Fig. 2.

This figure displays hypothetical results from a more carefully designed and analyzed cross-sectional study. (A) A whole brain analysis reveals an interaction effect with age (Age × Condition × Level) in the lateral prefrontal cortex, anterior cingulate and medial caudate. (B) An example of BOLD signal change plotted from one of these regions reveals similar activation in younger and older adults in condition R and significantly different activation between age groups in condition S. (C) A scatter plot displays a significant partial correlation (controlling for age) between neural signal (a difference score between levels in condition S) and behavior (a difference score in task performance between levels in condition S). (D) The individual data points (the difference scores in condition S used in the x-axis in the scatter plot) plotted here by age group reveal approximately normal distributions of the data in both younger and older adults groups. (E) In this hypothetical study, ROI masks were individually defined on single participants. These individualized masks are overlaid on a sample of participants’ T1 anatomy (axial; TT S = 10; MNI S = 12). The individual participant ages are listed below each image. It is clear from the images that these individual masks provide much better spatial accuracy than the single mask used in Figure 1. The data points for each of these four sample participants have been highlighted with red dots and labeled with the participant ages in panels (C) and (D). (F) A sample set of timecourses of activation from the ROI analysis visually confirm the age group by condition interaction. It is clear from these plots that the age differences are not due to group differences in the general shape, height or latency of the BOLD signal in this region of the brain.