Table 2:

Results of the individual linear regression models predicting cognition for each individual MRI measure

SemanticEpisodicExecutiveSpatial
GM6.14 (1.90)P = .001*10.87 (1.92) P < .001*6.95 (1.46)P < .001*7.65 (2.07)P < .001*
WMH−.06 (0.05)P = .27−.16 (.06)P = .0068−.09 (.04)P = .04−.08 (.06)P = .17
FA-CC.91 (.86)P = .301.82 (.92)P = .051.24 (.68)P = .071.70 (.93)P = .07
FA-Cingulum.11 (.58)P = .851.80 (.62)P = .004*1.32 (.46)P = .005*1.28 (.61)P = .04
FA-Fornix.83 (.70)P = .242.78 (.75)P < .001*1.71 (.57)P < .003*2.10 (.74)P < .005*
FA-SLF.27 (.82)P = .74.67 (.89)P = .45.69 (.67)P = .301.18 (.88)P = .18
FA-UNC1.45 (.80)P = .071.90 (.86)P = .031.49 (.64)P = .021.98 (.85)P = .02
  • Note:—Estimated from separate linear regression models that controlled for gender, education, and recruitment source. The α level was set at .01 to account for the 4 outcome measures (.05/4 = .0125). FA indicates ratio of young adults fractional anistropy value; GM, percentage of total cranial volume; WMH, white matter hyperintensities (log-transformed percentage of total cranial volume).

  • * Indicates P values not larger than the α level which was set at .0125.