Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
displaying one specific space as masked in self-paced reading?
#1
Hi!
I'm designing a self-paced reading experiment (tokenized text, masked, forward_mask: true). For one word 'he' I would like to have the mask be 3 characters (xxx) instead of two, to make 'he' indistinguishable from 'she' when they are masked. Is there a way to do this while keeping the general shape of the words/sentence visible (so not masking all spaces)?
Thanks!
Reply
#2
Hi atfrederiksen,

I am checking on this for you, and I will get back to you soon.

Best,
Rachel
Reply
#3
Hi atfrederiksen,

You could do this by using "+" in place of normal spaces and "+" as the delimiter for the sentence containing the word in which you'd like the mask to be padded ('he'). So, for example, you could define this stimulus like so:

"ssr_1": {
  "type": "tokenized_text",
  "content": "He +is+very+happy",
  "delimiter": "+",
  "self_paced": true,
  "mode": "masked",
  "forward_mask": true
}

For the other stimulus, you could either define like so:

"ssr_2": {
  "type": "tokenized_text",
  "content": "She+is+very+happy",
  "delimiter": "+",
  "self_paced": true,
  "mode": "masked",
  "forward_mask": true
}
 
Or as you would typically define it:
 
"ssr_2": {
  "type": "tokenized_text",
  "content": "She is very happy",
  "self_paced": true,
  "mode": "masked",
  "forward_mask": true
}

However, these sentences will show up as single strings of (equal length) masked characters. It's not possible to accomplish what you are asking with individually masked elements. 

I hope that helps!

Best,
Rachel
Reply
#4
(04-14-2022, 02:08 PM)Hi Rachel,Thanks for your help.It looks like this accomplishes more or less the same as setting masked_spaces to TRUE, then (except with one more character)?This doesn't solve my problem, but it looks like it is not currently solvable. Thanks again!Rachel - Participant Affairs Wrote: Hi atfrederiksen,

You could do this by using "+" in place of normal spaces and "+" as the delimiter for the sentence containing the word in which you'd like the mask to be padded ('he'). So, for example, you could define this stimulus like so:

"ssr_1": {
  "type": "tokenized_text",
  "content": "He +is+very+happy",
  "delimiter": "+",
  "self_paced": true,
  "mode": "masked",
  "forward_mask": true
}

For the other stimulus, you could either define like so:

"ssr_2": {
  "type": "tokenized_text",
  "content": "She+is+very+happy",
  "delimiter": "+",
  "self_paced": true,
  "mode": "masked",
  "forward_mask": true
}
 
Or as you would typically define it:
 
"ssr_2": {
  "type": "tokenized_text",
  "content": "She is very happy",
  "self_paced": true,
  "mode": "masked",
  "forward_mask": true
}

However, these sentences will show up as single strings of (equal length) masked characters. It's not possible to accomplish what you are asking with individually masked elements. 

I hope that helps!

Best,
Rachel
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)