CORRECTION

The authors of the article “Microglial Activity in People at Ultra High Risk of Psychosis and in Schizophrenia: An [¹¹C]PBR28 PET Brain Imaging Study” (doi: 10.1176/appi.ajp.2015.14101358), published in the January 2016 issue of the Journal, conducted new analyses of the data and discovered errors. The following corrections do not affect the main findings or conclusions of the article. In the abstract (3rd sentence of Method portion) and in the Method section (first sentence under “Participants”), it should state that 50 subjects completed the study, not 56. In Table 1, there were 12 subjects (85.71%) in the schizophrenia group, not 13, characterized as high-affinity binders. In Table 1, the positive, negative, general, and total symptom scores for the ultrahigh risk group were 10.7 (SD=4.3), 5.6 (SD=4.5), 17.1 (SD=12.6), and 44.4 (SD=23.5), respectively. The standard deviation for the total symptom score in the schizophrenia group should be 15.1, not 18.1. In Table 2, the values accompanying the means are standard errors of the mean (SEM), not standard deviations. This changes the effect size calculations, so the Cohen’s d values for the between-group analyses for total gray matter, frontal lobe, and temporal lobe were 1.358, 0.982, and 0.961, respectively, in the first cohort and 2.057, 0.596, and 1.269 in the second cohort. Similarly, the error bars in Figure 1 and the values accompanying the means in supplemental tables 1, 4, 5, 6, and 7 represent SEMs, not SDs. Supplemental tables 5 and 7 had values requiring updating. The tables are reproduced in their entirety here with the amended mean, SEM, and p values in bold italics.