Pavlovia not calculating mean reaction time

Online experiments
1

Using the latest version of Psychopy. URL for experiment is here: https://run.pavlovia.org/JDiggins/sart2

There is code in within the experiment which uses the .rt from a keyboard command, however this causes a crash and from what I can see it’s because Pavlovia is not recording the reaction times.

Is there a tickbox I am missing?

Jake

2

I think you are misunderstanding the error message

  • TypeError: Cannot read properties of undefined (reading ‘key_resp_3.rt’)

The error is from meanRt = trials.data["key_resp_3.rt"].mean();

so trials.data is undefined. If the rt was missing it woudl be a key error instead.

What I do is append the RT to a list (when correct) and then calculate the median of that list rather than trying to access the data frame.

3

I now have it appending the .rt to meanRT as a list. I then have meanRt = meanRt.mean() but it is returning that .mean is not a function.

4

It’s math.mean(meanRt)

5

So I have changed it to that and I am now getting KeyError: math is not defined.

6

I’ve just checked and math is Python. Math is JS, so you could try Math.mean

As I mentioned, I normally use median, but in one experiment I defined an average function in a JS component.

average = function(data){
  var sum = data.reduce(function(sum, value){
    return sum + value;
  }, 0);

  var avg = sum / data.length;
  return avg;
}
7

I think I’m getting somewhere (I hate JS!) so I have it plugged in as meanRt = average(meanRt), after declaring your function at the beginning of the experiment. The experiment now runs through but I get “NaN” as an answer.

8

I don’t like working in JavaScript either. Here’s a simpler function which should work that Auto translates. Note I’ve put it in Before Experiment

def average(x):
    return(sum(x)/len(x))

image
image979×224 7.58 KB

Try printing meanRt to check it’s a list of reaction times.

Also, I would recommend using separate variable names, e.g. meanRT = average(listRT)

9

Brilliant, that’s all fixed now.

I added a rounding to the function as well just for ease.

Thank you

1 Like