I’ve noticed a lot of posts mentioning gigantic increases in reading speed; one of them mentioned a claim of 25,000 WPM. I think if anyone had discovered a way to do this, it would have been reported in the literature of experimental psychology journals.
However, if you take a look, almost every study indicates that comprehension rapidly goes down as you increase your speed, even beyond 250 WPM.
I’d highly recommend to anyone interested in the topic to look at the research by Ronald P. Carver (you can find his 1997 paper “Reading for One Second, One Minute, or One Year From the Perspective of Rauding Theory” freely published online). He breaks down the actual processes involved when we do scanning, skimming, “rauding”, learning and memorising, and places some estimates on the upper limit of reading speed (assuming you want to comprehend what you read). According to his models, scanning is effective up to about 600 WPM, but reading for comprehension and learning is between 200-300 WPM.
Ultimately, we’re limited by the time it takes to recognise words and form the semantic links to the concepts they represent, and we’re severely limited by the amount of information that can fit into our short term memories, as well as the amount that can be encoded and stored in long-term memory.
I’d urge you to take claims about reading effectively at speeds in the thousands of WPM with a healthy dose of skepticism. As a kid, I was hooked on Tony Buzan’s books and really wanted speed reading to work, but it just didn’t live up to the promise.