Certaines applications de lecture rapide font de grandes déclarations, mais à quelle vitesse les humains peuvent- ils lire?

Speed-reading applications often promise users the ability to dramatically boost their reading pace, sometimes even claiming speeds of over 1,000 words per minute. Yet, scientific research into human reading capacity challenges these assertions, suggesting that such gains come at the cost of comprehension. According to a 2019 meta-analysis of 190 studies, the average adult reads English non-fiction at a rate of 175 to 300 words per minute. Fiction, which is often considered more engaging, sees slightly higher rates, ranging from 200 to 320 words per minute. These figures represent the typical silent reading speed under standard conditions. However, proponents of speed-reading argue that these numbers are based on inefficient reading practices and that individuals can surpass them with the right techniques or tools. Human reading involves a complex interplay of eye movement and cognitive processing. During reading, the eyes make rapid jumps known as saccades, interspersed with brief pauses called fixations. Visual processing occurs primarily during these fixations, as the brain interprets the information presented within the limited area of focus. Each fixation typically lasts around 200 to 300 milliseconds, though this can vary depending on factors such as word frequency and complexity. Research indicates that readers typically perceive two to three letters to the left of their point of focus and up to seven or eight letters to the right. This perceptual span differs across languages and writing systems. Despite this, speed-reading literature frequently asserts that individuals can learn to process 10 to 14 words in a single fixation. Such claims contradict empirical findings, as studies suggest that attempting to extract more information per fixation leads to reduced comprehension rather than improved efficiency. Several speed-reading technologies attempt to enhance reading speed by altering how text is displayed. Apps like Bionic Reading use bold formatting for the initial part of each word to aid in quick recognition. Similarly, BeeLine Reader employs color-coded lines to guide the reader's eye from one line to the next. Spritz, another application, presents each word in the same location on the screen, eliminating the need for eye movement and supposedly enabling faster reading. Despite these innovations, research has consistently shown minimal to no improvement in reading speed or comprehension when using these methods. A 2024 study comparing users of Bionic Reading with traditional reading methods found no significant differences in either reading rate or understanding. Likewise, a 2022 study involving children revealed that while BeeLine Reader might assist some young readers in navigating dense text more quickly, it did not improve comprehension for simpler material. These findings underscore the limitations of current speed-reading approaches and highlight the importance of maintaining comprehension alongside reading speed. While certain strategies may offer marginal benefits, they do not fundamentally alter the physiological and cognitive processes underlying reading. As such, the effectiveness of speed-reading apps remains questionable, and their impact on actual reading performance is largely overstated.