Change Blindness Task

SuperLab SuperLab 4 Support
1

Dear Cedrus Support People,

Hope you all had a nice Christmas/ festive break.

I am writing in the hope that you might be able to help me with an experiment I am trying to design:

Purpose: Undergraduate Final Year Dissertation project
Participants: Children with and without Autism Spectrum Disorder

Operating System: PC – Windows XP
SuperLab Version: 4.0.7b
Level of Expertise: Novice – this is the first experiment I have tried to create using SuperLab

Type of Experiment: Change Blindness Task – Flicker Paradigm
Stimuli: Bitmap Images
Response Mode: Just keys on Keyboard (no external devices)

Details (what I want experiment to do):

  1. A black fixation cross will appear in the center of the display – the participant will press the ‘1’ key when ready to proceed.
  2. A photographic scene (version A) will then be presented for 400 ms
  3. Followed by a grey masking screen for 100 ms.
  4. The alternate version of the scene (version B) will then be displayed for 400 ms,
  5. followed again by a grey masking screen for 100 ms
  6. This “flicker” sequence will be repeated until the participant detects the ‘change’ between these 2 scenes. (e.g. object in scene appearing/ disappearing)
  7. Participants will press the space bar when they spot the ‘change’ (the reaction time of interest is the time between initial onset of the first photo and the point in time in which the spacebar is pressed)
  8. They will then be presented with a static version of the photo (Version A) and asked to report the location of the change by pointing at the screen.
  9. A timeout error will be recorded if the participant does not spot the change in 30 seconds.
  10. This process will be repeated for ~ 24 trials.

I attach the experiment I have created thus far (as a single .SL4 file and as an experiment package). It contains only 2 trials at this stage. I have managed to create the ‘flicker’ effect by using the feedback commands to instruct the events to repeat themselves . . . . but have been unsuccessful in the following:

A. RESPONSE RECORDING – I need to record the time taken between the onset of the very first stimulus photo and when the participant presses the spacebar. The response times recorded do not look right. They seem too short. Is the timer able to record this response time across multiple cycles of the same trial? (Detail # 7)
B. TIMEOUT – I cannot seem to program each trial to end after 30 seconds if participant has not responded (Detail # 9)

I am also unsure whether I have created the experiment correctly thus far – it has crashed a few times and I am not sure if that is due to a mistake on my part or not.

I am currently based in Dubai and I am not sure whether there is any local support I can tap into ‘on the ground’. Unfortunately my tutor at my university has not been able to help me. I feel that I am feeling my way in the dark a little bit.

Any advice would be greatly appreciated,

Thanks in advance

Eleanore

2

Change Blindness ‘Flicker’ task - update

Hello again,

As a potential work around to solve the previously mentioned2 issues, I have tried:

1. deleting the feedback command that instructed the trial (or “flicker cycle”) to repeat if the participant had not responded

2. and instead have replicated the 4 events that made up the ‘flicker’ 30 times.

i.e. instead of having Trial 1 consist of:

ImageA
mask1
ImageB
mask2
(+plus a command on 4th event saying to repeat trial if no response)

I changed it so that Trial 1 consisted of:

ImageA
mask1
ImageB
mask2
ImageA-1
mask1-1
ImageB-1
mask2-1
ImageA-2
mask1-2
ImageB-2
mask2-2
.
.
.
. etc etc . . .
.
.
.
ImageA-29
mask1-29
ImageB-29
mask2-29
(+plus a command on the last mask event saying if no response, move to next trial)

This seems to have solved the issue of ensuring each trial does not last longer than 30seconds … as well as the response timing problem I was facing.

However, it does not seem a very efficient way of going about things. My concern is that this is a lot of events to have in each block. ~720 events per block (between 16-24 trials per block are needed for my study).

1. Is there a limit to the number of events you can create in SuperLab?
2. Is this likely to slow down the program at all? (note: I have only done this for one trial so far)
3. Do you have any other suggestions of how I could do this?

I look forward to hearing back from anyone on this,

Eleanore

3

Sorry for the delay. Our office is closed between Christmas and New Years.

  1. No, there is not a limit on how many events can be created in SuperLab.

  2. Since your images are duplicates being displayed over and over, it should not slow down SuperLab.

  3. Another suggestion is to use a movie editor and develop a 30 second clip.

4

Chnage Blindness Movie

Thank you so much for the reply Monika - very helpful

I have never used a Movie Editing Program. Can you suggest the name of one?

Also as I already have the static images, would I then just use a movie editor to put them all together?

Thank you

5

Unfortunately I am not familiar with movie editors. I know that using movie editors for the first time does include a learning curve. If you have a colleague or know somebody that has experience with movie editors I think that would help you a great deal.

6

Change Blindness - Stimulus Loading

Hi Monika,

I have almost finished creating my experiment and ironing out the glitches. However I am facing a problem that I have not been able to resolve.

Consistently, on the 5th Trial (out of 32), the program pauses… the ‘home’ window appears again (with the three boxes: blocks; trials; events) and then a mini window pops up onto the screen saying ““loading all stimulus files for the experiment”” with an associated status bar. Once this status bar has run it’s course (a few seconds), the changeblindness flicker stimuli continues - but is only visible in the background, at the edges. The ‘home’ window remains on the screen. The only way to get rid of it is to press ESC and then click “no” (i.e. do not want to exit). It then goes away.

This consistently happens once the 5th trial is reached and inevitably disrupts the experiment and response times. I am not sure how to resolve this. Any ideas?

Is there a function that enables one to load all the stimuli at the very beginning before the experiment is run? Do you know why it is happening on the 5th trial?

I look forward to hearing from you

Ellie

7

In SuperLab, go to the Experiment menu and select Options. Then, choose Memory Management. To the right you will see two options for loading stimulus files. Select Load them all before running the experiment.