The Feynman story:

I decided to investigate. I started by counting seconds - without looking at a clock, of course - up to 60 in a slow, steady rhythm: 1, 2, 3, 4, 5.... When I got to 60, only 48 seconds had gone by, but that didn't bother me: the problem was not to count for exactly one minute, but to count at a standard rate. The next time I counted to 60, 49 seconds had passed. The next time, 48. Then 47, 48, 49, 48, 48.... So I found I could count at a pretty standard rate.

Now, if I just sat there, without counting, and waited until I thought a minute had gone by, it was very irregular - complete variations. So I found it's very poor to estimate a minute by sheer guessing. But by counting, I could get very accurate.

[...]

For instance, when I put out the laundry, I had to fill out a form saying how many shirts I had, how many pants, and so on. I found I could write down “3” in front of “pants” or “4” in front of “shirts,” but I couldn't count my socks. There were too many of them: I'm already using my “counting machine” - 36, 37, 38 - and here are all these socks in front of me - 39, 40, 41.... How do I count the socks?

I found I could arrange them in geometrical patterns - like a square, for example: a pair of socks in this comer, a pair in that one; a pair over here, and a pair over there - eight socks.

I continued this game of counting by patterns, and found I could count the lines in a newspaper article by grouping the lines into patterns of 3, 3, 3, and 1 to get 10; then 3 of those patterns, 3 of those patterns, 3 of those patterns, and 1 of those patterns made 100. I went right down the newspaper like that. After I had finished counting up to 60, I knew where I was in the patterns and could say, “I'm up to 60, and there are 113 lines.” I found that I could even read the articles while I counted to 60, and it didn't affect the rate! In fact, I could do anything while counting to myself - except talk out loud, of course.

[...]

The next morning, over breakfast, I reported the results of all these experiments to the other guys at the table. I told them all the things I could do while counting to myself, and said the only thing I absolutely could not do while counting to myself was talk.

One of the guys, a fella named John Tukey, said, “I don't believe you can read, and I don't see why you can't talk. I'll bet you I can talk while counting to myself, and I'll bet you you can't read.”

So I gave a demonstration: they gave me a book and I read it for a while, counting to myself. When I reached 60 I said, “Now!” - 48 seconds, my regular time. Then I told them what I had read.

Tukey was amazed. After we checked him a few times to see what his regular time was, he started talking: “Mary had a little lamb; I can say anything I want to, it doesn't make any difference; I don't know what's bothering you” - blah, blah, blah, and finally, “Okay!” He hit his time right on the nose! I couldn't believe it!

We talked about it a while, and we discovered something. It turned out that Tukey was counting in a different way: he was visualizing a tape with numbers on it going by. He would say, “Mary had a little lamb,” and he would watch it! Well, now it was clear: he's “looking” at his tape going by, so he can't read, and I'm “talking” to myself when I'm counting, so I can't speak!

-- source

(The rest of the story is interesting as well.)