This post will focus primarily on visual stimuli and the
method used to deliver them to the projector in our imaging facility. In an
ideal world, you would be able to take everything I lay out here and translate
it directly to your own facility. Alas no two snowflakes are alike. But by the
end of this post you should have a better understanding of how a system like
this is established. Afterwards I’ll use the same method to set
up the auditory stimuli as well. The actual projector and its various
components will be covered in a follow-up post.
As we explore visual stimuli I want you to keep our
minimalistic design philosophy in mind. “Keep it simple, stupid” was pounded
into my head by my engineering professors early on and I still find myself repeating it whenever a new problem arises.
Since I’m a visual learner, I’ll be using a series of charts
to illustrate the process as I go through it. This also helps keep track of
cables when it comes time to actually make all the physical connections. The
coloring scheme used in the charts is simply to distinguish between the various
components. In general I like to keep the inputs one color palette and outputs
another.
Setting up the Video Switch
First begin by writing down every possible input you’re
going to need. If you are just starting out this can be a difficult task.
Without any prior experience I’d say it’s almost impossible! Luckily I have
years of experience and user feedback at my disposal, so feel free to use my
center’s setup as a template. I’m not saying it’s perfect but we do pride
ourselves on providing as many options as possible, including room to add to
the system later on if needed. The last thing we want to do is limit the research
being done due to an inadequate facility.
Figure 1 - video switch inputs |
As you can see we have six distinct inputs:
- Stimulus PC - for working with Presentation and E-Prime
- Stimulus Mac – for working with Matlab and the Psychtoolbox add-on
- Eyetracking PC – for working with our Avotec eye tracker and Biopac physiological recording system
- Laptop – for those that wish to bring in their own computer
- EyeLink PC – for using our EyeLink eye tracking system
- Old Stim Mac – our old stimulus Mac. Temporarily left in place due to a lingering study. This will be removed once the study is completed
Now look at the input type for the different sources. I know
what you’re thinking. How convenient that they are all the same, right? Absolutely!
Keep it simple, stupid. Our projector only accepts VGA inputs so the decision
of which signal type to use has already been made for us. Most computers come
standard with a VGA port and there are several after-market adapters for Macs, so
it’s an easy solution to go with even though it has limitations. (See Notes 1
and 2.)
Next up we add in the different outputs needed, which is again
based on previous experience and requests made by our user base:
Figure 2 - video switch inputs and outputs |
Our four outputs are:
- Projector – Avotec silent vision model SV-6011 LCD projection system
- Display monitor – In the control room. Useful so that researchers can see the same image that is being projected to their subject inside the magnet
- Stimulus input monitor – In case an extra monitor is needed in the control room
- Eyetracking goggles – For tracking eye movement via goggles that mount on the head coil [not currently being utilized because we already have two separate eye tracking units that are better and easier to use]
Figure 3 - video switch |
Figure 4 - Laptop VGA, RCA and USB cables |
Setting up the Audio Switch
Since it uses the same basic principle and we’re pretty much
experts by now, I’m going to go ahead and quickly run through the auditory
setup. Following the same procedure gives us this diagram:
Figure 5 - audio switch inputs and outputs |
It has even less components than the video setup:
- Stimulus PC – audio for the stem PC
- Stimulus Mac – audio for the stem Mac
- Laptop – audio for any computer our users wish to bring in
- Open connection – usually free, but currently being used by our old stem Mac
- Siemens Audio – communication system that comes standard with a 3T Trio
- Control speakers – a set of speakers setup in the control room. So researchers can hear what the subject inside the magnet is hearing
- 32 Ch. ear buds – a smaller profile audio system that is used with our 32 Ch. head coil. Expect a post on these in the future as well
The audio switch has less inputs simply because less are required and our users have not requested tapping in audio from other source yet. The audio switch we went with is this one by Burst
Electronics. It has stereo RCA connectors in the back, so you’ll need to
purchase a few adapters
to get everything to play nicely. Make
sure to future proof just like with the video switch and buy more than what you
currently need.
Figure 6 - audio switch |
The Finished Product
Once the equipment is mounted on a rack and all cables have been connected, your setup might look a little something like this:Figure 7 - equipment rank with video and audio switch |
There are components pictured that I have not mentioned, but
for now I will only point out what is specific to this post:
- Stimulus Mac - a Mac Mini encased in a rack mount
- Audio switch
- Video switch
- 32 Ch. ear bud system - three separate components mounted on top of each other
- Old stimulus Mac
- Stimulus PC
Not exactly the prettiest, but you should get the idea of what it all looks like when it comes together. By all means, feel free to arrange your physical setup however is most convenient and appropriate for your space. The racks and desks pictured work just fine for our facility. FYI although it's not labeled on the photo, the projector is located on the bottom of the rack. Next is a picture of the stimulus Mac, stimulus PC and display monitors:
And here’s what we end up with behind the scenes:
Figure 8 - Mac, PC and display monitor |
Figure 9 - video and audio switch cable webs with color coded tape |
I considered not including the last picture since it tends to
confuse and stress people out, but this will be unavoidable once
you start connecting everything up. I know it’s not pretty, but there is a
method to the madness! Just know you are going to have many cables running
through your facility. At times it may seem overwhelming and no matter how you
bundle them up and try to make them manageable they’ll still be a pain to go
through. Just remember to color code everything, maintain a detailed diagram or
list of the connections, and take plenty of reference photos. I’ve spent
countless hours tracing connection issues, unplugging incorrect cables, and
other nonsense that could have been avoided if I had kept detailed
documentation from the very beginning. Proper documentation can be the difference
between a quick solution and wasting an entire day. Trust me. I learned this
the hard way so that you don’t have to!
Notes:
- Each computer must also output a video signal to a primary monitor inside the control room, otherwise it would be extremely difficult to set up the experiment. Imagine having to look inside the magnet at the projector screen each time. Any two-port VGA splitter will work, or even an entirely different type of signal (DVI, HDMI or DisplayPort) if that’s easier for you to work with. In my experience however, computers can get a bit finicky when outputting multiple signal types.
- VGA signals are analog; where as the other types I mentioned are digital. This can lead to image degradation if the resolution is set too high. Our projector accepts signals with a maximum resolution of 1024 x 768 and a refresh rate of 60 Hz. We’ve had our share of headaches with these limitations, but thankfully nothing that prevented a study from continuing. I will be discussing some of this in greater detail in a future post centering around the projector and mirror systems used along with it.
A further note on the mass (mess?) of cables at the back, behind the scenes. We prefer that users do not make cable swaps, with the exception being the use of their own laptop. (Here we provide naked connectors, shown in Fig.4.) Swapped cables invariably fail to get swapped back properly, and it is nigh on impossible for the next user to identify and fix the problem in a pinch. By using only switch boxes to control all configurations it is trivial to set up a default configuration. A huge fraction of user-on-user malpractice, mostly requiring lost scan time as staff get involved, is eliminated this way.
ReplyDeleteYes! Thank you for including that information! All it takes is one wrong connection to completely screw over an entire session. We quickly establish a rule that absolutely no one was to touch the cables or disconnect them without first discussing it with staff and that eliminated the problem almost entirely
DeleteThanks a lot for the post. I am a researcher working on auditory stimuli and just transferred from MEG to fMRI. If possible, it would be great to see such a post on auditory stimulus presentation.
ReplyDeleteBest,
Shawn
Glad you found the post helpful! I do plan on doing a quick post in the future regarding our auditory stimuli setup, but it might be a while because I already have a few other posts lined up. Feel free to email me with any questions you may have though!
Delete