1. NIMH ML and, as far as I know, MATLAB itself are not dependent on Qt. The error is likely to come from the other software that you may run together. 2. What do you mean by "lost"? Did you compare the signal length between analog inputs and the button? You know that a trial starts from ITI when the ITI recording is enabled, don't you? https://monkeylogic.nimh.nih.gov/docs_Maiby Jaewon - Questions and Answers
I took the liberty of moving the timing script to the attachment. I have a few questions. 1. Is pressing the space bar the act of selecting a number? What should be displayed on the screen, when the subject enters the number? 2. Do you need to know the time when the enter is pressed? 3. Other than space, numbers and enter, what keys can the subject type?by Jaewon - Questions and Answers
I sent you a modified version of mlbhv2.m by email. Please check your inbox. The data could be read up to Trial200. The new mlbhv2.m will be included in the future release of NIMH ML, so you do not need to keep the file.by Jaewon - Questions and Answers
Now the message is supposed to be a warning, not an error, and anything readable up until the warning should be in the return value of mlread(). Did you see anything in the return value?by Jaewon - Questions and Answers
It means you are still running the old version. If you ran the new version, at least the line number in the error message would be different. Please check the MATLAB path and make sure that the commands in the new version are called. The new version returns the readable portion of data even when an error occurs.by Jaewon - Questions and Answers
Try to read again with the latest NIMH ML.by Jaewon - Questions and Answers
I do not see such a problem during my tests. The condition of the first trial in each block is as random as it can get on my computer. What you need to check is whether the condition assigned in the selector function and the current condition in your timing script are the same. If so, there is nothing wrong in getting the first condition at the beginning of each block. Using randi() does not eby Jaewon - Questions and Answers
Please explain what you are trying to achieve, not how you want to do it. Why do you need getkeypress and moviegraphic?by Jaewon - Questions and Answers
I guess you probably did not use lsl_loadlib correctly, but neither lsl_loadlib nor LabRecorder was built by me, so you have to contact the developers of those applications if you think there is something wrong with them. I do not know whether LabRecorder allows running other applications at the same time. You should try running it on another computer. It is not a good idea anyway to slow downby Jaewon - Questions and Answers
Read this manual page carefully. https://monkeylogic.nimh.nih.gov/docs_CreatingTask.html#HowToCreateAdapterChains In your code, TTLOutput is the topmost adapter, which means that the scene ends when TTLOutput stops. If fixation was the behavior you wanted to acquire in the scene, you should have made WaitThenHold the topmost adapter.by Jaewon - Questions and Answers
I see where the manual was not written clearly. You should set the duration of TTLOutput, like the following. ttl1.Duration = 50;by Jaewon - Questions and Answers
If you have a psychopy function that works for this already, why don't you write a MATLAB function that does the same thing? Are you saying that you did so but it did not work? By the way, you can call python functions from MATLAB, too.by Jaewon - Questions and Answers
Both errors are your programming mistakes and have nothing to do with NIMH ML. You know that your TimeCounter (tc1) is not included in the adapter chain and therefore does not do anything, don't you?by Jaewon - Questions and Answers
You are asking the question to a wrong person. You should ask it to the maker of the software that generates the .xdf file.by Jaewon - Questions and Answers
You cannot use just any I/O board. It has to be supported by NI-DAQmx. It is not impossible to support the drivers of other products, but I just do not have time to do so.by Jaewon - Questions and Answers
It is just impossible that some markers are stored and others are not. You should look harder what really happened.by Jaewon - Questions and Answers
A sound card can process multiple channels, so it is unusual to install more than one card. I think the best option for you now is to use two mono microphones and record them as stereo. You may need a stereo-to-dual-mono jack or splitter.by Jaewon - Questions and Answers
Did you start sending packets on the DQW side? It does not send anything until you click some button.by Jaewon - Questions and Answers
If you are scripting in the v1 style, you can monitor the eye position around the initial fixation point with eyejoytrack('holdfix',...) and, as soon as it returns with 0 (i.e., when the fixation is broken), call eventmarker. However, I advise against this unless you have a compelling reason to detect saccades in real time. Whenever possible, it is better to analyze saccades offline.by Jaewon - Questions and Answers
Depending on what you exactly want to do, there can be different suggestions, but usually this is what you need a mixer for.by Jaewon - Questions and Answers
I wrote an example task for you. This example requires you to define the boundaries of the pathways with non-overlapping rectangles. See the code in the attached file.by Jaewon - Questions and Answers
You can use the custom calibration function. See this manual page. https://monkeylogic.nimh.nih.gov/docs_CoordinateConversion.html There is also an example task of this in the "task\runtime v1\8 customized calibration\" directory.by Jaewon - Questions and Answers
The eye tracker has to have a voltage output option. If it provides only TCP/IP, it will not work. See this post. https://monkeylogic.nimh.nih.gov/board/read.php?3,1993by Jaewon - Questions and Answers
What is the width and height of your videos? I guess a video card with large memory and a powerful CPU can help with it, but it is not always possible to prevent skipped frames for large videos.by Jaewon - Questions and Answers
This is not something you can fix with a little tweak. I may be able to develop something without actual hardware, but, at some time point, the hardware will be required for testing whether everything works. Ask the vendor if they can send me a test unit. Tobii once lent me their top-of-the-line product for a month so that I could support their hardware. Maybe this company will do the same.by Jaewon - Questions and Answers
Due to a system upgrade made a couple of months ago, some functions of the board (e.g., posting new topics, creating new profiles, etc.) were disabled. Most of them are restored now, but please let me know if you find anything that does not work as expected.by Jaewon - News
* Changes in 2.0.270 (Dec 19, 2024) - OnsetDetector did not report RT due to a change made in 2.2.43. It is now working correctly. (Thanks to Chaoyi Zhang)by Jaewon - News
Establishing a TCP/IP connection alone does not guarantee any functionality. The critical factor is the structure of the transmitted data, which varies by vendor. Unless your eye tracker is listed among the supported hardware (https://monkeylogic.nimh.nih.gov/docs_TCPIPEyeTracker.html), it is not possible to retrieve eye positions from the connection. To support new hardware, I would need accessby Jaewon - Questions and Answers
Please do your part first before asking that kind of question. Take a look at the provided example tasks and manuals.by Jaewon - Questions and Answers
There are so many ways to detect such a behavior. You need to use a little imagination. For example, you can repeat the scene until the subject touches the target.by Jaewon - Questions and Answers