The objective of the study was to evaluate the effects of metacognitive skills training (MST) and behavioural practice on error self-regulation during a naturalistic task after traumatic brain injury (TBI). A single-case study design was used and three participants (two males, one female) aged 26-43 years with severe TBI were included in the study. In the first study, after a four-session baseline of behavioural practice, two participants received eight MST sessions followed by four maintenance sessions. In the second study, a third participant received 16 sessions of behavioural practice to assess the extent to which error self-regulation improves through long-term task practice and therapist corrections. Participants prepared two different meals with a novel meal introduced later to examine skills generalisation. Behavioural outcomes included error frequency, checking and self-corrections. Data analysis involved a combination of visual analysis and two standard deviation (2-SD) band analysis. In the MST study, the two participants demonstrated a 38% and 76% reduction in error frequency (p < .05), a significant decrease in checks (p < .05), and a significant increase in self-corrections (p < .05) relative to baseline. In the behavioural practice study, the participant demonstrated reduced errors (25%), although this was not statistically significant, and a significant increase in checks (p < .05), but self-corrections did not significantly change (p > .05). This exploratory research suggests that, firstly, by targeting error self-regulation MST can potentially promote independence on complex everyday tasks; and secondly, although behavioural practice alone may facilitate some functional gains, it fails to promote more independent self-regulatory behaviours.