An Important Update for EGI Users

Recently, while providing support for a lab that is using EGI, two issues came to our attention. If you have used any Cedrus equipment to send event markers, e.g. StimTracker, c-pod, or m-pod, this blog post applies to you; read on. Even if you used other software and hardware such as E-Prime, Chronos, or PsychoPy, you might still find this blog post relevant.

First Issue – It’s a Matter of Logic

Positive logic is what most users are familiar with. The output voltage from m-pod is normally zero volts (0V). When sending an event marker for, say, 10ms, the output voltage rises to +5V and drops back to zero after 10ms.

With negative logic, it’s the reverse: the output voltage is normally +5V. When sending an event marker, the voltage drops to zero and then goes back up to +5V after 10ms:

While positive logic is more popular, it is not unusual for electronic devices to adopt negative logic.

If you were wondering aloud “who cares?”, you would not be wrong. After all, we want our hardware to just work and we do not care which logic is used inside. The trouble arises when two devices want to communicate with each other and they use different logics.

What Happened?

After developing the interface between the original StimTracker ST-100 model and Net Amps, Cedrus worked closely with EGI to test compatibility and create a support page on how to setup Net Station for accepting event markers. This was done using Net Station version 4.5.5.

During this joint effort, the fact the Cedrus and EGI are using different logics was somehow missed.

In practice, this means that Net Station was recording the offset of a signal instead of its onset. For example, if you were playing a sound for 100ms and StimTracker was marking this sound and generating a corresponding 100ms TTL pulse, Net Station was recording when the sound has finished playing instead of when it started to play.

The Fix

After we became aware of this difference, we updated the firmware for StimTracker Duo and Quad, m-pod, and c-pod. Version 2.2.7 provides support for negative logic; see Changing m-pod’s Output Mode and Logic:

The RB Series response pads and Lumina do not require a firmware update. The hardware in the original StimTracker ST-100 model is hardwired for positive logic and cannot be changed via a firmware update.

Second Issue – Net Station 5 Does Not Record Offsets

Net Station 4.5 offered the option to record the onset of an event marker or its offset (but not both). Starting with Net Station version 5, some software options were removed. When using negative logic with Cedrus products, Net Station 5 no longer recorded the offset of a signal — only its onset. This is not unusual either: once in a while, companies will remove some features in order to tame software complexity. Cedrus is currently doing something similar.

If you are using an m-pod, a workaround is to change its output mode from the default “Reflective” to “Double Pulse” mode. Using the example above of playing a 100ms audio, Reflective mode would have m-pod generate a 100ms pulse but Double Pulse mode would have it generate two short pulses instead, allowing Net Station to record both the onset and the offset; see Changing m-pod’s Output Mode and Logic. A pulse duration of 1ms is recommended.

There is no built-in workaround if you are using c-pod.

How You Can Correct Your Data

This is undoubtedly the harder part. If you were not aware of negative logic being used, then your data analysis is off. For example:

If you were sending a 20ms event marker via USB at time T, Net Station 5 was recording the event at time T+20ms.
If you were having StimTracker mark the onset of audio or visual stimuli, Net Station 5 was recording the offset instead.

The best way to correct, we think, depends on the types of event markers that you were collecting:

If you were sending only event markers via USB, the correction is easy: subtract the pulse duration from your recorded offsets in order to obtain the onset.
If you were having StimTracker mark the onset of visual and auditory stimuli, make sure to update the firmwares first and then re-run the same experiment, preferably a few times. You will then be obtaining the onset of events and can compare to the offsets and subtract the difference in order to correct.

In Summary

EGI uses negative logic. It always has.
Cedrus products used positive logic until recently. Firmware version 2.2.7 or later lets you use either.
Net Station 4 was able to record the onset or the offset of an event marker, but not both.
Net Station 5 records only the onset.
When the logic does not match, i.e. Cedrus products using positive logic, then what you thought was the onset was actually the offset.

Other Devices

As of this writing, the only other device that we know of that uses negative logic is SAGA by TMSi.

Blog