Outpt with zeros when should be values

I have a matrix of two columns. The first column is cue times, as in times an animal was cued to do something. The second column is times that the animal responded. If the animal correctly responded then the animal will respond within a range greater than .15 seconds of cue and less then 5 seconds from cue. If so this is a correct behavior and needs to go into one matrix(correcttrials), if not then this is an error behavior and needs to go in a second matrix(errortrials).

The original Matrix (I have also attached it):

intervals = 
81.4358          735.1798
272.0970  1442.0839
515.2985  1682.2252
575.41885  1961.58675
734.95985  2507.14965
915.20085  2681.83055
1319.4032  2682.0705
1441.0439  2777.6710
1526.62435  3498.9947
1681.4852  0
1961.38675  0
2087.0074  0
2372.6089  0
2559.9099  0
2681.73055  0
2838.63135  0
3049.3324  0
3140.97285  0
3322.3738  0
3498.87465    0

This is the code I have written, but I am not sure why the correct output has zeros in it and the error output gets all zeros.

Here is my code:

data=intervals;
cue=data(:,1);
response=data(:,2);
 for k=1:length(cue)
    for i=1:length(response)
        if reward(i)>= (cue(k)+.15) && reward(i)<= (cue(k)+5)
            correcttrials(i)=response(i); 
        else
            errortrials(i)=response(i);
        end
    end
end

Here are the Matrices I am getting out:

Correct Trials:

correcttrials=
735.1798  1442.0839  1682.2252  1961.58675  735.1798  0  2682.0705  1442.0839  0  1682.2252  1961.58675  0  0  0  2682.0705

Error Trials:

errortrials = 
0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0  0

I am intending to get:

Correct Trials:

correcttrials = 
735.1798  1442.0839  1682.2252  1961.58675  2682.0705

Error Trials:

errortrials = 
2507.14965  2681.83055  2777.6710  3498.9947

[Image Analyst]

We can't run this code because you have not defined the reward array.

Answer 1

[Andrei Bobrov]

intervals = [...
81.4358          735.1798
272.0970  1442.0839
515.2985  1682.2252
575.41885  1961.58675
734.95985  2507.14965
915.20085  2681.83055
1319.4032  2682.0705
1441.0439  2777.6710
1526.62435  3498.9947
1681.4852  0
1961.38675  0
2087.0074  0
2372.6089  0
2559.9099  0
2681.73055  0
2838.63135  0
3049.3324  0
3140.97285  0
3322.3738  0
3498.87465    0];

d = intervals;
dn0 = d(d(:,2)~=0,2)';
da = abs(d(:,1) - dn0);
t = any(da > .15 & da < 15);
correcttrials = dn0(t)
errortrials = dn0(~t)

[Krispy Scripts]

Thank you!

[Krispy Scripts]

If I wanted to instead get the values in column one that meet the criteria instead of column 2 could that be easily done?

Answer 2

[Image Analyst]

Try this vectorized way:

load('intervals.mat');
cueTimes = intervals(:,1); % Extract the first column.
responseTimes = intervals(:,2); % Extract the second column.
% Get the time differences.
timeDifferences = responseTimes - cueTimes
% Convert from milliseconds to seconds. ??????????
timeDifferences = timeDifferences / 1000;
correctTrialsRows = (timeDifferences > 0.15) & (timeDifferences < 5)
correctTrials = responseTimes(correctTrialsRows)
errorTrials = responseTimes(~correctTrialsRows)

None of the things was correct so I figured the times might be in milliseconds whereas your criteria was in seconds, so I converted. I'm still not sure if this gives you what you want though.

[Krispy Scripts]

It does not need to be converted into milliseconds. It also cannot be assumed that the value in one column corresponds to another. So thats why I did not look to get a time difference.

My method was to take each value in the first column, for your code cueTimes, and read through the entire second column. If any value is within the cutTime window, greater than .15 and less than 5 (seconds), than put that second column responeTime in correctTrials. If it does, not then it should go in the errorTrials.

Does that help at all? I am not sure about how to write for this though..

[Image Analyst]

Then none of your time differences are in the range 0.15 to 5 seconds. Let's take a typical start/stop pair - the pair in row #5. Start/cue time of 734.95985 seconds, and response (stop time) at 2507.14965 seconds. That's 1771 seconds. And all the rest are like that. Not one of them is in the range 0.15 to 5 - they're all hundreds.