With my sample size of 1, I can say that decoding and comprehension aren't necessarily mutually exclusive at the early stages of reading, particularly for children who rely on sight words.
When DS first began reading, he relied mostly on phonetic decomposition, clues from first letters of words, and context.
Now, as an "older" early reader, he blends phonetic decomposition and sight words. Because the average complexity of the words he reads is higher, more information has to be stored in his working memory at any time when decoding. However, with a larger bank of retained sight words, understanding of sight words is immediate. So there's a natural trade-off in the speed of a cycle of decoding to comprehension as he acquires a richer spoken and read vocabulary. I imagine this process, or one like it, is seen in kindergarten HP readers, with a discrete jump as the child first makes the transition from classroom easy readers. That's learning.
From what I see, my son will gladly accept a slower processing speed in the short term for the benefit of greater ease in reading later. He's intrinsically motivated by the challenge and seems thrilled when he reads more (for him) complex words like "stopped" or "hippocampus". Having the power to unlock new knowledge seems intoxicating to him.
The notion expressed by the author that you should have full comprehension of all nuances of a passage is preposterous. That is tantamount to saying that there should never be any learning from reading. Total hogwash.
