Specialized early intervention (EI) following a first episode of psychosis (FEP) are effective at reducing negative symptoms, although its trajectory warrants systematic assessment. However, findings are equivocal as to whether extended gains are made post 2 years of EI and whether there is additional benefit of extending EI for an additional 3 years. Data on 178 FEP patients, from a randomized controlled trial of a 3-year extension of EI service v. transfer to regular care following 2 years of EI service, were used for this report. Repeated measures analysis of variance were conducted separately for the initial 2 years of treatment in an EI service, and for the 3-year post-randomization to examine trajectories of negative symptoms over the two periods in the two arms of the study. There were significant improvements in total negative symptoms over the first 2 years of EI F(4.612, 797.905) = 25.263, p < 0.001 and in domains of 'expressivity' and 'motivation'. In the following 3 years, there were further significant improvements in negative symptoms F(4.318, 759.908) = 4.182, p = 0.002 with no difference between groups F(4.318, 759.908) = 1.073, p = 0.371. Changes in negative symptoms over the extension period were driven by expressivity F(4.01, 674.73) = 7.19, p < 0.01, but not motivation F(6.58, 1112.18) = 0.95, p = 0.46. Negative symptoms improve significantly over the first 2 years of EI. Subsequent amelioration was largely the result of expressivity. Motivation deficits remained stable. Extended EI offered no advantage over regular care post-randomization.