Introduction: Laws, Kondel, and McKenna (1999) previously reported a case study analysis of a schizophrenic patient (TC) with severe formal thought disorder (FTD). Examining consistency across item and modality of input, Laws et al. documented an impairment of access to semantic knowledge in TC. Method: Following substantial improvement in his FTD, we readministered the same extensive battery of neuropsychological tests tapping semantic memory functioning. Results: Whilst TC's naming remained relatively good, it also became more consistent across both time and modality. Tasks tapping language comprehension and understanding of semantic association revealed some significant improvements. Nevertheless, TC showed a residual propensity to verify false information. Conclusion: Improvement in FTD in schizophrenia was accompanied by a better and more stable semantic memory performance in TC. The findings are consistent with, and expand upon the original suggestion that thought disorder reflects disorganised access to semantic memory.