Since roughly 1960, the typographical style has shifted from two spaces after a period to one space after a period. Some people have claimed that the old convention of two spaces can improve speed reading.

Found in comments for the same article:

I am a speed reader (trained and practicing).

Comprehension being equal, I can read faster with double spacing between sentences. It is a tested and quantifiable fact, and this is generally true for the majority of speed readers.

And also here:

[Speed readers] read in word "chunks" or "grabs", but doesn't that actually suggest that double-spacing after periods improves readability? Most people absorb maybe two to three, perhaps up to five words per "chunk", not two or more sentences per chunk. Having sentences separated by spaces would make it easier to "grab" the words in related groups according to sentence structure.

I also remember reading a claim on Facebook that "studies" show that double spaces improve speed reading, but I can't find that claim any more. The first quote above seems to suggest the same, though (talking about the "quantifiable fact").

Are there any studies which actually show this claim to be true?


It has been brought up in comments below that "speed reading" is not a well defined term. For the purpose of this question, I'm willing to entertain any reasonable definition of the term, even possibly including "self-identified speed readers." But I do expect an answer to specifically address "speed readers", since the claim clearly makes the distinction between "speed readers" and the implied "average" or "non-speed reader".

  • 4
    The answer might be different depending on whether you're using a monospaced or a proportional font, so there may be two valid studies with opposite results because of that.
    – Rob Watts
    Commented Jan 20, 2015 at 16:53
  • 1
    @RobWatts: Perhaps, although as demonstrated in the first article I linked, the double space convention has existed for centuries in proportional spaced typesetting. Mono-spaced type didn't come about as a real thing until the typewriter, and even then it wasn't seen except in manuscripts, letters, and other non-typeset formats. So I'm primarily interested in a study about proportional (typset) print. But if there is another study about monospace type, that would of course be relevant and interesting as well.
    – Flimzy
    Commented Jan 20, 2015 at 16:57
  • 2
    The point is that in proportion fonts the space after a full stop might be larger but this is not a double space. Double space only make sense in a monospaced font
    – mmmmmm
    Commented Jan 20, 2015 at 19:08
  • @Mark: Indeed, the proper term is "emspace", which is distinct from two normal spaces. But that doesn't change the nature of the question.
    – Flimzy
    Commented Jan 20, 2015 at 20:22
  • 1
    This seems to be confusing the implementation details (e.g. hitting the space bar twice on a typewriter) with the result (English versus French spacing, i.e. en-space versus em-space.) In some implementations (HTML and LaTeX, from memory), the number of space characters is irrelevant to the type-setting.
    – Oddthinking
    Commented Jan 21, 2015 at 0:48

1 Answer 1


Though not addressing any sort of "speed" reading, a review of literature suggests no statistically significant difference.

Loh et al. studied 66 college students in 2002. They found the mean times to read passages with two spaces after each period was longer than reading passages with one space, but with their small sample size were unable to conclude statistical significance.

We have concluded that there was not enough evidence to suggest that a significance difference exists in an on-screen reading task, between text formatting using single space or double space in sentence separation.

Another study by Ni et al. in 2009 found:

The results showed that there is insufficient evidence to support the claim that a significant performance difference exists, in an on-screen reading task, between text formatting using single space, double space, or triple space in sentence separation.

  • "Loh et al" 2002 ≠ "Ni et al." 2009.
    – jmabs
    Commented Jan 21, 2015 at 3:50
  • Well then contact the 40 year old International Journal of Instructional Media and tell them they published a peer-reviewed paper that is plagiarized. Good luck!
    – jmabs
    Commented Jan 21, 2015 at 3:57
  • 1
    Neither of these studies address speed reading, which is essential to address the claim.
    – Flimzy
    Commented Jan 21, 2015 at 3:58
  • @Flimzy the abstracts mention performance but not explicitly speed reading, maybe it's the same but one needs to read the article. jferr, could you clarify which is the case in the answer please (the answer is fine no matter what the articles say, but without a clarification it's hard for many of our readers to judge your post).
    – Sklivvz
    Commented Jan 21, 2015 at 10:18
  • 5
    Flimzy is correct in that neither article addresses "speed" reading. You will need to define this term further if you would like a more specific answer. I'm not aware of any academic definition for it nor method of measuring whether a person looking at words is "speed" reading or just "regular" reading.
    – jmabs
    Commented Jan 21, 2015 at 18:45

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .