EarlyFlank3 Specs
This is fairly simple, with only one factor (Trial Condition = Weird or Normal) that is established by random independent drawing as described below.
Basic Task: This is a Choice RT task with 2 possible stimuli and 2 possible button-push responses.
Fixation point appears. After 500 msec, fixation point disappears and two “flanker characters” appear (identical copies of the same letter to the left and right of the central fixation -- D in the figure below). After another 500 msec, the target letter appears (single letter in the center -- M in the figure below) appears in the center of the screen. The subject’s task is to press a button depending on the identity of the target. The target and flanker letters remain present until response. We are interested in both reaction time and errors.
The letters should be in 16 point font size. There should be one letter spacing between the left flanker and the target, and one letter spacing between the target and the right flanker.
When the subject responds, the square and the fixation point disappear and there is a blank screen shown for 2000 msec. If the subject makes an error, the first portion of this time is filled with a windows error sound. Try to time it so that whether or not there was an error sound played, the time between the response and the fixation point for the next trial is still always 2000 msec.
Note: Be sure you are always clearing the response buffer
before you present the fixation point, so that if subjects respond several
times, it doesn’t count a zero RT on the next trial. Try hitting keys multiple times and make sure the program behaves
appropriately (discarding all but the first hit!).
Task Mapping:
From the letters O, E, V, N
randomly assign two to be Target1 and Target2.
E.g., you might assign V to be Target1, and N to be Target2.
Target1 is assigned to the “Z” key on the keyboard.
Target2 is assigned to the “/” key on the keyboard.
The flankers are correlated with the flanker, in the sense that 90% of the time, the target is the same as the flanker. Those are “normal” trials. The other 10% of the trials are “weird” trials, where the flanker is the other target.
Pseudocode to implement the
appropriate mapping:
Take the vector (1,…,100) and
permute it. Refer to the randomized
list as P’. So P’(i) is the ith item on
the permuted list.
For trial = 1 to 100
Determine
trial based on P’(trial):
If
P’(trial) is between 1 and 45
Then
TrialCondition = Normal
Target
= Target1
Flanker =
Target1
If
P’(trial) is between 46 and 90
Then
TrialCondition = Normal
Target
= Target2
Flanker
= Target2
If
P’(trial) is between 91 and 95
Then
TrialCondition = Weird
Target
= Target1
Flanker
= Target2
If
P’(trial) is between 96 and 100
Then
TrialCondition = Weird
Target
= Target2
Flanker
= Target1
Note that this accomplishes the same
probability distribution, except that it insists that exactly 10% of the trials
within a block are weird, rather than letting it just be approximately 10%.
Blocks and Trials
On first page collect subject ID# and the usual (there is no subject condition in this study). The instructions need to be “personalized” based on the selection of Target1 and Target2 (you don’t need to tell the subject what flanker1 and flanker2 are).
During Block1 (but not the remaining blocks) put a big display up the whole time reminding them what they are doing, as follows:
Your Task M -> Left Key R -> Right Key R
M R
There should be 8 blocks each composed of 100 trials per block.
At the end of each block, put up a message saying “Rest Period – Click to Resume” and then continue when they mouse click take away that message and insert a 3 second pause before the first plus sign appears to give them time to put their fingers on the response keys. During the rest period show them their mean RT and number of errors for that block and each of the preceding blocks.
Instructions for subject:
Thank you for participating in our experiment. This experiment is composed of repeated “trials”, in which a letter is presented and you make a response. On every trial, a plus sign appears to mark the beginning of the trial. Please look directly at the plus sign.
Shortly after the plus sign appears, two “distractor letters” will appear to the left and right of fixation. You should just ignore these letters. A half-second later, a letter will appear in between the distractor letters, in the center of the screen. This is the letter you respond to.
Your task is simply to press one button if the central letter is ___ or the other button if the letter is ___. That is your whole job here today!
We are interested in your speed and accuracy. Try to respond as quickly and accurately as you can. The computer will beep at you every time you make an error.
We will have you do 8 blocks of 100 trials, each lasting just a few minutes. During each block, please keep the index and middle fingers of your right hand rested on the response keys.
The first block is a practice block, and during this block, there will be a display on the screen to remind you which letter goes with which response.
Between blocks there will be a pause for you to rest, and the computer will tell you your average response time and number of responses. Thanks again for participating, and we hope you find this task fairly interesting.
File Structure
In the R-file save the
<subject_ID#, block number, trial number, target letter; flanker letter; Trial Condition (“Weird” or “Common”); Response Time; Correct response; Identity of Actual Response; Correct vs. Error>