Need help plotting error bars to my data on a line plot

I'm having difficulty plotting error bars along the line plot of my data.

I'm getting the following error:

Error using errorbar>checkSingleInput (line 272)

YNegativeDelta must be empty or the same size as YData.

Error in errorbar (line 135)

yneg = checkSingleInput(neg, sz, 'YNegativeDelta');

Error in Experiment1_ParticipantLearningRate_Figure2 (line 41)

errorbar(y, eb_interval, allsubs_cond0_stderr)

Below is the script I'm using. For added context, allsubs_response_cond0_trials and allsubs_response_cond1_trials are both 120 x 40 matrices that represent the probability correct from 120 trials of 40 participants.

%%%Participant Learning Rate Figure%%%
load('Experiment1_Allsubs.mat')
%Creates Figure and plots data
figure; plot((smooth(mean(allsubs_response_cond0_trials, 2))),'LineWidth',2);  
hold on 
plot((smooth(mean(allsubs_response_cond1_trials, 2))),'LineWidth',2);

%Creates title 
title('Participant Learning Curve')

%Creates legend in bottom-left corner
legend({'Concrete First','Abstract First'},'Location','southeast')

% Create ylabel
ylabel('Probability of Correct Response','FontSize',24,...
    'FontName','Times New Roman');

% Create xlabel
xlabel({'Trial Number'},'FontSize',24,'FontName','Times New Roman');

% Defines participant learning rate standard error for each condition
allsubs_cond0_stderr = std(allsubs_p_correct_hist_cond0)./sqrt(length(allsubs_p_correct_hist_cond0));
allsubs_cond1_stderr = std(allsubs_p_correct_hist_cond1)./sqrt(length(allsubs_p_correct_hist_cond1));

%Transposes learning curve dimensions to compute error bars(changes 120x40
%to 40x120)
allsubs_response_cond0_trials =  allsubs_response_cond0_trials'; 
allsubs_response_cond1_trials =  allsubs_response_cond1_trials';

% Creates errorbars interval at every 10 trials along the x-axis from 10 to
% 120
eb_interval = 10:10:120; 

% Creates point on y-axis along data to plot error bars. My intention here
% is to take the mean probability of correct of all 40 participants at each
% desired interval point. 
y = mean((allsubs_response_cond0_trials(:,eb_interval))); 

% Plots error bars
errorbar(y, eb_interval, allsubs_cond0_stderr)

%Formats current axis and current figure
set(gca, 'FontSize', 18, 'FontWeight', 'bold')
set(gcf, 'Color', 'w')

[Star Strider]

The problem is that ‘allsubs_cond0_stderr’ is (1x40) and the other vectors in the errorbar call are (1x12). The other problem is that ‘eb_interval’ won’t work as an index into it (and similar vectors ) because they go from 10 to 120 in steps of 10, exceeding the size of the vector.

I am not certain what you are doing. All I can find is that (with synthetic matrices for ‘allsubs_response_cond0_trials’ and ‘allsubs_response_cond1_trials’) the vector lengths do not match, and I am at a loss as to how to make them match. There are also msssing variables.

Perhaps with the missing variables and actual data, this problem could be solved.

.

[Ryan Gallagher]

This is a screen shot of the .mat file that is loaded in the beginning of the script. For clarification, "allsubs_response_cond0_trials" is the probability of correct for 120 trials and all 40 participants. The other variable you reference "allsubs_cond0_stderr" is the singular value representing the standard error calculated for each condition.

Answer 1

[Shanmukha Voggu]

Hi Ryan,

The error is due to the third argument of the errorbar function is not of the same size as the first argument.

The third argument allsubs_cond0_stderr is a constant and the first argument y is [1 x 12] matrix.

A possible fix that will resolve the error is to make the allsubs_cond0_stderr as [1 x 12] matrix (changing constant to matrix that has each and every value equal to constant)

% Plots error bars
allsubs_cond0_stderr=allsubs_cond0_stderr(ones(size(y)));% add this line above the errorbar function
errorbar(y, eb_interval, allsubs_cond0_stderr);

Refer to this for more information.