Mipellssed Wdors

Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a total mses and you can sitll raed it wouthit a porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe. Amzanig huh?


  1. Jeremy Flint says:

    Kind of the same way that you can read most sentences faster by only looking at the top half of the words….or something like that.
    I think my Type professor told me that when i was in school.
    The big question is how long did it take you to write that senctence.

  2. ACJ says:

    Heh. I received a similar message in my mail box some time ago.
    “wouthit a porbelm”
    Little problem, true… no problem at all, that I don’t know. Then again, I’m dyslexic.
    Seriously though, it really slows down the reading — for me.

  3. mike says:

    pertty amzaing…!

  4. Darice says:

    I’m with what Jeremy said, that’s why I have lots of misspelled words after writing up something. But it is amazing how you can read and understand a whole paragraph of misspelled words.

  5. Josh S. says:

    This has already been debunked by lingustic professors, maybe someone has the link.
    It doesn’t seem to be very reliable, especially since it doesn’t even seem to be from Cambridge. Check this article out at Cambridge. It has even been labeled an urban legend by some.
    If anyone has the link where the linguistic guys made (was it Harvard?) a response in the same format but very hard to read, it would be appreciated if you could post it. Thanks!

  6. stew says:

    The brain actually recognises patterns rather than a “word” its more a collection of certain letters, so what we actually see as say “hello” our mind simply sees the letter group in a particular way and make s a descision as to if its hello or something else using a similar word pattern.
    I cant rememebr exactly but I’ll check with my typography :)

  7. James Craig says:

    Yeah, it requires a bit of context in the misspelling. Certain two-letter phonemes like /th/ and /ng/ would definately not make sense if they were more than a letter or so apart. If the misspellings were truly random, we’d have a much harder time with it.

  8. Egor Kloos says:

    As far as I can see this does show how humans, to a certain extent, read patterns and not correctly spelled sentences and words. In fact this doesn’t only work in English, I’ve seen this before in Dutch and it works like a charm. In fact Josh S’s Cambridge link shows even more languages. And also doesn’t dispel the fact that reading is a letter by letter affair.
    I would find it unbelievable that human language would rely entirely on correct spelling when every other human ability is dependant on some form of complex pattern and or pulse recognition. So although this is a bit pseudo science it isn’t really a lie either. Okay the bit about Cambridge is apparently a bit of a fib.
    Still very cool though :)

  9. Keith says:

    Hloy cow! Taht’s pterty amziang.
    I gesus I can sotp wroring auobt my selpling mtiskes now.
    Ok, that is hard to write…fun though.

  10. monkeyinabox says:

    Ouch! My head is hurting!!!!

  11. Anonymous says:

    To me is no urban legend… Its amazing, but I can easly read that paragraph..

  12. Nick Finck says:

    The text could be reduced even more… simply crop off the text a little bit above the base line, but no more than half of the full height of text. Check it out:

  13. Jenn says:

    I just assumed I could read spelling mistakes because I always make spelling mistakes. :)

  14. Rene says:

    I guess what we can all take away from this is that it will be next to impossible to pick out the spelling errors when proofreading a hardcopy of something :) I suddenly have a whole new appreciation for the editors out there.

  15. David House says:

    Not to sound cocky, but I got that in my inbox a looong time ago ;P

  16. Tanny O'Halely says:

    I’m with ACL, it really slows down my ability to read the paragraph.

  17. Beth says:

    Yes, the last time this went around, an interesting exercise I undertook was to sort the words alphabetically, rather than in sentence order. Kind of interesting to see how less readable the majority of words become then.

  18. Skardhamar says:

    Welcome to yesterday :)

  19. Peter says:

    Sadly, this finding has lately been exploited by spammers to get pass spam filters.

  20. Jeremy says:

    That may be true, but I look down upon all the AOL kiddies, who still use the oddest spellings for everyone. ;)

  21. Stephen says:

    Hey! This idea has been driving me nuts all day. I read it this morning but only had a chance to code in the last two hours…
    So I made this script that automatically misspells webpages and the results don’t look too legible. :( Not even google is legible at times (it is random, so your results will vary).

  22. Kitta says:

    That is pretty cool.

  23. cafe fort says:

    That’s really fine ! But, that information isn’t so accurate. Sorry Dan, The story, IMHO, hapened in the automn 2003 ;-) See you. :)

  24. andreas says:

    it is the context an what you know about something: The last of tsehe trehe ofcfies ehbastelisd was that of the Ahcron, the inoiuttitsn of wihch is daetd by a mjarioty of aortuhtiies in the time of Moedn, toughh smoe put it in taht of Atsucas, adicudng in enivdcee the fact that the Nnie Ahcrons saewr taht they wlil prrofem thier ohats eevn as in the tmie of Atsacus, sniowhg that in his time the hsoue of Courds rreetid from the Khgnsiip in rretun for the peeivglris bosweetd on the Ahcron. Whechievr of the two acocnuts is ture, it wuold mkae vrey llitte dffenercie in the dteas; but that tihs was the lsat of tsehe oceiffs to be iesuttintd is aslo iecadntid by the fact taht the Acorhn deos not amistdienr any of the artcasenl reits, as do the Knig and the War-lrod, but meelry the dtuies added ltear; on acncuot of wihch also the Aochnirhsp olny bcmaee great in reenct temis, when aeungemtd by the adedd duties.
    This is a text from Aristotle, and I am pretty sure, must of us have a hard time reading it

  25. doug says:

    It’s interesting how this phenomenon follows the political ‘sound byte’ model – at first glance it seems to make sense, but upon closer scrutiny it’s just a load of carp :)

  26. Stephen says:

    I don’t know doug, I can still read the above paragraph so there is definately something going on. We just don’t know what yet. :)

  27. Fredrik says:

    Funny, that it’s pretty old news. Most Typographers and web designers know that we don’t read per letter, but a word-image. We form the letters into words, and the words to images. That’s why w0rd would look more disturbing than wrod.
    The human memory works in the same way. It’s almost impossible for us to memorize a group of letters, or a series of digits. But if we can form these into words or an image it is usually a lot easier…

  28. Sam Walker says:

    While it’s certainly easy to read, it is by no means just as easy as regularly spelled words. I don’t know what those “Cambridge researches” were smoking..

  29. Kronn says:

    I totally disagree. It may be possible to read, but this is just because its simple text. Try this with a complex text or try to shuffle the letters across syllable borders. It will be very hard to read.
    Nice idea, though.

  30. Messing middle letters is okay in English.
    Now try it in French, it won’t work so well, mainly because of (disturbing) accents.
    Halés, les trmees dveninent ililslbes !

  31. mjr says:

    This research did actually take place, although it was undertaken at my alma mater Edinburgh University and not Cambridge. This article from the Student newspaper has a little more detail.

  32. thomas says:

    I dont know, I went tried #21′s (Stephen) script on cnn.com and I could read it pretty well. Im guessing that the people at Cambridge put together that snippet so that it looked dissheveled but was still legibile, in the same way that you could put together a paragraph that looked equally messed-up, but was a lot less legible. The random script proves it pretty well, though – just go to some text-rich page and try to read it!
    Nice coding, Stephen!

  33. andy says:

    Humans read via pattern and shape, this is the same reason why words in all caps take so long to read. When a word is randomized but the high characters and low characters are fairly close to the same place it is easy enough to read, but if that displacement becomes too large it becomes very difficult even if the first and last letters are the same. The addition of the first/last letters rule makes it much easier to read, but it is not the true reason why we are able to read it so easily.
    Some examples of this phenomenon would be moving the tall letters all to one side:
    Adoccrnig to a rhsceearch at Cdbmarige Utinervisy
    Your eyes sort of break the words up into sectiosn based on teh tall sections. As you scan the first word of that phrase if you see a “d” around the middle you assume it is “according” and move on. With the “d” at the beginning it isn’t until you reach the end of the word that you can decide what it is, as you have seen no familiar sections, so you are forced to go back and re-evaluate once you have already read it once. The first/last allows you to fairly simply discern the word at that point, but reading every word twice will slow you down significantly.

  34. Gambit says:

    Since I was a kid I could never spell. Something about the way we pronounce words and then right them differently to the way they sound.
    At leased now I know that if I just manage to get the first and the last right people will be able to understand me. Cool.

  35. vlado says:

    There was a follow-up that debunked the theory
    Anidroccg to crad cniyrrag lcitsiugnis planoissefors at an uemannd,
    utisreviny in Bsitirh Cibmuloa, and crartnoy to the duoibus cmials of
    the ueticnd rcraeseh, a slpmie, macinahcel ioisrevnn of ianretnl
    cretcarahs araepps sneiciffut to csufnoe the eadyrevy oekoolnr.

    Like andreas’ example, it’s possible to read, but no way near as easy as the original example of scrambled words.

  36. leticia says:

    you know. the scary thing is that i read that faster than i read normallly…. hmmm what’s that mean?

  37. Stephen says:

    Haha vlado!!! Just what I was waiting for! :D

  38. SRC says:

    wow yuo did a gpod jbo terhe!

  39. Michaël Guitton says:

    Leçon de lecture
    Here’s the same hoax in French
    Sleon une édtue de l’Uvinertisé de Cmabrigde, l’odrre des ltteers dnas un mtos n’a pas d’ipmrotncae, la suele coshe ipmrotnate est que la pmeirère et la drenèire soit à la bnnoe pclae. Le rsete peut êrte dnas un dsérorde ttoal et vuos puoevz tujoruos lrie snas porlblème. C’est prace que le creaveu hmauin ne lit pas chuaqe ltetre elle-mmêe, mias le mot cmome un tuot.
    Etonnant non ?

  40. Vegard says:

    Heh… excellent!

  41. Jesper says:

    Wsna’t taht the ogrinial emaxlpe thugoh?

  42. Chris McDougall says:

    Note that this phenomenon shouldn’t excuse poor spelling. Take, for example, Keith’s sentence:
    “I gesus I can sotp wroring auobt my selpling mtiskes now.”
    The words “worrying” and “mistakes” were hard for me to read because he’s actually missing letters. So words can be jumbled up, but they have to have all the appropriate letters in order to work. I’m not trying to single Keith out. I’m just pointing out something relevant. His sentence just happened to be a good example.
    As for the responses that contest the phenomenon, I’d have to agree with them. Certain combinations are very difficult. Though it is cool that such “readable” paragraphs can be made in the first place.

  43. aaron wall says:

    As long as you know most of the concepts and words which are supposed to be represented and they are in a somewhat logical order it works, but confusing high level literature becomes unbearable when juxtaposed. Its still cool though :)
    People still do take the time to tell me that my websites suck for poor spelling…perhaps they only notice because they are mean people?

  44. I agree with Peter in that I, too, have noticed more spam using this concept to avoid spam filters.

  45. Eugene says:

    Is this still going on? I’m almost positive that this email has gone round the weblog circles last year. If I remember correctly, some weblogger made an experiment in how fast this meme spreads by putting different versions of that misspelled paragraph in his weblog.

  46. Steve says:

    I teach speed reading courses for a brain based accelerated learning company called Quantum Learning. What I have learned in my research and direct application with regularly skilled readers is this: (1) our eyes look for meaning–this includes when we read; (2) meaning comes from our eyes recognizing shapes and patterns; (3) the more we read, the more shape and pattern recognition skill we gain. Therefore, this activity would not be as effective with beginning readers or slower skilled readers. Also, people with more adept visual skills will be more successful and quicker to be able to recognize the reordered words. The ability to decipher the reordered words will wholly depend on the brain and eyes looking at them.

  47. NeSt says:

    I tinhk taht is amzanig!

  48. Russell says:

    Sounds like bulhsit to me

  49. renaldo says:

    we have become accustomed to race through our daily routine to accomplish more. reading images rather then actual words, is an example of such