OS: Windows 11 PsychoPy version : PsychoPy Standalone 2026.1.3
Hello,
I have a recognition memory experiment created in Builder.
The experiment runs correctly and saves data correctly, but the keyboard RT values are incorrect. Instead of normal reaction times (e.g., 0.5–2 s), the column contains values around 8.30277E+15.
What I have already checked:
Latest PsychoPy Standalone (2026.1.3)
Previously tested with 2025.1.1 (same problem)
Tried both PTB and ioHub keyboard backends
Keyboard component starts with the stimulus
Register on = press
Store= last key
Sync timing with screen
Save onset/offset times
CSV data are saved correctly
The experiment itself runs normally; only RT values are wrong.
This is completely normal. “Technically” you are not going to get a perfect 1/60th increment (or whatever your framerate is) due to how internal clocks work. You can just take the first 3-4 digits. So your first cell is 1.034s, then 1.607, etc.
Excel may give you a prompt like this (or it is in your settings)
Thank you for your answer. However, I’m not sure if we’re seeing the same issue. For example, when a value appears as 2.03462E+16 in Excel, clicking the cell shows 20346197999897400 (without a decimal point). So it doesn’t seem to be just a display issue or scientific notation. Because of that, I don’t think converting it would produce something like 2.03462 seconds. Am I misunderstanding something?
Use a text-editor to take a look at your data or import your data with LibreOffice. I agree with @stanley1O1 that this is an Excel conversion result. Use a proper statistics software for data analysis.
