As an example, Dennett invites us to consider tiled wallpaper, displaying hundreds of identical Marilyn Monroe portraits. Upon seeing such a wall, you would immediately see all the images in reasonable detail. Yet, as Dennett points out, "you could foveate only one or two Marilyns in the time it takes you to jump to the conclusion and thereupon to see hundreds of identical Marilyns." (p.354) We could not possibly have discriminated them all, for our parafoveal vision is very poor. What we should see, if based only on the raw visual data, is one or two Marilyns "surrounded by various indistinct Marilyn-shaped blobs."
So how are we to explain this clarity of vision? The proponent of "filling in" might suggest that we make a great many mental 'photocopies' of the more detailed image, and plaster them all over our theatre walls. But Dennett suggests a much more economical alternative: your brain simply judges that the rest are the same, and "labels the whole region 'more Marilyns' without any further rendering of Marilyn at all." (p.355)
We can then apply this insight to understanding the blind spot:
The brain doesn't have to "fill in" for the blind spot, since the region in which the blind spot falls is already labeled (e.g., "plaid" or "Marilyns" or just "more of the same"). If the brain received contradictory evidence from some region, it would abandon or adjust its generalization, but not getting any evidence from the blind spot region is not the same as getting contradictory evidence. [...] [T]he brain has no precedent of getting information from that gap of the retina, [...] so when no information arrives from those sources, no one complains. The area is simply neglected. In other words, all normally sighted people "suffer" from a tiny bit of "anosognosia." We are unaware of our "deficit" - of the fact that we are receiving no visual information from our blind spots. [...] The fundamental flaw in the idea of "filling in" is that it suggests that the brain is providing something when in fact the brain is ignoring something. (p.355-6)