behavior_logistic_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from utils_sampling import *\n",
    "from funcs_behavior_logistic_analysis import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import set_matplotlib_formats\n",
    "set_matplotlib_formats('png', 'pdf')\n",
    "plt.rcParams.update({'font.size': 28})\n",
    "plt.rcParams.update({'font.family': 'arial'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "datasetName = 'takahashi2016roesch2009burton2018Valid' # or 'initialTrainingValid'\n",
    "\n",
    "# import data (valid trials only)\n",
    "dataValid = pd.read_csv('data/' + datasetName + '.csv')\n",
    "\n",
    "# mark if a trial is correct or not\n",
    "dataValid['correctChoice'] = 1*(dataValid['odor']=='left') + 2*(dataValid['odor']=='right') + (dataValid['odor']=='free')*(\n",
    "    1*((dataValid['blockType']=='short_long')|(dataValid['blockType']=='big_small')) +\n",
    "    2*((dataValid['blockType']=='long_short')|(dataValid['blockType']=='small_big')) )\n",
    "dataValid['correct'] = (dataValid['correctChoice'] == dataValid['choice'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "stanCodeName = 'logisticRegressionMultiVar'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "tags": []
   },
   "source": [
    "### logistic regression: correct/incorrect free-choice vs number of trials of correct forced-choice following the last incorrect free-choice + intercept"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "analysisName = 'lastWrong_NCorrectForced_improvement'\n",
    "fitName = datasetName + '_' + analysisName"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# run hierarchical logistic regression (skip this if fits are already saved locally)\n",
    "if not os.path.exists('model_fits/' + fitName + '_' + stanCodeName + '_allSamples.csv'):\n",
    "    ratList, startSubject, NCorrectForcedBetweenList, accuracyList = prep_data(dataValid, fitName)\n",
    "    dd = dict(P=1, Ns=dataValid['rat'].unique().size, Nt=len(ratList), startSubject=startSubject, X=np.array(NCorrectForcedBetweenList)[:,np.newaxis], y=np.array(accuracyList))\n",
    "    fit = fitModel(modelName=stanCodeName, datasetName=fitName, dd=dd, samplingInfo=samplingInfo, moreControl={'max_treedepth':10, 'adapt_delta':0.99})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "takahashi2016roesch2009burton2018Valid_lastWrong_NCorrectForced_improvement\n",
      "(-0.022, 0.035)\n"
     ]
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_hdi(fitName, stanCodeName, iRegressor=1, sort=False, reorder=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Add trial index as an additional regressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "analysisName = 'lastWrong_NCorrectForced_improvement_wiTrial'\n",
    "fitName = datasetName + '_' + analysisName"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# run hierarchical logistic regression (skip this if fits are already saved locally)\n",
    "if not os.path.exists('model_fits/'+fitName+'_'+stanCodeName+'_allSamples.csv'):\n",
    "    ratList, startSubject, NCorrectForcedBetweenList, accuracyList = prep_data(dataValid, fitName)\n",
    "    X = np.stack((np.array(NCorrectForcedBetweenList),np.array(iTrialList))).T\n",
    "    dd = dict(P=2, Ns=dataValid['rat'].unique().size, Nt=len(ratList), startSubject=startSubject, X=X, y=np.array(accuracyList))\n",
    "    fit = fitModel(modelName=stanCodeName, datasetName=fitName, dd=dd, samplingInfo=samplingInfo, moreControl={'max_treedepth':10, 'adapt_delta':0.99})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "takahashi2016roesch2009burton2018Valid_lastWrong_NCorrectForced_improvement_wiTrial\n",
      "(-0.035, 0.021)\n"
     ]
    },
    {
     "data": {
      "application/pdf": "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\n",
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plot the slope of interest\n",
    "plot_hdi(fitName, stanCodeName, iRegressor=1)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}