Objective:In the present study, we investigated whether varying the task difficulty of the Sentence-Final Word Identification and Recall (SWIR) Test has an effect on the benefit of noise reduction, as well as whether task difficulty predictability affects recall. The relationship between working memory and recall was examined. Design:Task difficulty was manipulated by varying the list length with noise reduction on and off in competing speech and speech-shaped noise. Half of the participants were informed about list length in advance. Working memory capacity was measured using the Reading Span. Study sample:Thirty-two experienced hearing aid users with moderate sensorineural hearing loss. Results:Task difficulty did not affect the noise reduction benefit and task difficulty predictability did not affect recall. Participants may have employed a different recall strategy when task difficulty was unpredictable and noise reduction off. Reading Span scores positively correlated with the SWIR test. Noise reduction improved recall in competing speech. Conclusions:The SWIR test with varying list length is suitable for detecting the benefit of noise reduction. The correlation with working memory suggests that the SWIR test could be modified to be adaptive to individual cognitive capacity. The results on noise and noise reduction replicate previous findings.