processing.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import cPickle\n",
    "import os\n",
    "from tools import *\n",
    "d = load_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "data was created\n"
     ]
    }
   ],
   "source": [
    "# Create or load a clean data file\n",
    "\n",
    "if 'clean_data.pkl' in os.listdir('.'):\n",
    "    d_clean = manual_load_data('clean_data.pkl')\n",
    "    print 'data was loaded'\n",
    "else:\n",
    "    d_clean = {}\n",
    "    print 'data was created'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P07\n",
    "\n",
    "- removed t14: lag. Trial name is Roman_Occlusion_NoCue_LeftToRight0001  \n",
    "- removed t28: lag. Trial name is Roman_Visible_Cue_LeftToRight0004\n",
    "- removed t44: drop in data. Trial name: Roman_Visible_NoCue_RightToLeft0001\n",
    "- edited t13: startmovement. Tiral name: Roman_Occlusion_NoCue_LeftToRight0000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P07'] = d['P07']\n",
    "\n",
    "del d_clean['P07']['trials']['t14']\n",
    "del d_clean['P07']['trials']['t28']\n",
    "del d_clean['P07']['trials']['t44']\n",
    "\n",
    "curve = np.array(d_clean['P07']['trials']['t13']['startmovement'])\n",
    "curve[400:600] = 0\n",
    "d_clean['P07']['trials']['t13']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P08\n",
    "\n",
    "- removed t22: drop in the data. Name: Roman_Occlusion_NoCue_RightToLeft0004  \n",
    "- removed t26: index & thumb deop at the end + lag. Trial name: Roman_Visible_Cue_LeftToRight0001  \n",
    "- removed t30: lag. Trial name: 'Roman_Visible_Cue_LeftToRight0007'  \n",
    "- removed t36: lag. Trial name: 'Roman_Visible_Cue_RightToLeft0006'  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P08'] = d['P08']\n",
    "\n",
    "del d_clean['P08']['trials']['t22']\n",
    "del d_clean['P08']['trials']['t26']\n",
    "del d_clean['P08']['trials']['t30']\n",
    "del d_clean['P08']['trials']['t36']\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P09\n",
    "\n",
    "- removed t13: earlier reach. Trial name: Roman_Occlusion_NoCue_LeftToRight0000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P09'] = d['P09']\n",
    "\n",
    "del d_clean['P09']['trials']['t13']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P 11\n",
    "\n",
    "- removed t1: early reach. Name: Roman_Occlusion_Cue_RightToLeft0003  \n",
    "- removed t13: early reach. Name: Roman_Occlusion_NoCue_LeftToRight0000  \n",
    "- removed t10: too early reach. Name: Roman_Occlusion_Cue_LeftToRight0000  \n",
    "- removed t17: drop data. Name: Roman_Occlusion_NoCue_LeftToRight0004"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P11'] = d['P11']\n",
    "\n",
    "del d_clean['P11']['trials']['t10']\n",
    "del d_clean['P11']['trials']['t13']\n",
    "del d_clean['P11']['trials']['t1']\n",
    "del d_clean['P11']['trials']['t17']\n",
    "\n",
    "# curve = np.array(d_clean['P11']['trials']['t17']['startmovement'])\n",
    "# curve[0:300] = 0\n",
    "# d_clean['P11']['trials']['t17']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P 12\n",
    "\n",
    "- removed t31: no data at the end of trial + lag. Name: Roman_Visible_Cue_RightToLeft0000  \n",
    "- removed t22: lag. Name: Roman_Occlusion_NoCue_RightToLeft0004  \n",
    "- removed t12: looks like a lag, data missing. Name: Roman_Occlusion_Cue_RightToLeft0005\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P12'] = d['P12']\n",
    "\n",
    "del d_clean['P12']['trials']['t31']\n",
    "del d_clean['P12']['trials']['t22']\n",
    "del d_clean['P12']['trials']['t12']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P16\n",
    "\n",
    "- removed t1: no reach. Name: Roman_Occlusion_Cue_LeftToRight0000  \n",
    "- removed t16: looks like a lag, no data. Name: Roman_Occlusion_NoCue_LeftToRight0003  \n",
    "- removed t26: early reach. Name: Roman_Visible_Cue_LeftToRight0001  \n",
    "- removed t41: no data at the beginning. Name: Roman_Visible_NoCue_LeftToRight0004  \n",
    "- removed t46: no data. Name: Roman_Visible_NoCue_RightToLeft0003  \n",
    "- removed t48: lag. Name: Roman_Visible_NoCue_RightToLeft0005\n",
    "- recalculated startmovement with wrist12vel\n",
    "- edited t22: startmovement. Name: Roman_Occlusion_NoCue_RightToLeft0004\n",
    "- edited t47: startmovement. Name: Roman_Visible_NoCue_RightToLeft0004"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P16'] = d['P16']\n",
    "\n",
    "del d_clean['P16']['trials']['t1']\n",
    "del d_clean['P16']['trials']['t16']\n",
    "del d_clean['P16']['trials']['t26']\n",
    "del d_clean['P16']['trials']['t41']\n",
    "del d_clean['P16']['trials']['t46']\n",
    "del d_clean['P16']['trials']['t48']\n",
    "\n",
    "for trial in d_clean['P16']['trials'].values():\n",
    "    x = []\n",
    "    for point in trial['wrist12vel']:\n",
    "        if point > 0.05:\n",
    "            x.append(1)\n",
    "        else:\n",
    "            x.append(0)\n",
    "    trial['startmovement']= tuple(x)\n",
    "\n",
    "curve = np.array(d_clean['P16']['trials']['t22']['startmovement'])\n",
    "curve[200:400] = 0\n",
    "d_clean['P16']['trials']['t22']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P16']['trials']['t47']['startmovement'])\n",
    "curve[200:400] = 0\n",
    "d_clean['P16']['trials']['t47']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P 18\n",
    "\n",
    "- removed t12: drop in the data. Name: Roman_Occlusion_Cue_RightToLeft0005  \n",
    "- removed t15: drop. Name: Roman_Occlusion_NoCue_LeftToRight0002  \n",
    "- removed t19: drop. Name: Roman_Occlusion_NoCue_RightToLeft0000\n",
    "- removed t47: strange data at grasp. too rapid. Name: Roman_Visible_NoCue_RightToLeft0004"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P18'] = d['P18']\n",
    "\n",
    "del d_clean['P18']['trials']['t12']\n",
    "del d_clean['P18']['trials']['t15']\n",
    "del d_clean['P18']['trials']['t19']\n",
    "del d_clean['P18']['trials']['t47']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P19\n",
    "\n",
    "- removed t3: drops. Name: Roman_Occlusion_Cue_LeftToRight0002  \n",
    "- removed t5: drop. Name: Roman_Occlusion_Cue_LeftToRight0004  \n",
    "- removed t7: drop. Name: Roman_Occlusion_Cue_RightToLeft0000  \n",
    "- removed t12: drop. Name: Roman_Occlusion_Cue_RightToLeft0005  \n",
    "- removed t34: drop. Name: Roman_Visible_Cue_RightToLeft0003\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P19'] = d['P19']\n",
    "\n",
    "del d_clean['P19']['trials']['t3']\n",
    "del d_clean['P19']['trials']['t5']\n",
    "del d_clean['P19']['trials']['t7']\n",
    "del d_clean['P19']['trials']['t12']\n",
    "del d_clean['P19']['trials']['t34']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P21\n",
    "\n",
    "- removed t22: drop. Name: Roman_Occlusion_NoCue_RightToLeft0004  \n",
    "- recalculated startmovement with wrist11vel for all trials"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P21'] = d['P21']\n",
    "\n",
    "del d_clean['P21']['trials']['t22']\n",
    "\n",
    "for trial in d_clean['P21']['trials'].values():\n",
    "    sm = []\n",
    "    for point in trial['wrist11vel']:\n",
    "        if point < 0.05:\n",
    "            sm.append(0)\n",
    "        else:\n",
    "            sm.append(1)\n",
    "        trial['startmovement'] = tuple(sm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P22\n",
    "\n",
    "- removed t47: data drop. Name: Roman_Visible_NoCue_RightToLeft0004  \n",
    "\n",
    "\n",
    "- edited t27: startmovement. Name: Roman_Visible_Cue_LeftToRight0003  \n",
    "- edited t28: startmovement. Name: Roman_Visible_Cue_LeftToRight0004  \n",
    "- edited t33: startmovement. Name: Roman_Visible_Cue_RightToLeft0002  \n",
    "- edited t35: startmovement. Name: Roman_Visible_Cue_RightToLeft0005  \n",
    "- edited t36: startmovement. Name: Roman_Visible_Cue_RightToLeft0006  \n",
    "- edited t39: startmovement. Name: Roman_Visible_NoCue_LeftToRight0002  \n",
    "- edited t42: startmovement. Name: Roman_Visible_NoCue_LeftToRight0005  \n",
    "- edited t48: startmovement. Name: Roman_Visible_NoCue_RightToLeft0005  \n",
    "\n",
    "This participant made slight wist movements before the tone "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P22'] = d['P22']\n",
    "\n",
    "del d_clean['P22']['trials']['t47']\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t27']['startmovement'])\n",
    "curve[600:890] = 0\n",
    "d_clean['P22']['trials']['t27']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t28']['startmovement'])\n",
    "curve[600:902] = 0\n",
    "d_clean['P22']['trials']['t28']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t33']['startmovement'])\n",
    "curve[500:600] = 0\n",
    "d_clean['P22']['trials']['t33']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t35']['startmovement'])\n",
    "curve[500:870] = 0\n",
    "d_clean['P22']['trials']['t35']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t36']['startmovement'])\n",
    "curve[700:870] = 0\n",
    "d_clean['P22']['trials']['t36']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t39']['startmovement'])\n",
    "curve[600:870] = 0\n",
    "d_clean['P22']['trials']['t39']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t42']['startmovement'])\n",
    "curve[500:600] = 0\n",
    "d_clean['P22']['trials']['t42']['startmovement'] = tuple(curve)\n",
    "\n",
    "curve = np.array(d_clean['P22']['trials']['t48']['startmovement'])\n",
    "curve[400:700] = 0\n",
    "d_clean['P22']['trials']['t48']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P23\n",
    "\n",
    "- removed t3: drop in the data. Name: Roman_Occlusion_Cue_LeftToRight0002  \n",
    "- removed t31: early reach. Name: Roman_Visible_Cue_RightToLeft0000  \n",
    "- swapped index8 for index7 in t33. index8 only produced noisy data just in this trial. Name: Roman_Visible_Cue_RightToLeft0002  \n",
    "- removed t35: lag at the end. Name: Roman_Visible_Cue_RightToLeft0005  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P23'] = d['P23']\n",
    "\n",
    "del d_clean['P23']['trials']['t3']\n",
    "del d_clean['P23']['trials']['t31']\n",
    "del d_clean['P23']['trials']['t35']\n",
    "\n",
    "d_clean['P23']['trials']['t33']['index8x'] = d['P23']['trials']['t33']['index7x']\n",
    "d_clean['P23']['trials']['t33']['index8y'] = d['P23']['trials']['t33']['index7y']\n",
    "d_clean['P23']['trials']['t33']['index8z'] = d['P23']['trials']['t33']['index7z']\n",
    "\n",
    "d_clean['P23']['trials']['t33']['index7x'] = d['P23']['trials']['t33']['index8x']\n",
    "d_clean['P23']['trials']['t33']['index7y'] = d['P23']['trials']['t33']['index8y']\n",
    "d_clean['P23']['trials']['t33']['index7z'] = d['P23']['trials']['t33']['index8z']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P25\n",
    "\n",
    "- removed t12: data drop. Name: Roman_Occlusion_Cue_RightToLeft0005  \n",
    "- removed t14: data drop. Name: Roman_Occlusion_NoCue_LeftToRight0001  \n",
    "- removed t16: data drop + lag. Name: Roman_Occlusion_NoCue_LeftToRight0003  \n",
    "- removed t27: lag. Name: Roman_Visible_Cue_LeftToRight0003  \n",
    "- removed t39: data drop. Name: Roman_Visible_NoCue_LeftToRight0002  \n",
    "\n",
    "- recalclated all trials' startmovement with wrist11vel  \n",
    "- edited t20: startmovement. Name: Roman_Occlusion_NoCue_RightToLeft0002  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P25'] = d['P25']\n",
    "\n",
    "del d_clean['P25']['trials']['t12']\n",
    "del d_clean['P25']['trials']['t14']\n",
    "del d_clean['P25']['trials']['t16']\n",
    "del d_clean['P25']['trials']['t27']\n",
    "del d_clean['P25']['trials']['t39']\n",
    "\n",
    "for trial in d_clean['P25']['trials'].values():\n",
    "    velcurve = np.array(trial['wrist11vel'])\n",
    "    sm = velcurve >= 0.05\n",
    "    trial['startmovement'] = tuple(sm)\n",
    "\n",
    "curve = np.array(d_clean['P25']['trials']['t20']['startmovement'])\n",
    "curve[400:600] = 0\n",
    "d_clean['P25']['trials']['t20']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P26\n",
    "\n",
    "- removed t1: drop in the data. Name: Roman_Occlusion_Cue_LeftToRight0000  \n",
    "- removed t23: microsof security message during trial. Name: Roman_Occlusion_NoCue_RightToLeft0005"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P26'] = d['P26']\n",
    "\n",
    "del d_clean['P26']['trials']['t1']\n",
    "del d_clean['P26']['trials']['t23']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P29\n",
    "\n",
    "- removed t19: drop in the data. Name: Roman_Occlusion_NoCue_RightToLeft0000  \n",
    "- edited t24: startmovement. Name: Roman_Visible_Cue_LeftToRight0000  \n",
    "\n",
    "\n",
    "- t14 looks suspicious: very fast reach, mabe remove. Name: Roman_Occlusion_NoCue_LeftToRight0001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P29'] = d['P29']\n",
    "\n",
    "del d_clean['P29']['trials']['t19']\n",
    "\n",
    "curve = np.array(d_clean['P29']['trials']['t24']['startmovement'])\n",
    "curve[200:400] = 0\n",
    "d_clean['P29']['trials']['t24']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P30\n",
    "\n",
    "- removed t43: lag. Name: Roman_Visible_NoCue_RightToLeft0000 \n",
    "\n",
    "\n",
    "This participant did not do the task well: did not pursue the object when occluded, saccades to the centre of the screen, wait there and grasp in the centre. Mean grasp accuracy on the occluded trials is 11 cm. Exclude.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# d_clean['P30'] = d['P30']\n",
    "# del d_clean['P30']['trials']['t43']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P31\n",
    "\n",
    "- edited t1: startmovement. Name: Roman_Occlusion_Cue_LeftToRight0000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P31'] = d['P31']\n",
    "\n",
    "curve = np.array(d_clean['P31']['trials']['t1']['startmovement'])\n",
    "curve[400:600] = 0\n",
    "d_clean['P31']['trials']['t1']['startmovement'] = tuple(curve)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P34\n",
    "\n",
    "- removed t25: lag. Name: Roman_Visible_Cue_LeftToRight0000    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P34'] = d['P34']\n",
    "\n",
    "del d_clean['P34']['trials']['t25']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P35\n",
    "\n",
    "- removed t2: sensor came off. Name: Roman_Occlusion_Cue_LeftToRight0001  \n",
    "- removed t11: data drop. Name: Roman_Occlusion_Cue_RightToLeft0004  \n",
    "- removed t15: data drop. Name: Roman_Occlusion_NoCue_LeftToRight0003  \n",
    "- removed t24: lag+data drop. Name: Roman_Visible_Cue_LeftToRight0001  \n",
    "- removed t31: windows msg. Name: Roman_Visible_Cue_RightToLeft0003  \n",
    "\n",
    "This participant got tired at the end of session. Stopped the third block half-way. Remove in the future"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# d_clean['P35'] = d['P35']\n",
    "\n",
    "# del d_clean['P35']['trials']['t2']\n",
    "# del d_clean['P35']['trials']['t11']\n",
    "# del d_clean['P35']['trials']['t15']\n",
    "# del d_clean['P35']['trials']['t24']\n",
    "# del d_clean['P35']['trials']['t31']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P36\n",
    "\n",
    "- removed t2: drop in the data. Name: Roman_Occlusion_Cue_LeftToRight0001  \n",
    "- removed t11: drop. Name: Roman_Occlusion_Cue_RightToLeft0004  \n",
    "- removed t17: drop. Name: Roman_Occlusion_NoCue_LeftToRight0004  \n",
    "- removed t47: drop. Name: Roman_Visible_NoCue_RightToLeft0004  \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P36'] = d['P36']\n",
    "\n",
    "del d_clean['P36']['trials']['t2']\n",
    "del d_clean['P36']['trials']['t11']\n",
    "del d_clean['P36']['trials']['t17']\n",
    "del d_clean['P36']['trials']['t47']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# P37\n",
    "\n",
    "- removed t1: early reach. Name: Roman_Occlusion_Cue_LeftToRight0000\n",
    "- removed t10: data drops. name: Roman_Occlusion_Cue_RightToLeft0003  \n",
    "- removed t11: no data. Name: Roman_Occlusion_Cue_RightToLeft0004  \n",
    "- removed t29: data drop. Name: Roman_Visible_Cue_LeftToRight0006\n",
    "- removed t37: screen lag. Name: Roman_Visible_NoCue_LeftToRight0000  \n",
    "- removed t39: data dop. Name: Roman_Visible_NoCue_LeftToRight0002  \n",
    "- removed t47: no data collection. Name: Roman_Visible_NoCue_RightToLeft0004  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "d_clean['P37'] = d['P37']\n",
    "\n",
    "del d_clean['P37']['trials']['t1']\n",
    "del d_clean['P37']['trials']['t10']\n",
    "del d_clean['P37']['trials']['t11']\n",
    "del d_clean['P37']['trials']['t29']\n",
    "del d_clean['P37']['trials']['t37']\n",
    "del d_clean['P37']['trials']['t39']\n",
    "del d_clean['P37']['trials']['t47']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Roman_Occlusion_Cue_LeftToRight0001\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAD8CAYAAACb4nSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nqOiLiLiIir6IiIuo6IuIuIiKvoiIi/x/AKtq2ZuLGL8AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe1826422d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "trial = d['P37']['trials']['t2']\n",
    "\n",
    "print trial['name']\n",
    "\n",
    "plt.plot(trial['index8x'])\n",
    "plt.axvline(400, color = 'red')\n",
    "plt.axvline(600, color = 'red')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAARMAAAEJCAYAAAC35MNFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nEyGELSSYCCFsIcFECGGL/w+7bzjh3HrqXQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7fe183179e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean   = 0.99263 +- 0.801975106284\n",
      "Median = 0.75375\n",
      "\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "(0.99262999999999957, 0.8019751062844771)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "check_visible(d_clean, 'P37', 'index8', 'thumb9')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The data will be overwritten. Continue? Yes/noYes\n"
     ]
    }
   ],
   "source": [
    "manual_save_data(d_clean, 'clean_data.pkl')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}