psychopy.org | Reference | Downloads | Github

Memory error related to increased number of trials


#1

Dear users,
I’ve designed an experiment in builder which keeps crashing and reporting a “memory error” - I’ve provided the traceback below. The experiment is quite complicated, and I will try to summarize it briefly below.

Running: C:\Users\RIF\Desktop\WERP_Experiment 2\Replication study\Phase2B_lastrun.py
pyo version 0.6.6 (uses double precision)
Portmidi warning: no input device!
Traceback (most recent call last):
File “C:\Users\RIF\Desktop\WERP_Experiment 2\Replication study\Phase2B_lastrun.py”, line 2025, in
test_target.setImage(welsh_test_picture)
File “C:\Program Files (x86)\PsychoPy2\lib\site-packages\psychopy-1.83.04-py2.7.egg\psychopy\visual\image.py”, line 261, in setImage
setAttribute(self, ‘image’, value, log)
File “C:\Program Files (x86)\PsychoPy2\lib\site-packages\psychopy-1.83.04-py2.7.egg\psychopy\tools\attributetools.py”, line 100, in setAttribute
setattr(self, attrib, value) # set attribute, calling attributeSetter if it exists
File “C:\Program Files (x86)\PsychoPy2\lib\site-packages\psychopy-1.83.04-py2.7.egg\psychopy\tools\attributetools.py”, line 20, in set
newValue = self.func(obj, value)
File “C:\Program Files (x86)\PsychoPy2\lib\site-packages\psychopy-1.83.04-py2.7.egg\psychopy\visual\image.py”, line 249, in image
maskParams=self.maskParams, forcePOW2=False)
File “C:\Program Files (x86)\PsychoPy2\lib\site-packages\psychopy-1.83.04-py2.7.egg\psychopy\visual\basevisual.py”, line 697, in _createTexture
intensity = numpy.array(im)
MemoryError
Exception TypeError: “‘NoneType’ object is not callable” in <bound method Server.del of <pyolib.server.Server object at 0x04D4EE30>> ignored

^ Note that the final line concerning TypeError ALWAYS appears, regardless of whether or not the experiment runs to completion - this seems to be a minor/harmless bug caused by the audio backends

The experiment
The experiment begins with a list of 60 stimuli which are randomly assigned to one of three conditions. The experiment comprises three phases, and stimuli are assigned to phases depending on which condition they were allocated to. In the first two phases, words or pictures appear onscreen and participants must read/name them aloud. Each vocal response is detected by a microphone, which triggers onset of the next trial. During these phases, each stimulus is repeated n times. In the final phase, stimuli are presented onscreen in pairs (matched according to their assigned condition), and participants have to make a lexical decision task with a keyboard response.

Here’s the issue: when each stimulus appears only once in the first two phases, the experiments runs to completion. However, if I increase the number of repetitions to ten, the experiment crashes during the third phase. Stranger still, a different version of the experiment - in which the order of tasks in the first two phases is reversed - runs perfectly. Given that the error occurs AFTER the phases which seem to make a difference, I feel like this is somehow related to cumulative memory load.

I’d be very grateful for any insight / suggestions. I can also provide the psychopy script and necessary materials to replicate the problem

Specs:
PsychoPy version: 1.8.3.04
Windows 8.1
8.00GB RAM (6.94 GB usable)
System Type: 64-bit operating system, x64-based processor
Processor: AMD A10-7850K APU with Radeon™ R7 Graphics 3.70 GHz