I understand the concept of regression to the mean. However, does anyone know if there is a limit to how much of a drop in a WISC score can be attributed to regression to the mean?

My DS had the WISC six years apart (ages 7.5 & 13.5). The first time both his full scale and GAI were 133. Now, his Full Scale dropped to 120 and his GAI dropped to 116. The psychologist who did the second exam was quick to attribute the drop in the score to regression to the mean and considers the second test valid.

His processing speed index dropped 32 percentile points - from the 90th percentile to the 58th. His VCI dropped from 133 to 113.