speech reception process, traced by a statistically significant higher response time. These results shade a new light on the fluctuating masker release (FMR) phenomenon.