You can download the audio and PDFs from the 2007 Cognitive Aging Summit in Washington DC here; they're good listening.  But I want to draw your attention to the graphs on page 6 of Archana Singh-Manoux's presentation.  It shows the "social gradient" of intelligence.  The X-axis is decreasing socioeconomic status (SES); the Y-axis is increasing performance on tests of reasoning, memory, phonemic fluency, and vocabulary.  Each graph shows a line sloping from the upper left (high SES, high performance) downwards and to the right.

Does anything leap out at you as strange about these graphs?