Millisecond Online Community

Hello,

First I would like to congratulate on Inquisit which makes my life so much easier!

Now I am working on an experiment using a dual task paradigm. The primary task is product evaluation using a Likert scale, the secondary test is a probe reaction task (the subject has to say something as fast as possible when she hears a beep on the headset). I managed to program the two parts separately, but I do not know if it is possible to put the the parts together and run them at the same time. That is why I need your help.

Thanks for the answer,

Agnes Buvar

 

Hi Agnes,

implementing dual task paradigms can be really tricky. In order to possibly help you, you will have to be way more specific with the details of your experimental design (and maybe even make the things you already have programmed available online). How exactly do you want the two tasks to work together? If you could write up a short experiment description (as you would for a journal article's method section), that would really be helpful!

Hope to hear more from you,
~Dave

Hello,

 

I have sent the two programs at inquisit@millisecond.com.

Briefly, the primary task is to evaluate (global assessment like/do not like) textile samples. The samples are presented by the experiment leader (we cannot use virtual presentation because of the stimuli’s nature). We would like to measure the time that the subject spends on the observation of the sample. In order to do this, the subject has to press the space bar at the same time as the experiment leader shows the sample to him. Then, the Likert scale appears on the screen and the subject gives his response on the scale. We have 9 textiles to evaluate with the secondary task. Then the subject has to evaluate some characteristics of the given textile, but without secondary task.  

The secondary task is to say “beep” whenever the subject hears a sound. We measure the latency of the response.

So, the subject is doing the probe reaction test and at the same time he is observing the textiles and assessing them. I have already done the primary and secondary task (with 10 stimuli), but I do not know how to put them together.

If you have any further questions, do not hesitate to contact me (I do not know if I was clear enough! J)

 

Thanks,

 

Agnes

 

P.S. The experiment is in French.

Hi Agnes,

as I'm not associated with Millisecond Software (I am - much like you - just a normal user of Inquisit), I don't have access to the scripts you sent to inquisit@millisecond.com. But anyway, could you elaborate on the connection between your two tasks in terms of timing? I.e. when do you want the "beep" sounds to be presented during a trial? Do you want every evaluation trial to include your secondary task (i.e. the "beep" detection) or only n out of x trials? Please try to be as specific as you possibly can.

Best,
~Dave

Hello,

 

Ok. So the sounds are presented randomly during the global assessment (this is independent from the primary task), this is the first block of the experiment. In fact we start the primary and the secondary tasks at the same time, but then there is no connection between the timings unless the fact that the secondary task finishes at the same time as the first block of the experiment (global assessment).

So, first you have a description of the task, then you have the question of global assessment on the screen, and the secondary task begins when the subject press the space bar (he starts to observe the textile). The sounds are presented every 1-6 seconds and if the subject has 2 seconds to respond by saying something. Meanwhile he observes the textile as long as he wants, then he answers to the question. Then he presses the space bar, he observes the textile, responds and so on. When he finishes the block, the secondary task (the beeps) stops, and he moves on the assessment of characteristics (without secondary task).

 

Is it clearer now? Is there any other way to show you the programs that I have already done?

 

Thanks,

 

Agnes

Thanks, think I'm seeing a bit clearer now. If you want to make your scripts available to others (like me), there are several possibilites:

- If the scripts are rather small (in terms of filesize), you can attach them to your posts in this forum. See here for a description.

- If your scripts are too large, you can either upload them to some personal webspace available to you or you can upload them to one of the many free file hosting services on the net (like rapidshare or megaupload) and then post the download link for the files here.

Best,
~Dave

Thanks, I have attached it.

Thanks for attaching the script. From what I've seen so far, this is a non-trivial issue. I'd really have to put on my thinking cap and devote some time to this. Right now, I am  pessimistic as to whether your specific dual-task design can presently be implemented with Inquisit. The main problem is that Inquisit will generally only allow a single response modality (keyboard input OR mouse input OR vocal input...) per trial, while your task clearly would require at least two (mouse input AND vocal input) per trial. Finding a workaround will not be easy (if at all possible). That's all I can say for now. I'll post to this thread if enlightenment strikes me...

Best,
~Dave

As Dave noted, Inquisit currently doesn't dual tasks involving mixed response modalities. You are by no means the first person to ask about this feature, however, so it's definitely on our list for a future release.

One way you might be able to pull this off is by running the experiment simultaneously on two separate computers. To ensure the computers are in sync, you could run a cable between their parallel ports and program the experiments to communicate via TTL signals. I know of at least one case where the researcher has done this successfully.

-Sean

Hello,

Thanks for this information, I have doubted that it would be possible, but I just wanted to be sure of it.
However, I will try to run the programs on two computers simultaneously, and I report on the results.

Best wishes,
Agnes Buvar

Dear Agnes,

after some hours of thinking, running the task using two synced computers is exactly what I wanted to suggest today. Needless to say, Sean clearly beat me to it... I think this is a good solution. I'd be very interested in your experiences with this setup, so please do report them here once you've worked it out.

All the best,
~Dave

Hello,

 

Is it possible to connect the two computers via USB ports? I haven’t found the other cable yet with the parallel ports. (The problem is that I work with two laptops, and their parallel ports are not the same that the ones on a PC.)

 

Thanks,

 

Agnes Buvar

 

Hi,

as far as I know, this is not possible. What notebook types / models do you have to deal with?

~Dave

Hello,

 

Thanks for the quick answer, in fact I have to start the experiment Wednesday and I am not ready yet. L

So it is two Dell computers. I have asked for a parallel port cable, but we have only found one that is good for PCs (the head of the cable is to big for the laptops).

 

What can I do in this case?

 

Agnes  

Okay - do you have any more specific information on the model of the Dell notebooks? If so, I could check out their specifications online and possibly point you to a working connector cable. If both of your laptops indeed have a parallel port, basically any cable should fit (they are normed). On some laptops, certain cables won't fit just because their connectors are too bulky and other parts are in the way (due to a laptop's constrained space). If you could post the model number of your laptops here (or even take a picture of their connectors), we might see a bit clearer.

If you can't find any fitting cable to connect them, you can still try and 'manually' synchronize the computers, i.e. have the experimenter start the auditory task on the second machine when the participant begins with the evaluation task on the primary computer.

~Dave