Background: Errorless learning continues to be much debated in rehabilitation literature. Emerging data suggest that errorless learning is as effective as errorful learning when applied to the treatment of aphasic word-finding difficulties (Fillingham, Hodgson, Sage, & Lambon Ralph, 2003; Fillingham, Sage, & Lambon Ralph, in press; Fillingham, Sage, & Lambon Ralph, 2005).Aims: This paper presents a third investigation, which was designed to replicate this result and also to explore and extend other important and interesting findings from the previous empirical studies: (1) that withdrawing feedback during therapy (not giving information about whether a patient's response was correct or not) does not prevent learning; (2) that frontal executive skills are a predictor of therapy outcome but not language skill. We also used this third study to explore whether the number of naming attempts during therapy affects outcome.Methods & Procedures: Seven of the original eleven participants took part in a multiple baseline, crossover, case-series design.Outcomes & Results: The previous results were replicated: errorless and errorful therapy produced equivalent results immediately post-therapy and at follow-up. There was no effect of omitting feedback-the participants learned equally well without therapists' feedback. Also, executive/problem-solving skills and monitoring ability again predicted immediate naming improvements not language ability. In addition, we found that increasing the number of naming attempts during therapy affected learning outcome.Conclusions: The final section of the paper draws together the results of all three studies, and their implications for the treatment of aphasic word-finding difficulties are discussed.