Dot-Probe task using E-Prime

[Sirous]

Hi I wonder if anybody can provide me with some information about how
to set up dot-probe task using E-Prime. many thanks for your help
Sirous

[liwenna]

Are your stimuli pictures or words? Do you need fixed stimulus pairs
(for instance when using two facial pictures of the same person in
each trial) or should stimulus pairs be randomised?
Do you need control over the contingency with which the probe appears
on the position of either of the stimulus (or should it just be 50% on
each type of stimulus?)

I'll get back to you on monday with more details when I am on my 'e-
prime' computer.. I just happened to have been working on dot probes
for the past few weeks. ;)

[S.Mo...@uea.ac.uk]

Hi Many thanks for your message. I am using the word pairs (e.g., shy and
rug), one presented above the other (top and bottom position) but I need
to counterbalance this, so for some subjects (50%) shy is the top word and
for some others (50%) it is the bottom word.
Altogether, I have 45 fixed pairs of words as my verbal stimuli, i.e. 15
pairs of POSITIVE words and Neutral (e.g., happy and pivot), 15 pairs of
NEGATIVE words and Neutral (e.g., insecure and tortoise), and 15 pairs of
BODY SENSATION and Non-body sensation words (e.g., Edgy and Dial).
I need to randomise the order of word pair presentations for each subject
so that each subject will receive the word pairs in a randomised order. I
also need to control the appearance of the probe (50% at the position of
the top word and 50% at the position of the bottom word).
Thank you very much for your help.
Sirous

[liwenna]

I sent a complete script to sirous, but for everybody else I post a
screenshot and some explanation here too... (I so often found threads
that should contain the answer to my question... but do not as people
continued in private e-mail).

Screenshot:
http://images.redial.net/awdpsirous.bmp

Explanation:
You want to counterbalance the position of the 'type of words', the
position of the probe and the identitiy of the probe, as well as
randomise the trial order.
To do all this we use two lists... the first list contains the
stimuluswords (stimword1 and stimword2) and for logging purposes I
also included a variable 'trialtype' so you can easily see in your
data which trial is which type... you might consider recoding these
into 1, 2, 3 right away (simply change that in the list or add another
attribute) so that for instance spss will later be able to easy
separate the different type of trials. You can add your remaining 42
word-combinations in here. Every trial will start with e-prime
choosing one of the word combinations, at random. If you need blocks
of the same trials (neg-neut, hap-neut etc), then we'll need an extra
list before this one, but I don't think this is the case, right?

Now, nested in this list is the 'positions' list (or positions and
probe id would be more accurately). It seemed most simple to just make
a list with all possible combinations of word-positions, 'probetype',
and probepositions in it. The position of the stimwords are given in
'posstim1' and 'posstim2', if you'll look at the properties of the
text-displays in the 'stimuli-slide' you'll see that for the y-
position of the frame it refers to these attributes: [posstim1] and
[posstim2]. This way the text-displays (with the stimwords which it
gets trough calling into [stimword1] and [stimword2] ) will be
displayed at y-positions 25% and 75% of the screen, ergo, top and
bottom. Then the probe, I used eiterh * or ** here, obviously you can
change this into any probes you like. The attribute 'cresp' refers to
the correct response... for my probes * and ** I will use buttons on a
SR-box, but in this version the two mousebuttons are you used...
obviously you can change all this in the stimuli-slide properties.
Lastly the attribute 'posprobe': this one is called upon by the probe-
texdisplay in the prob-slide in the same way as the stimulusdisplays
called into this list for their position, and it determines whether
the probe will be shown in the position of stimword1 or stimword2.

Regards,


liwenna

[seo...@ku.edu.tr]

Hello Anne-Wil,

First, thanks for your detailed explanation regarding the design of a sample dot-probe using e-prime.

I am going to use dot-prime, modify it with the pictures. I guess this is muh similar to Sirous. 

I have previous experience on e-prime but I need to develop it. If possible, could you please send me a sample of it?  
So that I can work on it further to modify it.

Thanks in advance,

Sezin

[Isa Zed]

Hi,

The screenshot doesn't work anymore, is it possible to put it back? I know it is an old topic but I am trying to do a similar thing in e-prime and I am stuck...

If it is also possible to obtain a complete script as Sirous got a few years back,it will be so perfect!

Thank you!

Isabella