The GAI likely is technically correct, it just might not be the best representation of that individual's true range. In my experience, (backed up by the tables calculating predicted achievement from only one index score (e.g., VCI or PRI/VSI/FRI)) when an index score is significantly higher than the others, it is a better predictor of that individual's academic achievement in strength areas than composites (FSIQ or GAI) that combine cognitive strengths and weaknesses. This, of course, is why the aforementioned tables exist.
