problem with output of ga data in fitness function

I do have a problem with the output of my ga. I formulated a ga problem in any form. My fitness and output function are shown below. The output function is pritty much a modified copy of the Matlab Docu example.

I want my ga to give me other values, then the actual score as an output. I highlighted it down in the code:

problem.fitnessfcn = @(x) myfitnessfun(x,otherParameter); 
problem.options = optimoptions('ga','UseParallel', true,'PopulationSize',10,'OutputFcn',{@gaoutfun},'Display','iter');

[x,fval,exitflag,output,population,scores] = ga(problem);

%% This is my fitness function
function K = myfitnessfun(x,otherParameter)
% Definition of persistent variables for the history
persistent n k my_output_history
% the struct variable my_output is calculated with the function my_special_function as a function of x and otherParameter
[my_output] = my_special_function(x,otherParamter)
% The scalar fitness is calculted with a operator function make_scalar_value
K = make_scalar_value(my_output);

%% Here I use the persistent variables to store and save the values of my_output in the workspace. 
% This is only working in serial run of the ga. It is not working for the parallel run
% I save all the data of all runs of the finess function in the cell array my_out_history
if isempty(n)
    n=1;
    k=1;
    my_output_history = cell(1e4,2);
end

if rem(n,1) == 0
    my_output_history{k,1} = my_output;
     my_output_history{k,2} = x; % For the later assignement of the values i also store the vector of x in the cell array
    assignin('base','my_output_history', my_output_history);
    k = k+1;
end
n = n+1;

%% I want to access and save all the different values of my_output into the workspace. 
% Here i want something like:
my_output_history(state.Individual) = my_output;
assignin('base','my_output',my_output_history);
end

%% This is my output function 
% I cannot hand over the gaoutfun other values than options, state and flag
function [state,options,optchanged] = gaoutfun(options,state,flag)
persistent  pophistory scorehistory bestscorehistory
optchanged = false;
switch flag
    case 'init'
        pophistory(:,:,1) = state.Population;
        scorehistory(:,1) = state.Score;
        bestscorehistory(1) = min(scorehistory(:,1));
        assignin('base','gapopulationhistory',pophistory);
        assignin('base','gascorehistory',scorehistory);
        assignin('base','gabestscorehistory',bestscorehistory);
        
    case 'iter'
        % Update the pophistory every 10 generations.
        if rem(state.Generation,1) == 0
            ss = size(pophistory,3);
            pophistory(:,:,ss+1) = state.Population;
            scorehistory(:,ss+1) = state.Score;
            bestscorehistory(ss+1,1) = min(scorehistory(:,ss+1));
            assignin('base','gapopulationhistory',pophistory);
            assignin('base','gascorehistory',scorehistory);
            assignin('base','gabestscorehistory',bestscorehistory);

            ibest = state.Best(end);
            ibest = find(state.Score == ibest,1,'last');
            bestx = state.Population(ibest,:);
            disp(['best x in population: ', num2str(bestx)]);
%             pause(0.01)
        end
        
    case 'done'
        % Include the final population in the pophistory.
        ss = size(pophistory,3);
        pophistory(:,:,ss+1) = state.Population;
        scorehistory(:,ss+1) = state.Score;
        bestscorehistory(ss+1,1) = min(scorehistory(:,ss+1));
        assignin('base','gapopulationhistory',pophistory);
        assignin('base','gascorehistory',scorehistory);
        assignin('base','gabestscorehistory',bestscorehistory);
        
end
end

Is there a way, i can access the actual state of the ga in the fitness function like it is done in the output function? At the moment I`m using the version that can be found above, but this i not working for parallel computing.

Basically i want at the end: The history not only of the scores (scalar values) and the population information but also of my own variable (struct) my_output, that is calculated as a function of x in the fitness function.

Any good ideas how do do this?

Thanks a lot!

[Matt J]

Since you are using UseParallel what does "history" even mean? In parallel computing, there is no well-defined sequence of anything.