catFR1_analyses.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2b5aaa88-e071-4d63-962f-6f800e7c37ee",
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "import cmlreaders as cml\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "import sys\n",
    "sys.path.append('/home1/djhalp/pybeh_pd')\n",
    "import pybeh_pd as pb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e2f83696-f626-4a38-a423-b513cdf549b3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/cmlreaders/readers/readers.py:250: UserWarning: Changing events session field from 0 to 1 to match data index.\n",
      "  warnings.warn(f'Changing events session field from {df[\"session\"].unique()[0]} ' +\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/cmlreaders/readers/readers.py:250: UserWarning: Changing events session field from 1 to 0 to match data index.\n",
      "  warnings.warn(f'Changing events session field from {df[\"session\"].unique()[0]} ' +\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/cmlreaders/readers/readers.py:250: UserWarning: Changing events session field from 10 to 3 to match data index.\n",
      "  warnings.warn(f'Changing events session field from {df[\"session\"].unique()[0]} ' +\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/cmlreaders/readers/readers.py:250: UserWarning: Changing events session field from 1 to 0 to match data index.\n",
      "  warnings.warn(f'Changing events session field from {df[\"session\"].unique()[0]} ' +\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/cmlreaders/readers/readers.py:250: UserWarning: Changing events session field from 2 to 1 to match data index.\n",
      "  warnings.warn(f'Changing events session field from {df[\"session\"].unique()[0]} ' +\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/cmlreaders/readers/readers.py:250: UserWarning: Changing events session field from 3 to 2 to match data index.\n",
      "  warnings.warn(f'Changing events session field from {df[\"session\"].unique()[0]} ' +\n"
     ]
    }
   ],
   "source": [
    "all_catfr1_events = cml.CMLReader.load_events(experiments=[\"catFR1\"], data_type='task_events')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "d45a14bd-6d3f-412e-88dc-8f4d5b45caa5",
   "metadata": {},
   "outputs": [],
   "source": [
    "events = all_catfr1_events.query('type == [\"WORD\", \"REC_WORD\"]')\n",
    "events = events[events['list'] > 0] # removing practice lists"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "01c9d906-b8bb-4e70-b26c-049376023df0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#removing subjects who were presented with the same word multiple times (probably repeated data)\n",
    "word_evs = events.query('type == \"WORD\"')\n",
    "n_word_pres = word_evs.groupby(['subject', 'session', 'list', 'item_num'], as_index=False).size()\n",
    "bad_lists = n_word_pres.query('size > 2')\n",
    "bad_lists[['subject', 'session', 'list']].drop_duplicates()\n",
    "bad_subs = bad_lists.subject.unique().tolist()\n",
    "\n",
    "#drop subjects without 12 items per list\n",
    "n_list_pres = word_evs.groupby(['subject', 'session', 'list'], as_index=False).size()\n",
    "bad_lists = n_list_pres.query('size != 12')\n",
    "bad_subs += bad_lists.subject.unique().tolist()\n",
    "\n",
    "#drop subjects with no recalls\n",
    "n_types = events.groupby('subject', as_index=False).agg({'type': 'nunique'})\n",
    "no_recall_subs = n_types.query('type != 2')\n",
    "bad_subs += no_recall_subs.subject.unique().tolist()\n",
    "\n",
    "#drop subjects with messed up recalls\n",
    "rec_evs = events.query('type == \"REC_WORD\"')\n",
    "joint_evs = word_evs.merge(rec_evs[['subject', 'session', 'list', 'category', 'item_name']], how='left', indicator=True)\n",
    "bad_subs += joint_evs.query('(recalled == 1 and _merge != \"both\") or (recalled == 0 and _merge == \"both\")').subject.unique().tolist()\n",
    "\n",
    "#drop subjects with rare_items\n",
    "item_sub_df = word_evs.groupby(['item_name'], as_index=False).agg({'subject': 'count'})\n",
    "rare_items = item_sub_df.query('subject < 20').item_name.unique()\n",
    "bad_subs += word_evs.query('item_name in @rare_items').subject.unique().tolist()\n",
    "\n",
    "events.query('subject not in @bad_subs', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "64bed631-650f-404c-86f5-33f1b9bd74a0",
   "metadata": {},
   "outputs": [],
   "source": [
    "word_evs = events.query('type == \"WORD\"')\n",
    "rec_evs = events.query('type == \"REC_WORD\"')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "158afea8-5b6d-4403-844d-f12e6d0068e4",
   "metadata": {},
   "outputs": [],
   "source": [
    "spc_sess_df = word_evs.groupby(['subject', 'session', 'serialpos'], as_index=False).agg({'recalled': 'mean'})\n",
    "spc_df = spc_sess_df.groupby(['subject', 'serialpos'], as_index=False).agg({'recalled': 'mean'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8cbb6fe5-e30f-4ff9-bd62-b055e772972a",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n",
      "  self._figure.tight_layout(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x2b3f11d36190>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.relplot(x='serialpos', y='recalled', data=spc_df, kind='line')\n",
    "g.set(xlabel='Serial Position', ylabel='P(recalled)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b71aa237-9cd4-429f-ba3c-1c5ee9b3c3e9",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_88677/1371011759.py:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  rec_evs['outpos'] = rec_evs.groupby(['subject', 'session', 'list']).cumcount()\n"
     ]
    }
   ],
   "source": [
    "rec_evs['outpos'] = rec_evs.groupby(['subject', 'session', 'list']).cumcount()\n",
    "n_list_df = word_evs.groupby(['subject', 'session']).agg({'list': 'nunique'}).reset_index().rename(columns={'list': 'n_lists'})\n",
    "first_recall_sess_df = rec_evs.query('outpos == 0 and serialpos != -999')\n",
    "#make sure there are 0s in serialpos that never get recalled first\n",
    "first_recall_sess_df = first_recall_sess_df.groupby(\n",
    "    ['subject', 'session', 'serialpos']).agg(\n",
    "    {'recalled': 'count'}).unstack(fill_value=0).stack().reset_index()\n",
    "first_recall_sess_df = first_recall_sess_df.merge(n_list_df)\n",
    "first_recall_sess_df['pfr'] = first_recall_sess_df['recalled'] / first_recall_sess_df['n_lists']\n",
    "pfr_df = first_recall_sess_df.groupby(['subject', 'serialpos']).agg({'pfr': 'mean'}).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "1621287c-dac8-444b-b5ec-267849473be6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n",
      "  self._figure.tight_layout(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x2b3f1d450cd0>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.relplot(x='serialpos', y='pfr', data=pfr_df, kind='line')\n",
    "g.set(xlabel='Serial Position', ylabel='P(recalled first)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "35981de7-6ec1-4f9d-bac8-abc2306bf935",
   "metadata": {},
   "outputs": [],
   "source": [
    "spc_sess_df = word_evs.groupby(['subject', 'session', 'serialpos'], as_index=False).agg({'recalled': 'mean'})\n",
    "spc_df = spc_sess_df.groupby(['subject', 'serialpos'], as_index=False).agg({'recalled': 'mean'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "bd6850af-c2b7-4f65-a488-43fa5009d3c2",
   "metadata": {},
   "outputs": [],
   "source": [
    "pb_crp = events.groupby('subject').apply(pb.pd_crp, \n",
    "                                itemno_column='item_num').reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "95c8809f-367a-4805-ba1c-6621ea84c3cd",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/seaborn/algorithms.py:98: RuntimeWarning: Mean of empty slice\n",
      "  boot_dist.append(f(*sample, **func_kwargs))\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/numpy/lib/nanfunctions.py:1556: RuntimeWarning: All-NaN slice encountered\n",
      "  return function_base._ureduce(a,\n",
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n",
      "  self._figure.tight_layout(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x2b3f10f5a110>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.catplot(x='lag', y='prob', kind='point', \n",
    "                  data=pb_crp.query('lag <= 5 and lag >= -5')\n",
    "           )\n",
    "g.set(xlabel='Lag', ylabel='P(transition)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "ab2f331d-7071-435c-bf00-c5740a481177",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>subject</th>\n",
       "      <th>temp_fact</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>R1004D</td>\n",
       "      <td>0.677778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>R1013E</td>\n",
       "      <td>0.614322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>R1015J</td>\n",
       "      <td>0.584127</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>R1016M</td>\n",
       "      <td>0.598297</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>R1021D</td>\n",
       "      <td>0.630544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>254</th>\n",
       "      <td>R1680T</td>\n",
       "      <td>0.611158</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>255</th>\n",
       "      <td>R1681J</td>\n",
       "      <td>0.545422</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256</th>\n",
       "      <td>R1683A</td>\n",
       "      <td>0.581063</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>257</th>\n",
       "      <td>R1689J</td>\n",
       "      <td>0.863415</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>258</th>\n",
       "      <td>R1696L</td>\n",
       "      <td>0.454117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>259 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    subject  temp_fact\n",
       "0    R1004D   0.677778\n",
       "1    R1013E   0.614322\n",
       "2    R1015J   0.584127\n",
       "3    R1016M   0.598297\n",
       "4    R1021D   0.630544\n",
       "..      ...        ...\n",
       "254  R1680T   0.611158\n",
       "255  R1681J   0.545422\n",
       "256  R1683A   0.581063\n",
       "257  R1689J   0.863415\n",
       "258  R1696L   0.454117\n",
       "\n",
       "[259 rows x 2 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pb_temp_fact = events.groupby(\"subject\").apply(pb.pd_temp_fact,\n",
    "                                                   itemno_column='item_num', \n",
    "                                                   list_index=['subject', 'session', 'list'], \n",
    "                                                   pres_type=\"WORD\", rec_type=\"REC_WORD\", type_column='type'\n",
    "                                                   ).to_frame(name='temp_fact').reset_index()\n",
    "\n",
    "pb_temp_fact"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "49b41ad8-ecc0-4469-86e3-8a6de4813af6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/global/miniconda/py310_23.1.0-1/envs/workshop_311/lib/python3.11/site-packages/seaborn/axisgrid.py:118: UserWarning: The figure layout has changed to tight\n",
      "  self._figure.tight_layout(*args, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.FacetGrid at 0x2b3f1b5e2650>"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 500x500 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "g = sns.displot(x='temp_fact', data=pb_temp_fact)\n",
    "g.set(xlabel='Temporal Factor Score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "5435512c-6db7-470b-81f3-a20832decc12",
   "metadata": {},
   "outputs": [],
   "source": [
    "# replace item nums so they are consistent across subjects\n",
    "item_col = 'item_name'\n",
    "item_num_col = 'item_num'\n",
    "item_num_df = events.query('type == \"WORD\"').drop_duplicates(subset=item_col, ignore_index=True\n",
    "                                    )[item_col].reset_index().rename(columns={'index': item_num_col})\n",
    "item_num_df['item_num'] = item_num_df['item_num'] + 1\n",
    "events_new = events.merge(item_num_df, on=item_col, suffixes=('', '_new'), \n",
    "                          how='left', sort=False)#.sort_values('mstime')\n",
    "events_new.fillna({'item_num_new': -999}, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "2c9d8001-c746-4238-bf6d-4c945946d3dc",
   "metadata": {},
   "outputs": [
    {
     "ename": "ModuleNotFoundError",
     "evalue": "No module named 'gensim'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[20], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mgensim\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdownloader\u001b[39;00m\n\u001b[1;32m      2\u001b[0m word2vec_vectors \u001b[38;5;241m=\u001b[39m gensim\u001b[38;5;241m.\u001b[39mdownloader\u001b[38;5;241m.\u001b[39mload(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mword2vec-google-news-300\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
      "\u001b[0;31mModuleNotFoundError\u001b[0m: No module named 'gensim'"
     ]
    }
   ],
   "source": [
    "import gensim.downloader\n",
    "word2vec_vectors = gensim.downloader.load('word2vec-google-news-300')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b8473ce5-a5b2-46a8-bde0-bae2f69bc3a5",
   "metadata": {},
   "outputs": [],
   "source": [
    "def word_similarity(df, col1, col2, keyed_vector=None):\n",
    "    try:\n",
    "        return keyed_vector.similarity(df[col1].lower(), df[col2].lower())\n",
    "    except:\n",
    "        return np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "25b1e07a-e248-4466-8753-9b8dbd5b4c4c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# get all pairs of items\n",
    "items = item_num_df[item_col].values\n",
    "sem_sim_df = pd.MultiIndex.from_product([items, items], names=['wordpool_item_1', 'wordpool_item_2']).to_frame(index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9ddba187-bc99-4b6e-a0a0-5f6f4238c3a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "sem_sim_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "443f0751-c0ea-423d-860c-26de5889942c",
   "metadata": {},
   "outputs": [],
   "source": [
    "# compute similarity of all pairs\n",
    "sem_sim_df['similarity'] = sem_sim_df.apply(word_similarity, \n",
    "               axis=1, col1='wordpool_item_1', \n",
    "               col2='wordpool_item_2', \n",
    "               keyed_vector=word2vec_vectors)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5f9e349e-c11c-450d-9abb-b51bb784ff29",
   "metadata": {},
   "outputs": [],
   "source": [
    "sem_sim_df.to_csv('/scratch/djh/catfr1_sem_sim_df.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "096eb21a-8adf-4248-92e5-ee2b0cada796",
   "metadata": {},
   "outputs": [],
   "source": [
    "sem_sim_df = pd.read_csv('/scratch/djh/catfr1_sem_sim_df.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a5adec3-1b8f-4df0-a7cf-74f9416dc44e",
   "metadata": {},
   "outputs": [],
   "source": [
    "sem_sim_num_df = sem_sim_df.merge(\n",
    "    item_num_df, left_on='wordpool_item_1', right_on=item_col).merge(\n",
    "    item_num_df, left_on='wordpool_item_2', right_on=item_col, suffixes=('_1', '_2')).drop(columns=['wordpool_item_1', 'wordpool_item_2'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7414a870-0597-4087-8a16-a40f18763828",
   "metadata": {},
   "outputs": [],
   "source": [
    "sem_sims = pd.pivot_table(sem_sim_num_df, index='item_num_1', columns='item_num_2', values='similarity', dropna=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e585f7d7-94ee-4962-9cc1-f8144fa81911",
   "metadata": {},
   "outputs": [],
   "source": [
    "events_new.query('item_num_new == -999')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a1183727-8e80-4fab-bd37-27e76ca89cc6",
   "metadata": {},
   "outputs": [],
   "source": [
    "pb_sem_crp = events_new.groupby('subject').apply(pb.pd_sem_crp, \n",
    "                                itemno_column='item_num_new', sem_sims=sem_sims).reset_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1121a0d2-f6de-4d22-87d1-49f4a4fd8aa0",
   "metadata": {},
   "outputs": [],
   "source": [
    "g = sns.catplot(x='sem_bin', y='prob', kind='point', \n",
    "                  data=pb_sem_crp)\n",
    "g.set(xlabel='Similarity Bin', ylabel='P(transition)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "1962d276-facb-42f7-a0d5-7dce697ee5f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "#using average bin value\n",
    "pb_sem_crp['avg_sem_bin_mean'] = pb_sem_crp.groupby('sem_bin')['sem_bin_mean'].transform('mean')\n",
    "g = sns.relplot(x='avg_sem_bin_mean', y='prob', data=pb_sem_crp, kind='line')\n",
    "g.set(xlabel='Similarity', ylabel='P(transition)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "846bf074-a63a-43e4-a137-eafe89973f63",
   "metadata": {},
   "outputs": [],
   "source": [
    "pb_sem_dist_fact = events_new.groupby(\"subject\").apply(pb.pd_dist_fact,\n",
    "                                                       rec_itemnos=None, \n",
    "                                                       itemno_column='item_num_new', \n",
    "                                                       list_index=['subject', 'session', 'list'], \n",
    "                                                       dist_mat=sem_sims, \n",
    "                                                       sim_columns=None, \n",
    "                                                       is_similarity=True, \n",
    "                                                       dist_columns=None,\n",
    "                                                       skip_first_n=0,\n",
    "                                                       pres_type=\"WORD\", rec_type=\"REC_WORD\", type_column='type', ret_counts=False\n",
    "                                                       ).to_frame(name='sem_fact').reset_index()\n",
    "\n",
    "pb_sem_dist_fact"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "72431e8a-24a8-444c-b670-f8367317324d",
   "metadata": {},
   "outputs": [],
   "source": [
    "g = sns.displot(x='sem_fact', data=pb_sem_dist_fact)\n",
    "g.set(xlabel='Semantic Factor Score')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b4375a8d-8215-47cb-8b9b-c3b9e15b1e54",
   "metadata": {},
   "outputs": [],
   "source": [
    "rec_evs['repeat'] = rec_evs.duplicated(subset=['subject', 'session', 'list', 'item_name'])\n",
    "rec_evs['correct'] = (rec_evs['repeat'] == False) & (rec_evs['intrusion'] == 0)\n",
    "rec_evs['next_rectime'] = rec_evs.groupby(['subject', 'session', 'list']).rectime.shift(-1)\n",
    "rec_evs['next_correct'] = rec_evs.groupby(['subject', 'session', 'list']).correct.shift(-1)\n",
    "rec_evs['irt'] = rec_evs['next_rectime'] - rec_evs['rectime']\n",
    "rec_evs.loc[(rec_evs['correct'] == False) | (rec_evs['next_correct'] == False), 'irt'] = np.nan\n",
    "rec_evs['correct_outpos'] = rec_evs.groupby(['subject', 'session', 'list'])['correct'].cumsum()\n",
    "rec_evs['n_correct'] = rec_evs.groupby(['subject', 'session', 'list'])['correct_outpos'].transform('max')\n",
    "rec_evs['irt_s'] = rec_evs['irt'] / 1000\n",
    "rec_evs['next_serialpos'] = rec_evs.groupby(['subject', 'session', 'list']).serialpos.shift(-1)\n",
    "rec_evs['transition'] = rec_evs['next_serialpos'] - rec_evs['serialpos']\n",
    "rec_evs.loc[(rec_evs['correct'] == False) | (rec_evs['next_correct'] == False), 'transition'] = np.nan\n",
    "rec_evs['next_item_name'] = rec_evs.groupby(['subject', 'session', 'list']).item_name.shift(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dbfc08fa-1186-46ee-9554-ba40546207da",
   "metadata": {},
   "outputs": [],
   "source": [
    "rec_evs = rec_evs.merge(sem_sim_df, how='left', left_on=['item_name', 'next_item_name'], \n",
    "                        right_on=['wordpool_item_1', 'wordpool_item_2'])\n",
    "rec_evs.loc[(rec_evs['correct'] == False) | (rec_evs['next_correct'] == False), 'similarity'] = np.nan"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ef0638de-3795-4451-acbf-9511d80eae1b",
   "metadata": {},
   "outputs": [],
   "source": [
    "rec_evs['similarity_bin'] = pd.qcut(rec_evs.similarity, 10, labels=np.arange(0, 10)).astype(float)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4df56744-7de2-46c2-aaa3-35f5efde1bd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "irt_sess_df = rec_evs.groupby(['subject', 'session', 'correct_outpos', 'n_correct'], as_index=False).agg({'irt_s': 'mean'})\n",
    "irt_df = irt_sess_df.groupby(['subject', 'correct_outpos', 'n_correct'], as_index=False).agg({'irt_s': 'mean'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c319e691-cf3e-42c4-99a4-89860719ffd6",
   "metadata": {},
   "outputs": [],
   "source": [
    "g = sns.catplot(x='correct_outpos', y='irt_s', hue='n_correct', data=irt_df.query('n_correct > 1 and n_correct < 10'), kind='point')\n",
    "g.set(xlabel='Output Position', ylabel='IRT (s)', title='catFR1')\n",
    "g._legend.set_title('Total Recalled')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "45b28204-d665-4e34-8eb3-247d7bebac9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "crl_sess_df = rec_evs.groupby(['subject', 'session', 'transition'], as_index=False).agg({'irt_s': 'mean'})\n",
    "crl_df = crl_sess_df.groupby(['subject', 'transition'], as_index=False).agg({'irt_s': 'mean'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a57ea306-21bc-4ed2-83d1-959877c12dc0",
   "metadata": {},
   "outputs": [],
   "source": [
    "g = sns.catplot(x='transition', y='irt_s', data=crl_df.query('abs(transition) < 6'),\n",
    "                order=np.arange(-5, 6),\n",
    "                kind='point')\n",
    "g.set(xlabel='Transition', ylabel='IRT (s)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "27089301-101b-4eb1-9b70-efa7b5352bc5",
   "metadata": {},
   "outputs": [],
   "source": [
    "sem_crl_sess_df = rec_evs.groupby(['subject', 'session', 'similarity_bin'], as_index=False).agg({'irt_s': 'mean'})\n",
    "sem_crl_df = sem_crl_sess_df.groupby(['subject', 'similarity_bin'], as_index=False).agg({'irt_s': 'mean'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ec34da1f-f9cd-48e6-a80f-723241b8afda",
   "metadata": {},
   "outputs": [],
   "source": [
    "g = sns.catplot(x='similarity_bin', y='irt_s', data=sem_crl_df,\n",
    "                kind='point')\n",
    "g.set(xlabel='Similarity', ylabel='IRT (s)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "89b55b3f-14a2-4440-9103-5fc73481fc05",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "workshop_311",
   "language": "python",
   "name": "workshop_311"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}