Hi, 

I need help on a lexical decision task.

For the non-word I got, each word has three non-words accordingly (add a letter, subtract a letter, and switch order). For example, apple can be applle, aple, alppe.  So now I have four word sets (original words, add a letter words, subtract a letter words, and switch order words).

I need the software to random select from the four word sets without repetition of the same word (For example, applle and aple can not be shown in the same experiment since they are from the same word.) I don't know how to do it. Can anyone help me please?

Thank you!

You first need to flesh this out more. I very much doubt you'd want the script to randomly pick one of the 4 sets for every stimulus. Certainly you want every subject to receive so and so many original words, so and so many add a letter words, so and so many subtract a letter words, etc.

I want the script to randomly pick one of the 4 sets for every stimulus. Then we can get for example 100 words. Then I want it to randomly pick 20 words out of these 100 words (as a secondary random selection).  Is it possible to do so?

You realize that this means that a person might actually only see stims of the same type, do you?

NO no no... 

What I mean is I want it to randomly select every word for each four same words first (for example, select applle from apple, applle, aple, alppe; select bear from bear, beear, ber, brae, etc....).

Then we can get 100 words with out any repetitions. (e.g., applle, bear, banaa, etc.)

Then I'd like to randomly select 20 words from these 100 words. 

Is it possible?

That's exactly what I'm talking about. It means a given subject might end up getting

applle, beear, etc.

*which are all of the same type*.

is it possible to get different type of words without repeating the same word? 

Anything's possible. Suppose you have 6 base stims, each existing in 4 different versions a to d. Then you do:

<block myblock>
/ trials = [1-3=mytrial]
</block>

<trial mytrial>
/ stimulusframes = [1=mytext]
/ validresponse = (57)
</trial>

<text mytext>
/ items = myitems
/ select = counter.baseword.selectedvalue+counter.version.selectedvalue
</text>

<item myitems>
/ 1 = "1a"
/ 2 = "1b"
/ 3 = "1c"
/ 4 = "1d"

/ 5 = "2a"
/ 6 = "2b"
/ 7 = "2c"
/ 8 = "2d"

/ 9 = "3a"
/ 10 = "3b"
/ 11 = "3c"
/ 12 = "3d"

/ 13 = "4a"
/ 14 = "4b"
/ 15 = "4c"
/ 16 = "4d"

/ 17 = "5a"
/ 18 = "5b"
/ 19 = "5c"
/ 20 = "5d"

/ 21 = "6a"
/ 22 = "6b"
/ 23 = "6c"
/ 24 = "6d"
</item>

<counter baseword>
/ items = (1,5,9,13,17,21)
/ select = noreplace
</counter>

<counter version>
/ items = (0,1,2,3)
/ select = replace
</counter>

Regards,

~Dave

Thanks so much!!! It works perfectly!