But children with CDD and other forms of regressive autism don't really grow backwards. If someone was growing backwards, then they'd act like they did at a younger age. When you have a hyperverbal, sociable 2 year old growing into a shy, somewhat quirky 4 year old who then becomes an aloof, obsessive, echolalic 6 year old, that 6 year old is clearly not like he was at any age before 4. Before 2 years, he was a sociable nonverbal child who couldn't walk becoming a sociable walking child who used single words becoming a sociable running child who used communicative sentences. He was never echolalic, obsessive or aloof.
This is part of the mental age fallacy. An autistic child at any age is not the same as a same-age, younger or older child. And an autistic child who regressed is not like they were at a younger age.
So far, I've just been looking at outside appearances. If you look at what is going on inside the child, the analogy of growing backward fits even less.
Amanda Baggs described 'regression' like this:
"it’s your entire brain shifting around, focusing on some things that a lot of people find unimportant or bad, failing to focus on some things that a lot of people find important or good. Imagine that the things it is focusing on are exactly what you need to be focusing on. You are becoming the sort of person you need to be."
This sounds pretty much like a lot of the kind of changes that normally go with growing up. Not growing backward.
She also talks about how someone with brain damage can change afterward, which is another way you could view some cases of autistic regression. In that case, you're certainly not growing backward, and if your behaviour post-brain injury gets viewed at all as like a younger version of yourself, it's purely coincidence. Saying a person with brain damage is 'growing younger' because they have lost skills they previously had is like saying a 4 year old who went deaf and then forgot how to talk has 'grown younger' because xyr communication skills superficially resemble a younger child. The kid's expressive language skills may be said to be at a 1 year old level, but firstly, their receptive language skills are poorer than a 1 year old while their motor and cognitive skills are much better, and secondly, when that kid was 1, xe could hear. Xe couldn't speak for very different reasons than now.
Here's an analogy I thought of awhile back:
Three children are traveling along three paths, that appear to be pretty close to identical. They are traveling at the same rate.
One kid is normal - xyr path may be straight or winding, but other people view that as the expected path for a child to take.
The second child seems to be heading along that path, but then xyr path turns when the other one doesn't, or doesn't turn when the other one does, and you realize xe was actually heading along a different path all along. This is what autistic regression usually is like.
The third child really is heading down the usual path, but then xe gets jostled onto a different path and start walking along it instead. This is more like brain damage.
None of those children ever went backward. They never retraced a part of the path they'd seen before. Instead, they all kept going forward, but the path ahead diverges from the other two children in various ways.