Quota for balancing inter-sessions stimuli

[leya-p]

Hello,
I'm preparing an experiment with jsPsych and the MindProbe server. I'm still discovering coding. In this experiment, I have a list of 128 audio stimuli: each participant is exposed to only 1 stimulus from this list.
How can I ensure that the 128 stimuli are presented in roughly equivalent proportions when all the completed sessions are taken into account?
I've figured out how to randomize the choice of stimuli within a run for a participant, but here I'd like the script to be able to propose one of the 128 stimuli on the list according to those already proposed before to the other participants. I would therefore need to have counters that could store which stimuli have already been proposed in the previously recorded results, in order to choose which stimuli to present to the current participant.

Here's some of my code. Currently, the one stimulus to be presented is chosen randomly from the list of 128 stimuli, regardless of previous runs. It works.
Thank you if you can help me :)

var list_stimuli= [

{context: "sound/LUI_1contexte_FSO.wav", stimulus: "sound/LUI_1E_DisDom_FSO.wav",fillercontext: "sound/FILLER_1contexte_FSO.wav",fillerstimulus: "sound/FILLER_1_FSO.wav"},
{context: "sound/LUI_1contexte_FSO.wav", stimulus: "sound/LUI_1F_DisNoDom_FSO.wav",fillercontext: "sound/FILLER_1contexte_FSO.wav",fillerstimulus: "sound/FILLER_1_FSO.wav"},
{context: "sound/LUI_1contexte_FSO.wav", stimulus: "sound/LUI_1G_NoDisDom_FSO.wav",fillercontext: "sound/FILLER_1contexte_FSO.wav",fillerstimulus: "sound/FILLER_1_FSO.wav"},
{context: "sound/LUI_1contexte_FSO.wav", stimulus: "sound/LUI_1H_NoDisNoDom_FSO.wav",fillercontext: "sound/FILLER_1contexte_FSO.wav",fillerstimulus: "sound/FILLER_1_FSO.wav"},
{context: "sound/LUI_2contexte_FSO.wav", stimulus: "sound/LUI_2E_DisDom_FSO.wav",fillercontext: "sound/FILLER_1contexte_FSO.wav",fillerstimulus: "sound/FILLER_1_FSO.wav"},
{context: "sound/LUI_2contexte_FSO.wav", stimulus: "sound/LUI_2F_DisNoDom_FSO.wav",fillercontext: "sound/FILLER_1contexte_FSO.wav",fillerstimulus: "sound/FILLER_1_FSO.wav"},
…. (the list is not exhaustive here)
];

var itemContext= {
  type: jsPsychAudioKeyboardResponse,
  stimulus: jsPsych.timelineVariable("context"),
  prompt: "<img src='img/bulles.png'></img>",
  trial_ends_after_audio: true,
  choices: "NO_KEYS",
  post_trial_gap:500
};

var fillerContext= {
  type: jsPsychAudioKeyboardResponse,
  stimulus: jsPsych.timelineVariable("fillercontext"),
  prompt: "<img src='img/bulles.png'></img>",
  trial_ends_after_audio: true,
  choices: "NO_KEYS",
  post_trial_gap: 500
};

var itemStimulus= {
  type: jsPsychAudioButtonResponse,
  stimulus: jsPsych.timelineVariable("stimulus"),
  prompt: "<br /><p>1 = pas du tout acceptable <br/> 7 = parfaitement acceptable</p>",
  choices: ["réécouter","1", "2", "3", "4", "5", "6", "7"],
  margin_vertical: '8px',
  response_allowed_while_playing: false,
  post_trial_gap:800
};

var fillerStimulus= {
  type: jsPsychAudioButtonResponse,
  stimulus: jsPsych.timelineVariable("fillerstimulus"),
  prompt: "<br /><p>1 = pas du tout acceptable <br/> 7 = parfaitement acceptable</p>",
  choices: ["réécouter","1", "2", "3", "4", "5", "6", "7"],
  margin_vertical: '8px',
  response_allowed_while_playing: false,
  post_trial_gap:800
};

var itemBoucle= {
  timeline: [itemStimulus],
loop_function: function (data){
    if(data.values()[0].response == [0]){
        return true; 
    }   else {
        return false; 
    } 
  } 
};

var fillerBoucle= {
  timeline: [fillerStimulus],
loop_function: function (data){
    if(data.values()[0].response == [0]){
        return true; 
    }   else {
        return false; 
    } 
  } 
};

var jugements= {
  timeline: [fillerContext, fillerBoucle, itemContext, itemBoucle],
  timeline_variables: list_stimuli,
  sample: {
    type: "without-replacement",
    size: 1
  }
};

var timeline= [preload, consentement, instructions, jugements, sac, choco, age, geo, cloture];

[leya-p]

Hello again,
I have find a solution with the exemple study: "Randomize Tasks Between Workers" on https://www.jatos.org/Example-Studies
If someone is looking for the same thing.
Léa