As always, polarbear makes good points. You'll note also that the outlier in the PRI is block design, which is the timed subtest in the cluster, suggesting that the PRI may actually be a low estimate of her ability under untimed conditions. It would not be surprising if the VCI were a low estimate due to her presentation of selective mutism, but she did respond on verbal tasks, which is not as characteristic. There is the possibility that there is a genuine expressive language disorder feeding into the mutism behaviors. Did the psych make additional comments about her verbalizations during the evaluation, that support the contention that mutism lowered the quality of her verbal responses?
Others who have mentioned percentiles, extended norms, and the GAI are also correct. Technically, there is no such thing as the 100th %ile, as that would mean that one's rank order within the norm group would be better than 100% of the population. This is why the upper extremes are reported as >99.9th %ile. The description of stopping at the ceiling and moving on may not mean that extended norms are not available. There are two different kinds of ceilings: test ceilings, which means that the test does not have enough range to capture the skills of this individual, and ceiling (or discontinue) rules, which define when the examiner is supposed to stop (usually after a certain number of incorrect responses). It is unclear which of these two ceilings is referenced in the comment you received. The maximum possible scaled score is 19, which means +3 SDs above the mean. A range of raw scores can result in this scaled score, some of which can be converted to higher extended scaled scores. It is usually recommended that extended norms be used when there are at least two 19s on the WISC-IV. The GAI is usually reported if it is more than about 4 points off of the FSIQ. And yes, the GAI can be calculated with or without extended norms. Yes, you can download the tables for doing it yourself, and, finally, yes, I would be happy to do it for you, with the scaled scores (for the standard GAI) or the raw scores (for the extended GAI).
The anxiety could be driving the slow processing speed, and the lower verbal/mutism, or slow processing/motor speed could be driving anxiety and mutism, or language processing delays could be driving anxiety and slow processing speed. I think you may need a more comprehensive evaluation, including social-emotional and speech/language evaluation, in order to tease out what the core deficit is.
