diff --git a/content/.jupyter_cache/executed/b1c4fd30be37983707d37bf89e60a186/base.ipynb b/content/.jupyter_cache/executed/b1c4fd30be37983707d37bf89e60a186/base.ipynb
new file mode 100644
index 0000000..ab7628a
--- /dev/null
+++ b/content/.jupyter_cache/executed/b1c4fd30be37983707d37bf89e60a186/base.ipynb
@@ -0,0 +1,2789 @@
+{
+ "cells": [
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "id": "87742e2d",
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import IJulia\n",
+ "\n",
+ "# The julia kernel has built in support for Revise.jl, so this is the \n",
+ "# recommended approach for long-running sessions:\n",
+ "# https://github.com/JuliaLang/IJulia.jl/blob/9b10fa9b879574bbf720f5285029e07758e50a5e/src/kernel.jl#L46-L51\n",
+ "\n",
+ "# Users should enable revise within .julia/config/startup_ijulia.jl:\n",
+ "# https://timholy.github.io/Revise.jl/stable/config/#Using-Revise-automatically-within-Jupyter/IJulia-1\n",
+ "\n",
+ "# clear console history\n",
+ "IJulia.clear_history()\n",
+ "\n",
+ "fig_width = 7\n",
+ "fig_height = 5\n",
+ "fig_format = :retina\n",
+ "fig_dpi = 96\n",
+ "\n",
+ "# no retina format type, use svg for high quality type/marks\n",
+ "if fig_format == :retina\n",
+ " fig_format = :svg\n",
+ "elseif fig_format == :pdf\n",
+ " fig_dpi = 96\n",
+ " # Enable PDF support for IJulia\n",
+ " IJulia.register_mime(MIME(\"application/pdf\"))\n",
+ "end\n",
+ "\n",
+ "# convert inches to pixels\n",
+ "fig_width = fig_width * fig_dpi\n",
+ "fig_height = fig_height * fig_dpi\n",
+ "\n",
+ "# Intialize Plots w/ default fig width/height\n",
+ "try\n",
+ " import Plots\n",
+ "\n",
+ " # Plots.jl doesn't support PDF output for versions < 1.28.1\n",
+ " # so use png (if the DPI remains the default of 300 then set to 96)\n",
+ " if (Plots._current_plots_version < v\"1.28.1\") & (fig_format == :pdf)\n",
+ " Plots.gr(size=(fig_width, fig_height), fmt = :png, dpi = fig_dpi)\n",
+ " else\n",
+ " Plots.gr(size=(fig_width, fig_height), fmt = fig_format, dpi = fig_dpi)\n",
+ " end\n",
+ "catch e\n",
+ " # @warn \"Plots init\" exception=(e, catch_backtrace())\n",
+ "end\n",
+ "\n",
+ "# Initialize CairoMakie with default fig width/height\n",
+ "try\n",
+ " import CairoMakie\n",
+ "\n",
+ " # CairoMakie's display() in PDF format opens an interactive window\n",
+ " # instead of saving to the ipynb file, so we don't do that.\n",
+ " # https://github.com/quarto-dev/quarto-cli/issues/7548\n",
+ " if fig_format == :pdf\n",
+ " CairoMakie.activate!(type = \"png\")\n",
+ " else\n",
+ " CairoMakie.activate!(type = string(fig_format))\n",
+ " end\n",
+ " CairoMakie.update_theme!(resolution=(fig_width, fig_height))\n",
+ "catch e\n",
+ " # @warn \"CairoMakie init\" exception=(e, catch_backtrace())\n",
+ "end\n",
+ " \n",
+ "# Set run_path if specified\n",
+ "try\n",
+ " run_path = raw\"C:\\Users\\domma\\Dropbox\\Software\\CognitiveModels\\content\"\n",
+ " if !isempty(run_path)\n",
+ " cd(run_path)\n",
+ " end\n",
+ "catch e\n",
+ " @warn \"Run path init:\" exception=(e, catch_backtrace())\n",
+ "end\n",
+ "\n",
+ "\n",
+ "# emulate old Pkg.installed beahvior, see\n",
+ "# https://discourse.julialang.org/t/how-to-use-pkg-dependencies-instead-of-pkg-installed/36416/9\n",
+ "import Pkg\n",
+ "function isinstalled(pkg::String)\n",
+ " any(x -> x.name == pkg && x.is_direct_dep, values(Pkg.dependencies()))\n",
+ "end\n",
+ "\n",
+ "# ojs_define\n",
+ "if isinstalled(\"JSON\") && isinstalled(\"DataFrames\")\n",
+ " import JSON, DataFrames\n",
+ " global function ojs_define(; kwargs...)\n",
+ " convert(x) = x\n",
+ " convert(x::DataFrames.AbstractDataFrame) = Tables.rows(x)\n",
+ " content = Dict(\"contents\" => [Dict(\"name\" => k, \"value\" => convert(v)) for (k, v) in kwargs])\n",
+ " tag = \"\"\n",
+ " IJulia.display(MIME(\"text/html\"), tag)\n",
+ " end\n",
+ "elseif isinstalled(\"JSON\")\n",
+ " import JSON\n",
+ " global function ojs_define(; kwargs...)\n",
+ " content = Dict(\"contents\" => [Dict(\"name\" => k, \"value\" => v) for (k, v) in kwargs])\n",
+ " tag = \"\"\n",
+ " IJulia.display(MIME(\"text/html\"), tag)\n",
+ " end\n",
+ "else\n",
+ " global function ojs_define(; kwargs...)\n",
+ " @warn \"JSON package not available. Please install the JSON.jl package to use ojs_define.\"\n",
+ " end\n",
+ "end\n",
+ "\n",
+ "\n",
+ "# don't return kernel dependencies (b/c Revise should take care of dependencies)\n",
+ "nothing\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "id": "0832bae0",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "1 1 Speed 0.7 false Low 2 1 Speed 0.392 true Very Low 3 1 Speed 0.46 false Very Low 4 1 Speed 0.455 false Very Low 5 1 Speed 0.505 true Low 6 1 Speed 0.773 false High 7 1 Speed 0.39 false High 8 1 Speed 0.587 true Low 9 1 Speed 0.603 false Low 10 1 Speed 0.435 false High
\n",
+ "\n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ " \n",
+ "\n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ " \n"
+ ]
+ },
+ "execution_count": 4,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "function plot_distribution(df, title=\"Empirical Distribution of Data from Wagenmakers et al. (2018)\")\n",
+ " fig = Figure()\n",
+ " ax = Axis(fig[1, 1], title=title,\n",
+ " xlabel=\"RT (s)\",\n",
+ " ylabel=\"Distribution\",\n",
+ " yticksvisible=false,\n",
+ " xticksvisible=false,\n",
+ " yticklabelsvisible=false)\n",
+ " CairoMakie.density!(df[df.Condition .== \"Speed\", :RT], color=(\"#EF5350\", 0.7), label = \"Speed\")\n",
+ " CairoMakie.density!(df[df.Condition .== \"Accuracy\", :RT], color=(\"#66BB6A\", 0.7), label = \"Accuracy\")\n",
+ " CairoMakie.axislegend(\"Condition\"; position=:rt)\n",
+ " CairoMakie.ylims!(ax, (0, nothing))\n",
+ " return fig\n",
+ "end\n",
+ "\n",
+ "plot_distribution(df, \"Empirical Distribution of Data from Wagenmakers et al. (2018)\")"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "id": "ffb2b20a",
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\u001b[36m\u001b[1m┌ \u001b[22m\u001b[39m\u001b[36m\u001b[1mInfo: \u001b[22m\u001b[39mFound initial step size\n",
+ "\u001b[36m\u001b[1m└ \u001b[22m\u001b[39m ϵ = 0.00625\n"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\r",
+ "\u001b[32mSampling: 0%|█ | ETA: 0:01:00\u001b[39m"
+ ]
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\r",
+ "\u001b[32mSampling: 100%|█████████████████████████████████████████| Time: 0:00:01\u001b[39m\n"
+ ]
+ },
+ {
+ "data": {
+ "text/plain": [
+ "Quantiles\n",
+ " \u001b[1m parameters \u001b[0m \u001b[1m 2.5% \u001b[0m \u001b[1m 97.5% \u001b[0m\n",
+ " \u001b[90m Symbol \u001b[0m \u001b[90m Float64 \u001b[0m \u001b[90m Float64 \u001b[0m\n",
+ "\n",
+ " σ² 0.1651 0.1699\n",
+ " intercept 0.5072 0.5166\n",
+ " slope_accuracy 0.1327 0.1451\n"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "#| code-fold: false\n",
+ "\n",
+ "@model function model_linear(rt; condition=nothing)\n",
+ "\n",
+ " # Set priors on variance, intercept and effect of ISI\n",
+ " σ² ~ truncated(Normal(0, 1); lower=0)\n",
+ " intercept ~ truncated(Normal(0, 1); lower=0)\n",
+ " slope_accuracy ~ Normal(0, 0.5)\n",
+ "\n",
+ " for i in 1:length(rt)\n",
+ " μ = intercept + slope_accuracy * condition[i]\n",
+ " rt[i] ~ Normal(μ, σ²)\n",
+ " end\n",
+ "end\n",
+ "\n",
+ "\n",
+ "model = model_linear(df.RT, condition=df.Accuracy)\n",
+ "chain_linear = sample(model, NUTS(), 200)\n",
+ "\n",
+ "# Summary (95% CI)\n",
+ "quantile(chain_linear; q=[0.025, 0.975])"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "id": "be300190",
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "200×9561 Matrix{Float64}:\n",
+ " 0.848602 0.705962 0.331378 0.620355 … 0.8028 0.518335 0.636211\n",
+ " 0.353569 0.647323 0.474041 0.588218 0.873261 0.884756 0.349058\n",
+ " 0.566442 0.642284 0.200491 0.469634 0.710634 0.810325 0.557676\n",
+ " 0.588596 0.465895 0.758861 0.43862 0.733499 0.708981 0.899049\n",
+ " 0.417891 0.347546 0.664938 0.507277 0.761035 0.656961 0.943823\n",
+ " 0.578149 0.417844 0.503022 0.740248 … 0.619439 1.04465 0.706872\n",
+ " 0.621813 0.326415 0.446741 0.383701 0.678272 0.610847 0.95635\n",
+ " 0.509265 0.350711 0.892334 0.336679 0.673803 0.621157 0.478052\n",
+ " 0.374051 0.589038 0.23259 0.581075 0.978464 0.678663 0.592232\n",
+ " 0.959078 0.803121 0.789548 -0.033704 0.867528 0.82184 0.82722\n",
+ " 0.429074 0.431844 0.544975 0.394383 … 0.77922 0.517494 0.514502\n",
+ " 0.576759 0.551092 0.397767 0.360882 0.668173 0.494338 0.715017\n",
+ " 0.258516 0.652359 0.673459 0.527017 0.839784 0.55494 0.625319\n",
+ " ⋮ ⋱ ⋮\n",
+ " 0.354336 0.385163 0.457614 0.324024 0.858301 0.51403 0.662216\n",
+ " 0.236944 0.668239 0.479412 0.675953 0.852875 0.750858 0.733875\n",
+ " 0.746798 0.528858 0.402718 0.450647 … 0.632083 0.607853 0.871148\n",
+ " 0.770152 0.425221 0.44569 0.497566 0.659614 0.768911 0.804448\n",
+ " 0.452424 0.871695 0.684568 0.5451 0.499917 0.37564 0.82096\n",
+ " 0.382002 0.515305 0.466115 0.502877 0.643914 0.794465 0.733294\n",
+ " 0.701579 0.807031 0.570335 0.38965 0.9301 0.571509 0.709906\n",
+ " 0.535645 0.277891 0.433616 0.680988 … 0.579326 0.548402 0.558554\n",
+ " 0.848265 0.484902 0.700802 0.621703 0.67408 0.728538 0.703432\n",
+ " 0.427934 0.661744 0.20302 0.478689 0.684985 0.63059 0.418275\n",
+ " 0.838115 0.490248 0.47848 0.61049 0.747242 1.20735 0.496999\n",
+ " 0.57673 0.597777 0.49274 0.355549 0.587818 0.653531 0.578666"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "#| output: false\n",
+ "\n",
+ "pred = predict(model_linear([(missing) for i in 1:length(df.RT)], condition=df.Accuracy), chain_linear)\n",
+ "pred = Array(pred)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "id": "07909a8b",
+ "metadata": {
+ "fig-height": 7,
+ "fig-width": 10
+ },
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\u001b[33m\u001b[1m┌ \u001b[22m\u001b[39m\u001b[33m\u001b[1mWarning: \u001b[22m\u001b[39mFound `resolution` in the theme when creating a `Scene`. The `resolution` keyword for `Scene`s and `Figure`s has been deprecated. Use `Figure(; size = ...` or `Scene(; size = ...)` instead, which better reflects that this is a unitless size and not a pixel resolution. The key could also come from `set_theme!` calls or related theming functions.\n",
+ "\u001b[33m\u001b[1m└ \u001b[22m\u001b[39m\u001b[90m@ Makie C:\\Users\\domma\\.julia\\packages\\Makie\\VRavR\\src\\scenes.jl:220\u001b[39m\n"
+ ]
+ },
+ {
+ "data": {
+ "image/png": "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",
+ "image/svg+xml": [
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ " \n",
+ "\n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ "\n",
+ "\n",
+ " \n",
+ " \n"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "fig = plot_distribution(df, \"Predictions made by Linear Model\")\n",
+ "for i in 1:length(chain_linear)\n",
+ " lines!(Makie.KernelDensity.kde(pred[:, i]), color=ifelse(df.Accuracy[i] == 1, \"#388E3C\", \"#D32F2F\"), alpha=0.1)\n",
+ "end\n",
+ "fig"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Julia 1.10.2",
+ "language": "julia",
+ "name": "julia-1.10"
+ },
+ "language_info": {
+ "file_extension": ".jl",
+ "mimetype": "application/julia",
+ "name": "julia",
+ "version": "1.10.2"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
\ No newline at end of file
diff --git a/content/.jupyter_cache/global.db b/content/.jupyter_cache/global.db
index b65302c..815f79b 100644
Binary files a/content/.jupyter_cache/global.db and b/content/.jupyter_cache/global.db differ
diff --git a/content/.quarto/_freeze/4_rt/execute-results/html.json b/content/.quarto/_freeze/4_rt/execute-results/html.json
index 6969ef0..fddac1f 100644
--- a/content/.quarto/_freeze/4_rt/execute-results/html.json
+++ b/content/.quarto/_freeze/4_rt/execute-results/html.json
@@ -1,10 +1,10 @@
{
- "hash": "26c45ef47fe7e81fdb051963e24f2bf1",
+ "hash": "80e397fa7a24ec91634f00cf896c2e90",
"result": {
"engine": "jupyter",
- "markdown": "# Reaction Times\n\n\nThis repository contain the following vignettes:\n\n- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)\n- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)\n\n## Descriptive Models (ExGaussian, LogNormal, Wald)\n\n### Modelling RT with a Bayesian Linear Model\n\n#### The Data\n\n::: {#718393b9 .cell execution_count=1}\n``` {.julia .cell-code}\nusing Downloads, CSV, DataFrames\nusing Turing, Distributions\nusing CairoMakie\n\ndf = CSV.read(Downloads.download(\"https://raw.githubusercontent.com/RealityBending/DoggoNogo/main/study1/data/data_game.csv\"), DataFrame)\nfirst(df, 10)\n```\n\n::: {.cell-output .cell-output-display execution_count=2}\n```{=html}\n1 0.517 1.638 1 S002 S1 NA Positive 0 0 2 0.433 2.282 2 S002 S1 0.475 Positive 1 0 3 0.533 0.739 3 S002 S1 0.494 Negative 2 0 4 0.35 0.602 4 S002 S1 0.458 Positive 0 1 5 0.517 1.951 5 S002 S1 0.47 Negative 1 0 6 0.383 0.952 6 S002 S1 0.456 Positive 0 1 7 0.383 0.875 7 S002 S1 0.445 Positive 1 0 8 0.417 2.662 8 S002 S1 0.442 Positive 2 0 9 0.367 2.8 9 S002 S1 0.433 Positive 3 0 10 0.417 0.258 10 S002 S1 0.432 Positive 4 0
Quantiles\n parameters 2.5% 97.5% \n Symbol Float64 Float64 \n σ² 0.0662 0.0741\n intercept 0.3432 0.3485\n slope_isi -0.0215 -0.0188\n \n```\n:::\n\n:::\n:::\n\n\n::: {.callout-tip title=\"Code Tip\"}\nWe first initialize the model by passing the `RT` and `ISI` columns.\n:::\n\n#### Posterior Predictive Check\n\n::: {#c0d701ef .cell execution_count=3}\n``` {.julia .cell-code}\npred = predict(model_linear([(missing) for i in 1:length(df.RT)]; isi=df.ISI), chain_linear)\npred = Array(pred)\n```\n:::\n\n\n::: {#5a9bdace .cell fig-height='7' fig-width='10' execution_count=4}\n``` {.julia .cell-code}\nf = Figure()\nax = Axis(f[1, 1], title=\"Predicted Data by Linear Model\",\n xlabel=\"RT (s)\",\n ylabel=\"Distribution\",\n yticksvisible=false,\n xticksvisible=false,\n yticklabelsvisible=false)\n\nCairoMakie.density!(df.RT, color=\"grey\")\nfor i in 1:length(chain_linear)\n lines!(ax, Makie.KernelDensity.kde(pred[:, i]), color=\"orange\", alpha=0.1)\nend\nCairoMakie.ylims!(ax, (0, nothing))\nf\n```\n\n::: {.cell-output .cell-output-stderr}\n```\n┌ Warning: Found `resolution` in the theme when creating a `Scene`. The `resolution` keyword for `Scene`s and `Figure`s has been deprecated. Use `Figure(; size = ...` or `Scene(; size = ...)` instead, which better reflects that this is a unitless size and not a pixel resolution. The key could also come from `set_theme!` calls or related theming functions.\n└ @ Makie C:\\Users\\domma\\.julia\\packages\\Makie\\VRavR\\src\\scenes.jl:220\n```\n:::\n\n::: {.cell-output .cell-output-display execution_count=5}\n{}\n:::\n:::\n\n\n### The Problem with Linear Models\n\nReaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the \"best fitting\" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\\mu$ and standard deviation $\\sigma$) are not good descriptors of the data.\n\n\n\n> Linear models try to find the best fitting Normal distribution for the data. However, for reaction times, even the best fitting Normal distribution (in red) does not capture well the actual data (in grey).\n\nA popular mitigation method to account for the non-normality of RTs is to transform the data, using for instance the popular *log-transform*. \nHowever, this practice should be avoided as it leads to various issues, including loss of power and distorted results interpretation [@lo2015transform; @schramm2019reaction].\nInstead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.\n\n\n### Shifted LogNormal Models\n\nOne of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.\n\n\n### Wald\n\nMoe from statistical models that *describe* to models that *generate* RT-like data.\n\n### Generative Models (DDM)\n\nUse DDM as a case study to introduce generative models\n\n### Other Models (LBA, LNR)\n\n\n## Additional Resources\n\n- [**Lindelov's overview of RT models**](https://lindeloev.github.io/shiny-rt/): An absolute must-read.\n- [**De Boeck & Jeon (2019)**](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00102/full): A paper providing an overview of RT models.\n- [https://github.com/vasishth/bayescogsci](https://github.com/vasishth/bayescogsci)\n\n",
+ "markdown": "# Reaction Times\n\n\nThis repository contain the following vignettes:\n\n- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)\n- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)\n\n## The Data\n\nData from @wagenmakers2008diffusion - Experiment 1.\nWe excluded all trials with uninterpretable response time [see @theriault2024check] such as too fast response (<180 ms) and too slow response (>2 sec).\n\n::: {#0832bae0 .cell execution_count=1}\n``` {.julia .cell-code code-fold=\"false\"}\nusing Downloads, CSV, DataFrames\nusing Turing, Distributions, SequentialSamplingModels\nusing CairoMakie\n\ndf = CSV.read(Downloads.download(\"https://raw.githubusercontent.com/DominiqueMakowski/CognitiveModels/main/data/wagenmakers2008.csv\"), DataFrame)\nfirst(df, 10)\n```\n\n::: {.cell-output .cell-output-display execution_count=2}\n```{=html}\n1 1 Speed 0.7 false Low 2 1 Speed 0.392 true Very Low 3 1 Speed 0.46 false Very Low 4 1 Speed 0.455 false Very Low 5 1 Speed 0.505 true Low 6 1 Speed 0.773 false High 7 1 Speed 0.39 false High 8 1 Speed 0.587 true Low 9 1 Speed 0.603 false Low 10 1 Speed 0.435 false High
Quantiles\n parameters 2.5% 97.5% \n Symbol Float64 Float64 \n σ² 0.1651 0.1699\n intercept 0.5072 0.5166\n slope_accuracy 0.1327 0.1451\n \n```\n:::\n\n:::\n:::\n\n\nThe effect of Condition is significant, people are on average slower (higher RT) when condition is `\"Accuracy\"`.\nBut is our model good?\n\n#### Posterior Predictive Check\n\n::: {#be300190 .cell execution_count=5}\n``` {.julia .cell-code}\npred = predict(model_linear([(missing) for i in 1:length(df.RT)], condition=df.Accuracy), chain_linear)\npred = Array(pred)\n```\n:::\n\n\n::: {#07909a8b .cell fig-height='7' fig-width='10' execution_count=6}\n``` {.julia .cell-code}\nfig = plot_distribution(df, \"Predictions made by Linear Model\")\nfor i in 1:length(chain_linear)\n lines!(Makie.KernelDensity.kde(pred[:, i]), color=ifelse(df.Accuracy[i] == 1, \"#388E3C\", \"#D32F2F\"), alpha=0.1)\nend\nfig\n```\n\n::: {.cell-output .cell-output-stderr}\n```\n┌ Warning: Found `resolution` in the theme when creating a `Scene`. The `resolution` keyword for `Scene`s and `Figure`s has been deprecated. Use `Figure(; size = ...` or `Scene(; size = ...)` instead, which better reflects that this is a unitless size and not a pixel resolution. The key could also come from `set_theme!` calls or related theming functions.\n└ @ Makie C:\\Users\\domma\\.julia\\packages\\Makie\\VRavR\\src\\scenes.jl:220\n```\n:::\n\n::: {.cell-output .cell-output-display execution_count=7}\n{}\n:::\n:::\n\n\n#### The Problem with Linear Models\n\nReaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the \"best fitting\" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\\mu$ and standard deviation $\\sigma$) are not good descriptors of the data.\n\n\n\n> Linear models try to find the best fitting Normal distribution for the data. However, for reaction times, even the best fitting Normal distribution (in red) does not capture well the actual data (in grey).\n\nA popular mitigation method to account for the non-normality of RTs is to transform the data, using for instance the popular *log-transform*. \nHowever, this practice should be avoided as it leads to various issues, including loss of power and distorted results interpretation [@lo2015transform; @schramm2019reaction].\nInstead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.\n\n\n### Shifted LogNormal Model\n\nOne of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.\n\n### ExGaussian Model\n\n\n### Wald Model\n\nMoe from statistical models that *describe* to models that *generate* RT-like data.\n\n## Generative Models (DDM)\n\nUse DDM as a case study to introduce generative models\n\n## Other Models (LBA, LNR)\n\n\n## Additional Resources\n\n- [**Lindelov's overview of RT models**](https://lindeloev.github.io/shiny-rt/): An absolute must-read.\n- [**De Boeck & Jeon (2019)**](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00102/full): A paper providing an overview of RT models.\n- [https://github.com/vasishth/bayescogsci](https://github.com/vasishth/bayescogsci)\n\n",
"supporting": [
- "4_rt_files"
+ "4_rt_files\\figure-html"
],
"filters": [],
"includes": {
diff --git a/content/.quarto/_freeze/4_rt/figure-html/cell-4-output-2.svg b/content/.quarto/_freeze/4_rt/figure-html/cell-4-output-2.svg
new file mode 100644
index 0000000..3bf0846
--- /dev/null
+++ b/content/.quarto/_freeze/4_rt/figure-html/cell-4-output-2.svg
@@ -0,0 +1,647 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/content/.quarto/_freeze/4_rt/figure-html/cell-7-output-2.svg b/content/.quarto/_freeze/4_rt/figure-html/cell-7-output-2.svg
new file mode 100644
index 0000000..c5d19fb
--- /dev/null
+++ b/content/.quarto/_freeze/4_rt/figure-html/cell-7-output-2.svg
@@ -0,0 +1,1695 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/content/.quarto/cites/index.json b/content/.quarto/cites/index.json
index 8fcd73c..a9c52cb 100644
--- a/content/.quarto/cites/index.json
+++ b/content/.quarto/cites/index.json
@@ -1 +1 @@
-{"index.qmd":[],"references.qmd":[],"4_rt.qmd":["lo2015transform","schramm2019reaction"],"2_predictors.qmd":[],"3_scales.qmd":[],"5_individual.qmd":[],"1_introduction.qmd":[]}
+{"3_scales.qmd":[],"5_individual.qmd":[],"references.qmd":[],"2_predictors.qmd":[],"4_rt.qmd":["wagenmakers2008diffusion","theriault2024check","lo2015transform","schramm2019reaction"],"index.qmd":[],"1_introduction.qmd":[]}
diff --git a/content/.quarto/idx/2_predictors.qmd.json b/content/.quarto/idx/2_predictors.qmd.json
index c04ce7f..fa4ae8b 100644
--- a/content/.quarto/idx/2_predictors.qmd.json
+++ b/content/.quarto/idx/2_predictors.qmd.json
@@ -1 +1 @@
-{"title":"Predictors","markdown":{"headingText":"Predictors","containsRefs":false,"markdown":"\n\n## Categorical predictors (Condition, Group, ...)\n\nNested interactions, contrasts, ...\n\n## Ordered predictors (Likert Scales)\n\nLikert scales, i.e., ordered multiple *discrete* choices are often used in surveys and questionnaires. While such data is often treated as a *continuous* variable, such assumption is not necessarily valid. Indeed, distance between the choices is not necessarily equal. For example, the difference between \"strongly agree\" and \"agree\" might not be the same as between \"agree\" and \"neutral\". Even when using integers like 1, 2, 3, 4; people might implicitly process \"4\" as more extreme relative to \"3\" as \"3\" to \"2\".\n\n\n\n> The probabilities assigned to discrete probability descriptors are not necessarily equidistant (https://github.com/zonination/perceptions)\n\nWhat can we do to better reflect the cognitive process underlying a Likert scale responses? [Monothonic effects](https://cran.r-project.org/web/packages/brms/vignettes/brms_monotonic.html).\n\n## Interactions\n\nTodo. \n\n## Non-linear relationships (polynomial, GAMs)\n\nTodo. ","srcMarkdownNoYaml":""},"formats":{"html":{"identifier":{"display-name":"HTML","target-format":"html","base-format":"html"},"execute":{"fig-width":7,"fig-height":5,"fig-format":"retina","fig-dpi":96,"df-print":"default","error":false,"eval":true,"cache":true,"freeze":"auto","echo":true,"output":true,"warning":true,"include":true,"keep-md":false,"keep-ipynb":false,"ipynb":null,"enabled":null,"daemon":null,"daemon-restart":false,"debug":false,"ipynb-filters":[],"ipynb-shell-interactivity":null,"plotly-connected":true,"engine":"markdown"},"render":{"keep-tex":false,"keep-typ":false,"keep-source":false,"keep-hidden":false,"prefer-html":false,"output-divs":true,"output-ext":"html","fig-align":"default","fig-pos":null,"fig-env":null,"code-fold":true,"code-overflow":"scroll","code-link":false,"code-line-numbers":false,"code-tools":false,"tbl-colwidths":"auto","merge-includes":true,"inline-includes":false,"preserve-yaml":false,"latex-auto-mk":true,"latex-auto-install":true,"latex-clean":true,"latex-min-runs":1,"latex-max-runs":10,"latex-makeindex":"makeindex","latex-makeindex-opts":[],"latex-tlmgr-opts":[],"latex-input-paths":[],"latex-output-dir":null,"link-external-icon":false,"link-external-newwindow":false,"self-contained-math":false,"format-resources":[],"notebook-links":true},"pandoc":{"standalone":true,"wrap":"none","default-image-extension":"png","to":"html","output-file":"2_predictors.html"},"language":{"toc-title-document":"Table of contents","toc-title-website":"On this page","related-formats-title":"Other Formats","related-notebooks-title":"Notebooks","source-notebooks-prefix":"Source","other-links-title":"Other Links","code-links-title":"Code Links","launch-dev-container-title":"Launch Dev Container","launch-binder-title":"Launch Binder","article-notebook-label":"Article Notebook","notebook-preview-download":"Download Notebook","notebook-preview-download-src":"Download Source","notebook-preview-back":"Back to Article","manuscript-meca-bundle":"MECA Bundle","section-title-abstract":"Abstract","section-title-appendices":"Appendices","section-title-footnotes":"Footnotes","section-title-references":"References","section-title-reuse":"Reuse","section-title-copyright":"Copyright","section-title-citation":"Citation","appendix-attribution-cite-as":"For attribution, please cite this work as:","appendix-attribution-bibtex":"BibTeX citation:","title-block-author-single":"Author","title-block-author-plural":"Authors","title-block-affiliation-single":"Affiliation","title-block-affiliation-plural":"Affiliations","title-block-published":"Published","title-block-modified":"Modified","title-block-keywords":"Keywords","callout-tip-title":"Tip","callout-note-title":"Note","callout-warning-title":"Warning","callout-important-title":"Important","callout-caution-title":"Caution","code-summary":"Code","code-tools-menu-caption":"Code","code-tools-show-all-code":"Show All Code","code-tools-hide-all-code":"Hide All Code","code-tools-view-source":"View Source","code-tools-source-code":"Source Code","tools-share":"Share","tools-download":"Download","code-line":"Line","code-lines":"Lines","copy-button-tooltip":"Copy to Clipboard","copy-button-tooltip-success":"Copied!","repo-action-links-edit":"Edit this page","repo-action-links-source":"View source","repo-action-links-issue":"Report an issue","back-to-top":"Back to top","search-no-results-text":"No results","search-matching-documents-text":"matching documents","search-copy-link-title":"Copy link to search","search-hide-matches-text":"Hide additional matches","search-more-match-text":"more match in this document","search-more-matches-text":"more matches in this document","search-clear-button-title":"Clear","search-text-placeholder":"","search-detached-cancel-button-title":"Cancel","search-submit-button-title":"Submit","search-label":"Search","toggle-section":"Toggle section","toggle-sidebar":"Toggle sidebar navigation","toggle-dark-mode":"Toggle dark mode","toggle-reader-mode":"Toggle reader mode","toggle-navigation":"Toggle navigation","crossref-fig-title":"Figure","crossref-tbl-title":"Table","crossref-lst-title":"Listing","crossref-thm-title":"Theorem","crossref-lem-title":"Lemma","crossref-cor-title":"Corollary","crossref-prp-title":"Proposition","crossref-cnj-title":"Conjecture","crossref-def-title":"Definition","crossref-exm-title":"Example","crossref-exr-title":"Exercise","crossref-ch-prefix":"Chapter","crossref-apx-prefix":"Appendix","crossref-sec-prefix":"Section","crossref-eq-prefix":"Equation","crossref-lof-title":"List of Figures","crossref-lot-title":"List of Tables","crossref-lol-title":"List of Listings","environment-proof-title":"Proof","environment-remark-title":"Remark","environment-solution-title":"Solution","listing-page-order-by":"Order By","listing-page-order-by-default":"Default","listing-page-order-by-date-asc":"Oldest","listing-page-order-by-date-desc":"Newest","listing-page-order-by-number-desc":"High to Low","listing-page-order-by-number-asc":"Low to High","listing-page-field-date":"Date","listing-page-field-title":"Title","listing-page-field-description":"Description","listing-page-field-author":"Author","listing-page-field-filename":"File Name","listing-page-field-filemodified":"Modified","listing-page-field-subtitle":"Subtitle","listing-page-field-readingtime":"Reading Time","listing-page-field-wordcount":"Word Count","listing-page-field-categories":"Categories","listing-page-minutes-compact":"{0} min","listing-page-category-all":"All","listing-page-no-matches":"No matching items","listing-page-words":"{0} words"},"metadata":{"lang":"en","fig-responsive":true,"quarto-version":"1.4.549","bibliography":["references.bib"],"theme":"cosmo"},"extensions":{"book":{"multiFile":true}}}},"projectFormats":["html"]}
\ No newline at end of file
+{"title":"Predictors","markdown":{"headingText":"Predictors","containsRefs":false,"markdown":"\n\n## Categorical predictors (Condition, Group, ...)\n\nNested interactions, contrasts, ...\n\n## Ordered predictors (Likert Scales)\n\nLikert scales, i.e., ordered multiple *discrete* choices are often used in surveys and questionnaires. While such data is often treated as a *continuous* variable, such assumption is not necessarily valid. Indeed, distance between the choices is not necessarily equal. For example, the difference between \"strongly agree\" and \"agree\" might not be the same as between \"agree\" and \"neutral\". Even when using integers like 1, 2, 3, 4; people might implicitly process \"4\" as more extreme relative to \"3\" as \"3\" to \"2\".\n\n\n\n> The probabilities assigned to discrete probability descriptors are not necessarily equidistant (https://github.com/zonination/perceptions)\n\nWhat can we do to better reflect the cognitive process underlying a Likert scale responses? [Monotonic effects](https://cran.r-project.org/web/packages/brms/vignettes/brms_monotonic.html).\n\n## Interactions\n\nTodo. \n\n## Non-linear relationships (polynomial, GAMs)\n\nTodo. ","srcMarkdownNoYaml":""},"formats":{"html":{"identifier":{"display-name":"HTML","target-format":"html","base-format":"html"},"execute":{"fig-width":7,"fig-height":5,"fig-format":"retina","fig-dpi":96,"df-print":"default","error":false,"eval":true,"cache":true,"freeze":"auto","echo":true,"output":true,"warning":true,"include":true,"keep-md":false,"keep-ipynb":false,"ipynb":null,"enabled":null,"daemon":null,"daemon-restart":false,"debug":false,"ipynb-filters":[],"ipynb-shell-interactivity":null,"plotly-connected":true,"engine":"markdown"},"render":{"keep-tex":false,"keep-typ":false,"keep-source":false,"keep-hidden":false,"prefer-html":false,"output-divs":true,"output-ext":"html","fig-align":"default","fig-pos":null,"fig-env":null,"code-fold":true,"code-overflow":"scroll","code-link":false,"code-line-numbers":false,"code-tools":false,"tbl-colwidths":"auto","merge-includes":true,"inline-includes":false,"preserve-yaml":false,"latex-auto-mk":true,"latex-auto-install":true,"latex-clean":true,"latex-min-runs":1,"latex-max-runs":10,"latex-makeindex":"makeindex","latex-makeindex-opts":[],"latex-tlmgr-opts":[],"latex-input-paths":[],"latex-output-dir":null,"link-external-icon":false,"link-external-newwindow":false,"self-contained-math":false,"format-resources":[],"notebook-links":true},"pandoc":{"standalone":true,"wrap":"none","default-image-extension":"png","to":"html","output-file":"2_predictors.html"},"language":{"toc-title-document":"Table of contents","toc-title-website":"On this page","related-formats-title":"Other Formats","related-notebooks-title":"Notebooks","source-notebooks-prefix":"Source","other-links-title":"Other Links","code-links-title":"Code Links","launch-dev-container-title":"Launch Dev Container","launch-binder-title":"Launch Binder","article-notebook-label":"Article Notebook","notebook-preview-download":"Download Notebook","notebook-preview-download-src":"Download Source","notebook-preview-back":"Back to Article","manuscript-meca-bundle":"MECA Bundle","section-title-abstract":"Abstract","section-title-appendices":"Appendices","section-title-footnotes":"Footnotes","section-title-references":"References","section-title-reuse":"Reuse","section-title-copyright":"Copyright","section-title-citation":"Citation","appendix-attribution-cite-as":"For attribution, please cite this work as:","appendix-attribution-bibtex":"BibTeX citation:","title-block-author-single":"Author","title-block-author-plural":"Authors","title-block-affiliation-single":"Affiliation","title-block-affiliation-plural":"Affiliations","title-block-published":"Published","title-block-modified":"Modified","title-block-keywords":"Keywords","callout-tip-title":"Tip","callout-note-title":"Note","callout-warning-title":"Warning","callout-important-title":"Important","callout-caution-title":"Caution","code-summary":"Code","code-tools-menu-caption":"Code","code-tools-show-all-code":"Show All Code","code-tools-hide-all-code":"Hide All Code","code-tools-view-source":"View Source","code-tools-source-code":"Source Code","tools-share":"Share","tools-download":"Download","code-line":"Line","code-lines":"Lines","copy-button-tooltip":"Copy to Clipboard","copy-button-tooltip-success":"Copied!","repo-action-links-edit":"Edit this page","repo-action-links-source":"View source","repo-action-links-issue":"Report an issue","back-to-top":"Back to top","search-no-results-text":"No results","search-matching-documents-text":"matching documents","search-copy-link-title":"Copy link to search","search-hide-matches-text":"Hide additional matches","search-more-match-text":"more match in this document","search-more-matches-text":"more matches in this document","search-clear-button-title":"Clear","search-text-placeholder":"","search-detached-cancel-button-title":"Cancel","search-submit-button-title":"Submit","search-label":"Search","toggle-section":"Toggle section","toggle-sidebar":"Toggle sidebar navigation","toggle-dark-mode":"Toggle dark mode","toggle-reader-mode":"Toggle reader mode","toggle-navigation":"Toggle navigation","crossref-fig-title":"Figure","crossref-tbl-title":"Table","crossref-lst-title":"Listing","crossref-thm-title":"Theorem","crossref-lem-title":"Lemma","crossref-cor-title":"Corollary","crossref-prp-title":"Proposition","crossref-cnj-title":"Conjecture","crossref-def-title":"Definition","crossref-exm-title":"Example","crossref-exr-title":"Exercise","crossref-ch-prefix":"Chapter","crossref-apx-prefix":"Appendix","crossref-sec-prefix":"Section","crossref-eq-prefix":"Equation","crossref-lof-title":"List of Figures","crossref-lot-title":"List of Tables","crossref-lol-title":"List of Listings","environment-proof-title":"Proof","environment-remark-title":"Remark","environment-solution-title":"Solution","listing-page-order-by":"Order By","listing-page-order-by-default":"Default","listing-page-order-by-date-asc":"Oldest","listing-page-order-by-date-desc":"Newest","listing-page-order-by-number-desc":"High to Low","listing-page-order-by-number-asc":"Low to High","listing-page-field-date":"Date","listing-page-field-title":"Title","listing-page-field-description":"Description","listing-page-field-author":"Author","listing-page-field-filename":"File Name","listing-page-field-filemodified":"Modified","listing-page-field-subtitle":"Subtitle","listing-page-field-readingtime":"Reading Time","listing-page-field-wordcount":"Word Count","listing-page-field-categories":"Categories","listing-page-minutes-compact":"{0} min","listing-page-category-all":"All","listing-page-no-matches":"No matching items","listing-page-words":"{0} words"},"metadata":{"lang":"en","fig-responsive":true,"quarto-version":"1.4.549","bibliography":["references.bib"],"theme":"cosmo"},"extensions":{"book":{"multiFile":true}}}},"projectFormats":["html"]}
\ No newline at end of file
diff --git a/content/.quarto/idx/4_rt.qmd.json b/content/.quarto/idx/4_rt.qmd.json
index debf4af..5b48326 100644
--- a/content/.quarto/idx/4_rt.qmd.json
+++ b/content/.quarto/idx/4_rt.qmd.json
@@ -1 +1 @@
-{"title":"Reaction Times","markdown":{"headingText":"Reaction Times","containsRefs":false,"markdown":"\n\nThis repository contain the following vignettes:\n\n- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)\n- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)\n\n## Descriptive Models (ExGaussian, LogNormal, Wald)\n\n### Modelling RT with a Bayesian Linear Model\n\n#### The Data\n\n```{julia}\nusing Downloads, CSV, DataFrames\nusing Turing, Distributions\nusing CairoMakie\n\ndf = CSV.read(Downloads.download(\"https://raw.githubusercontent.com/RealityBending/DoggoNogo/main/study1/data/data_game.csv\"), DataFrame)\nfirst(df, 10)\n```\n\n\n#### The Model\n\n```{julia}\n#| code-fold: false\n\n@model function model_linear(rt; isi=nothing)\n\n # Set priors on variance, intercept and effect of ISI\n σ² ~ truncated(Normal(0, 1); lower=0)\n intercept ~ truncated(Normal(0, 1); lower=0)\n slope_isi ~ Normal(0, 0.5)\n\n for i in 1:length(rt)\n μ = intercept + slope_isi * isi[i]\n rt[i] ~ Normal(μ, σ²)\n end\nend\n\nmodel = model_linear(df.RT, isi=df.ISI)\nchain_linear = sample(model, NUTS(), 200)\n\n# Summary (95% CI)\nquantile(chain_linear; q=[0.025, 0.975])\n```\n\n::: {.callout-tip title=\"Code Tip\"}\nWe first initialize the model by passing the `RT` and `ISI` columns.\n:::\n\n#### Posterior Predictive Check\n\n```{julia}\n#| output: false\n\npred = predict(model_linear([(missing) for i in 1:length(df.RT)]; isi=df.ISI), chain_linear)\npred = Array(pred)\n```\n\n```{julia}\n#| fig-width: 10\n#| fig-height: 7\n\nf = Figure()\nax = Axis(f[1, 1], title=\"Predicted Data by Linear Model\",\n xlabel=\"RT (s)\",\n ylabel=\"Distribution\",\n yticksvisible=false,\n xticksvisible=false,\n yticklabelsvisible=false)\n\nCairoMakie.density!(df.RT, color=\"grey\")\nfor i in 1:length(chain_linear)\n lines!(ax, Makie.KernelDensity.kde(pred[:, i]), color=\"orange\", alpha=0.1)\nend\nCairoMakie.ylims!(ax, (0, nothing))\nf\n```\n\n### The Problem with Linear Models\n\nReaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the \"best fitting\" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\\mu$ and standard deviation $\\sigma$) are not good descriptors of the data.\n\n\n\n> Linear models try to find the best fitting Normal distribution for the data. However, for reaction times, even the best fitting Normal distribution (in red) does not capture well the actual data (in grey).\n\nA popular mitigation method to account for the non-normality of RTs is to transform the data, using for instance the popular *log-transform*. \nHowever, this practice should be avoided as it leads to various issues, including loss of power and distorted results interpretation [@lo2015transform; @schramm2019reaction].\nInstead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.\n\n\n### Shifted LogNormal Models\n\nOne of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.\n\n\n### Wald\n\nMoe from statistical models that *describe* to models that *generate* RT-like data.\n\n### Generative Models (DDM)\n\nUse DDM as a case study to introduce generative models\n\n### Other Models (LBA, LNR)\n\n\n## Additional Resources\n\n- [**Lindelov's overview of RT models**](https://lindeloev.github.io/shiny-rt/): An absolute must-read.\n- [**De Boeck & Jeon (2019)**](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00102/full): A paper providing an overview of RT models.\n- [https://github.com/vasishth/bayescogsci](https://github.com/vasishth/bayescogsci)\n","srcMarkdownNoYaml":""},"formats":{"html":{"identifier":{"display-name":"HTML","target-format":"html","base-format":"html"},"execute":{"fig-width":7,"fig-height":5,"fig-format":"retina","fig-dpi":96,"df-print":"default","error":false,"eval":true,"cache":true,"freeze":"auto","echo":true,"output":true,"warning":true,"include":true,"keep-md":false,"keep-ipynb":false,"ipynb":null,"enabled":null,"daemon":null,"daemon-restart":false,"debug":false,"ipynb-filters":[],"ipynb-shell-interactivity":null,"plotly-connected":true,"engine":"jupyter"},"render":{"keep-tex":false,"keep-typ":false,"keep-source":false,"keep-hidden":false,"prefer-html":false,"output-divs":true,"output-ext":"html","fig-align":"default","fig-pos":null,"fig-env":null,"code-fold":true,"code-overflow":"scroll","code-link":false,"code-line-numbers":false,"code-tools":false,"tbl-colwidths":"auto","merge-includes":true,"inline-includes":false,"preserve-yaml":false,"latex-auto-mk":true,"latex-auto-install":true,"latex-clean":true,"latex-min-runs":1,"latex-max-runs":10,"latex-makeindex":"makeindex","latex-makeindex-opts":[],"latex-tlmgr-opts":[],"latex-input-paths":[],"latex-output-dir":null,"link-external-icon":false,"link-external-newwindow":false,"self-contained-math":false,"format-resources":[],"notebook-links":true},"pandoc":{"standalone":true,"wrap":"none","default-image-extension":"png","to":"html","output-file":"4_rt.html"},"language":{"toc-title-document":"Table of contents","toc-title-website":"On this page","related-formats-title":"Other Formats","related-notebooks-title":"Notebooks","source-notebooks-prefix":"Source","other-links-title":"Other Links","code-links-title":"Code Links","launch-dev-container-title":"Launch Dev Container","launch-binder-title":"Launch Binder","article-notebook-label":"Article Notebook","notebook-preview-download":"Download Notebook","notebook-preview-download-src":"Download Source","notebook-preview-back":"Back to Article","manuscript-meca-bundle":"MECA Bundle","section-title-abstract":"Abstract","section-title-appendices":"Appendices","section-title-footnotes":"Footnotes","section-title-references":"References","section-title-reuse":"Reuse","section-title-copyright":"Copyright","section-title-citation":"Citation","appendix-attribution-cite-as":"For attribution, please cite this work as:","appendix-attribution-bibtex":"BibTeX citation:","title-block-author-single":"Author","title-block-author-plural":"Authors","title-block-affiliation-single":"Affiliation","title-block-affiliation-plural":"Affiliations","title-block-published":"Published","title-block-modified":"Modified","title-block-keywords":"Keywords","callout-tip-title":"Tip","callout-note-title":"Note","callout-warning-title":"Warning","callout-important-title":"Important","callout-caution-title":"Caution","code-summary":"Code","code-tools-menu-caption":"Code","code-tools-show-all-code":"Show All Code","code-tools-hide-all-code":"Hide All Code","code-tools-view-source":"View Source","code-tools-source-code":"Source Code","tools-share":"Share","tools-download":"Download","code-line":"Line","code-lines":"Lines","copy-button-tooltip":"Copy to Clipboard","copy-button-tooltip-success":"Copied!","repo-action-links-edit":"Edit this page","repo-action-links-source":"View source","repo-action-links-issue":"Report an issue","back-to-top":"Back to top","search-no-results-text":"No results","search-matching-documents-text":"matching documents","search-copy-link-title":"Copy link to search","search-hide-matches-text":"Hide additional matches","search-more-match-text":"more match in this document","search-more-matches-text":"more matches in this document","search-clear-button-title":"Clear","search-text-placeholder":"","search-detached-cancel-button-title":"Cancel","search-submit-button-title":"Submit","search-label":"Search","toggle-section":"Toggle section","toggle-sidebar":"Toggle sidebar navigation","toggle-dark-mode":"Toggle dark mode","toggle-reader-mode":"Toggle reader mode","toggle-navigation":"Toggle navigation","crossref-fig-title":"Figure","crossref-tbl-title":"Table","crossref-lst-title":"Listing","crossref-thm-title":"Theorem","crossref-lem-title":"Lemma","crossref-cor-title":"Corollary","crossref-prp-title":"Proposition","crossref-cnj-title":"Conjecture","crossref-def-title":"Definition","crossref-exm-title":"Example","crossref-exr-title":"Exercise","crossref-ch-prefix":"Chapter","crossref-apx-prefix":"Appendix","crossref-sec-prefix":"Section","crossref-eq-prefix":"Equation","crossref-lof-title":"List of Figures","crossref-lot-title":"List of Tables","crossref-lol-title":"List of Listings","environment-proof-title":"Proof","environment-remark-title":"Remark","environment-solution-title":"Solution","listing-page-order-by":"Order By","listing-page-order-by-default":"Default","listing-page-order-by-date-asc":"Oldest","listing-page-order-by-date-desc":"Newest","listing-page-order-by-number-desc":"High to Low","listing-page-order-by-number-asc":"Low to High","listing-page-field-date":"Date","listing-page-field-title":"Title","listing-page-field-description":"Description","listing-page-field-author":"Author","listing-page-field-filename":"File Name","listing-page-field-filemodified":"Modified","listing-page-field-subtitle":"Subtitle","listing-page-field-readingtime":"Reading Time","listing-page-field-wordcount":"Word Count","listing-page-field-categories":"Categories","listing-page-minutes-compact":"{0} min","listing-page-category-all":"All","listing-page-no-matches":"No matching items","listing-page-words":"{0} words"},"metadata":{"lang":"en","fig-responsive":true,"quarto-version":"1.4.549","bibliography":["references.bib"],"theme":"cosmo"},"extensions":{"book":{"multiFile":true}}}},"projectFormats":["html"]}
\ No newline at end of file
+{"title":"Reaction Times","markdown":{"headingText":"Reaction Times","containsRefs":false,"markdown":"\n\nThis repository contain the following vignettes:\n\n- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)\n- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)\n\n## The Data\n\nData from @wagenmakers2008diffusion - Experiment 1.\nWe excluded all trials with uninterpretable response time [see @theriault2024check] such as too fast response (<180 ms) and too slow response (>2 sec).\n\n```{julia}\n#| code-fold: false\n\nusing Downloads, CSV, DataFrames\nusing Turing, Distributions, SequentialSamplingModels\nusing CairoMakie\n\ndf = CSV.read(Downloads.download(\"https://raw.githubusercontent.com/DominiqueMakowski/CognitiveModels/main/data/wagenmakers2008.csv\"), DataFrame)\nfirst(df, 10)\n```\n\nWe create a new column, `Accuracy`, which is the \"binarization\" of the `Condition` column, and is equal to 1 when the condition is `\"Accuracy\"` and 0 when it is `\"Speed\"`.\n\n```{julia}\n#| output: false\n\ndf = df[df.Error .== 0, :]\ndf.Accuracy = df.Condition .== \"Accuracy\"\n```\n\n\n::: {.callout-tip title=\"Code Tip\"}\nNote the usage of *vectorization* `.==` as we want to compare each element of the `Condition` vector to the target `\"Accuracy\"`.\n:::\n\n```{julia}\nfunction plot_distribution(df, title=\"Empirical Distribution of Data from Wagenmakers et al. (2018)\")\n fig = Figure()\n ax = Axis(fig[1, 1], title=title,\n xlabel=\"RT (s)\",\n ylabel=\"Distribution\",\n yticksvisible=false,\n xticksvisible=false,\n yticklabelsvisible=false)\n CairoMakie.density!(df[df.Condition .== \"Speed\", :RT], color=(\"#EF5350\", 0.7), label = \"Speed\")\n CairoMakie.density!(df[df.Condition .== \"Accuracy\", :RT], color=(\"#66BB6A\", 0.7), label = \"Accuracy\")\n CairoMakie.axislegend(\"Condition\"; position=:rt)\n CairoMakie.ylims!(ax, (0, nothing))\n return fig\nend\n\nplot_distribution(df, \"Empirical Distribution of Data from Wagenmakers et al. (2018)\")\n```\n\n\n## Descriptive Models (ExGaussian, LogNormal, Wald)\n\n### Modelling RT with a Bayesian Linear Model\n\n#### The Data\n\n\n#### The Model\n\n```{julia}\n#| code-fold: false\n\n@model function model_linear(rt; condition=nothing)\n\n # Set priors on variance, intercept and effect of ISI\n σ² ~ truncated(Normal(0, 1); lower=0)\n intercept ~ truncated(Normal(0, 1); lower=0)\n slope_accuracy ~ Normal(0, 0.5)\n\n for i in 1:length(rt)\n μ = intercept + slope_accuracy * condition[i]\n rt[i] ~ Normal(μ, σ²)\n end\nend\n\n\nmodel = model_linear(df.RT, condition=df.Accuracy)\nchain_linear = sample(model, NUTS(), 200)\n\n# Summary (95% CI)\nquantile(chain_linear; q=[0.025, 0.975])\n```\n\nThe effect of Condition is significant, people are on average slower (higher RT) when condition is `\"Accuracy\"`.\nBut is our model good?\n\n#### Posterior Predictive Check\n\n```{julia}\n#| output: false\n\npred = predict(model_linear([(missing) for i in 1:length(df.RT)], condition=df.Accuracy), chain_linear)\npred = Array(pred)\n```\n\n```{julia}\n#| fig-width: 10\n#| fig-height: 7\n\nfig = plot_distribution(df, \"Predictions made by Linear Model\")\nfor i in 1:length(chain_linear)\n lines!(Makie.KernelDensity.kde(pred[:, i]), color=ifelse(df.Accuracy[i] == 1, \"#388E3C\", \"#D32F2F\"), alpha=0.1)\nend\nfig\n```\n\n#### The Problem with Linear Models\n\nReaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the \"best fitting\" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\\mu$ and standard deviation $\\sigma$) are not good descriptors of the data.\n\n\n\n> Linear models try to find the best fitting Normal distribution for the data. However, for reaction times, even the best fitting Normal distribution (in red) does not capture well the actual data (in grey).\n\nA popular mitigation method to account for the non-normality of RTs is to transform the data, using for instance the popular *log-transform*. \nHowever, this practice should be avoided as it leads to various issues, including loss of power and distorted results interpretation [@lo2015transform; @schramm2019reaction].\nInstead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.\n\n\n### Shifted LogNormal Model\n\nOne of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.\n\n### ExGaussian Model\n\n\n### Wald Model\n\nMoe from statistical models that *describe* to models that *generate* RT-like data.\n\n## Generative Models (DDM)\n\nUse DDM as a case study to introduce generative models\n\n## Other Models (LBA, LNR)\n\n\n## Additional Resources\n\n- [**Lindelov's overview of RT models**](https://lindeloev.github.io/shiny-rt/): An absolute must-read.\n- [**De Boeck & Jeon (2019)**](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00102/full): A paper providing an overview of RT models.\n- [https://github.com/vasishth/bayescogsci](https://github.com/vasishth/bayescogsci)\n","srcMarkdownNoYaml":""},"formats":{"html":{"identifier":{"display-name":"HTML","target-format":"html","base-format":"html"},"execute":{"fig-width":7,"fig-height":5,"fig-format":"retina","fig-dpi":96,"df-print":"default","error":false,"eval":true,"cache":true,"freeze":"auto","echo":true,"output":true,"warning":true,"include":true,"keep-md":false,"keep-ipynb":false,"ipynb":null,"enabled":null,"daemon":null,"daemon-restart":false,"debug":false,"ipynb-filters":[],"ipynb-shell-interactivity":null,"plotly-connected":true,"engine":"jupyter"},"render":{"keep-tex":false,"keep-typ":false,"keep-source":false,"keep-hidden":false,"prefer-html":false,"output-divs":true,"output-ext":"html","fig-align":"default","fig-pos":null,"fig-env":null,"code-fold":true,"code-overflow":"scroll","code-link":false,"code-line-numbers":false,"code-tools":false,"tbl-colwidths":"auto","merge-includes":true,"inline-includes":false,"preserve-yaml":false,"latex-auto-mk":true,"latex-auto-install":true,"latex-clean":true,"latex-min-runs":1,"latex-max-runs":10,"latex-makeindex":"makeindex","latex-makeindex-opts":[],"latex-tlmgr-opts":[],"latex-input-paths":[],"latex-output-dir":null,"link-external-icon":false,"link-external-newwindow":false,"self-contained-math":false,"format-resources":[],"notebook-links":true},"pandoc":{"standalone":true,"wrap":"none","default-image-extension":"png","to":"html","output-file":"4_rt.html"},"language":{"toc-title-document":"Table of contents","toc-title-website":"On this page","related-formats-title":"Other Formats","related-notebooks-title":"Notebooks","source-notebooks-prefix":"Source","other-links-title":"Other Links","code-links-title":"Code Links","launch-dev-container-title":"Launch Dev Container","launch-binder-title":"Launch Binder","article-notebook-label":"Article Notebook","notebook-preview-download":"Download Notebook","notebook-preview-download-src":"Download Source","notebook-preview-back":"Back to Article","manuscript-meca-bundle":"MECA Bundle","section-title-abstract":"Abstract","section-title-appendices":"Appendices","section-title-footnotes":"Footnotes","section-title-references":"References","section-title-reuse":"Reuse","section-title-copyright":"Copyright","section-title-citation":"Citation","appendix-attribution-cite-as":"For attribution, please cite this work as:","appendix-attribution-bibtex":"BibTeX citation:","title-block-author-single":"Author","title-block-author-plural":"Authors","title-block-affiliation-single":"Affiliation","title-block-affiliation-plural":"Affiliations","title-block-published":"Published","title-block-modified":"Modified","title-block-keywords":"Keywords","callout-tip-title":"Tip","callout-note-title":"Note","callout-warning-title":"Warning","callout-important-title":"Important","callout-caution-title":"Caution","code-summary":"Code","code-tools-menu-caption":"Code","code-tools-show-all-code":"Show All Code","code-tools-hide-all-code":"Hide All Code","code-tools-view-source":"View Source","code-tools-source-code":"Source Code","tools-share":"Share","tools-download":"Download","code-line":"Line","code-lines":"Lines","copy-button-tooltip":"Copy to Clipboard","copy-button-tooltip-success":"Copied!","repo-action-links-edit":"Edit this page","repo-action-links-source":"View source","repo-action-links-issue":"Report an issue","back-to-top":"Back to top","search-no-results-text":"No results","search-matching-documents-text":"matching documents","search-copy-link-title":"Copy link to search","search-hide-matches-text":"Hide additional matches","search-more-match-text":"more match in this document","search-more-matches-text":"more matches in this document","search-clear-button-title":"Clear","search-text-placeholder":"","search-detached-cancel-button-title":"Cancel","search-submit-button-title":"Submit","search-label":"Search","toggle-section":"Toggle section","toggle-sidebar":"Toggle sidebar navigation","toggle-dark-mode":"Toggle dark mode","toggle-reader-mode":"Toggle reader mode","toggle-navigation":"Toggle navigation","crossref-fig-title":"Figure","crossref-tbl-title":"Table","crossref-lst-title":"Listing","crossref-thm-title":"Theorem","crossref-lem-title":"Lemma","crossref-cor-title":"Corollary","crossref-prp-title":"Proposition","crossref-cnj-title":"Conjecture","crossref-def-title":"Definition","crossref-exm-title":"Example","crossref-exr-title":"Exercise","crossref-ch-prefix":"Chapter","crossref-apx-prefix":"Appendix","crossref-sec-prefix":"Section","crossref-eq-prefix":"Equation","crossref-lof-title":"List of Figures","crossref-lot-title":"List of Tables","crossref-lol-title":"List of Listings","environment-proof-title":"Proof","environment-remark-title":"Remark","environment-solution-title":"Solution","listing-page-order-by":"Order By","listing-page-order-by-default":"Default","listing-page-order-by-date-asc":"Oldest","listing-page-order-by-date-desc":"Newest","listing-page-order-by-number-desc":"High to Low","listing-page-order-by-number-asc":"Low to High","listing-page-field-date":"Date","listing-page-field-title":"Title","listing-page-field-description":"Description","listing-page-field-author":"Author","listing-page-field-filename":"File Name","listing-page-field-filemodified":"Modified","listing-page-field-subtitle":"Subtitle","listing-page-field-readingtime":"Reading Time","listing-page-field-wordcount":"Word Count","listing-page-field-categories":"Categories","listing-page-minutes-compact":"{0} min","listing-page-category-all":"All","listing-page-no-matches":"No matching items","listing-page-words":"{0} words"},"metadata":{"lang":"en","fig-responsive":true,"quarto-version":"1.4.549","bibliography":["references.bib"],"theme":"cosmo"},"extensions":{"book":{"multiFile":true}}}},"projectFormats":["html"]}
\ No newline at end of file
diff --git a/content/.quarto/xref/04307669 b/content/.quarto/xref/04307669
index 5129a29..b25df94 100644
--- a/content/.quarto/xref/04307669
+++ b/content/.quarto/xref/04307669
@@ -1 +1 @@
-{"entries":[],"headings":[],"options":{"chapters":true}}
\ No newline at end of file
+{"options":{"chapters":true},"entries":[],"headings":[]}
\ No newline at end of file
diff --git a/content/.quarto/xref/15f266d2 b/content/.quarto/xref/15f266d2
index 42c82d3..5129a29 100644
--- a/content/.quarto/xref/15f266d2
+++ b/content/.quarto/xref/15f266d2
@@ -1 +1 @@
-{"headings":[],"options":{"chapters":true},"entries":[]}
\ No newline at end of file
+{"entries":[],"headings":[],"options":{"chapters":true}}
\ No newline at end of file
diff --git a/content/.quarto/xref/1a47137c b/content/.quarto/xref/1a47137c
index b466b0b..b25df94 100644
--- a/content/.quarto/xref/1a47137c
+++ b/content/.quarto/xref/1a47137c
@@ -1 +1 @@
-{"options":{"chapters":true},"headings":[],"entries":[]}
\ No newline at end of file
+{"options":{"chapters":true},"entries":[],"headings":[]}
\ No newline at end of file
diff --git a/content/.quarto/xref/ce37606d b/content/.quarto/xref/ce37606d
index b25df94..7f7a477 100644
--- a/content/.quarto/xref/ce37606d
+++ b/content/.quarto/xref/ce37606d
@@ -1 +1 @@
-{"options":{"chapters":true},"entries":[],"headings":[]}
\ No newline at end of file
+{"headings":[],"entries":[],"options":{"chapters":true}}
\ No newline at end of file
diff --git a/content/.quarto/xref/efe17597 b/content/.quarto/xref/efe17597
index fa9b76b..0640727 100644
--- a/content/.quarto/xref/efe17597
+++ b/content/.quarto/xref/efe17597
@@ -1 +1 @@
-{"entries":[],"options":{"chapters":true},"headings":["descriptive-models-exgaussian-lognormal-wald","modelling-rt-with-a-bayesian-linear-model","the-data","the-model","posterior-predictive-check","the-problem-with-linear-models","shifted-lognormal-models","wald","generative-models-ddm","other-models-lba-lnr","additional-resources"]}
\ No newline at end of file
+{"options":{"chapters":true},"entries":[],"headings":["the-data","descriptive-models-exgaussian-lognormal-wald","modelling-rt-with-a-bayesian-linear-model","the-data-1","the-model","posterior-predictive-check","the-problem-with-linear-models","shifted-lognormal-model","exgaussian-model","wald-model","generative-models-ddm","other-models-lba-lnr","additional-resources"]}
\ No newline at end of file
diff --git a/content/2_predictors.qmd b/content/2_predictors.qmd
index e4e3939..71e11b2 100644
--- a/content/2_predictors.qmd
+++ b/content/2_predictors.qmd
@@ -13,7 +13,7 @@ Likert scales, i.e., ordered multiple *discrete* choices are often used in surve
> The probabilities assigned to discrete probability descriptors are not necessarily equidistant (https://github.com/zonination/perceptions)
-What can we do to better reflect the cognitive process underlying a Likert scale responses? [Monothonic effects](https://cran.r-project.org/web/packages/brms/vignettes/brms_monotonic.html).
+What can we do to better reflect the cognitive process underlying a Likert scale responses? [Monotonic effects](https://cran.r-project.org/web/packages/brms/vignettes/brms_monotonic.html).
## Interactions
diff --git a/content/4_rt.qmd b/content/4_rt.qmd
index 8592092..5e40c03 100644
--- a/content/4_rt.qmd
+++ b/content/4_rt.qmd
@@ -6,57 +6,98 @@ This repository contain the following vignettes:
- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)
- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)
-## Descriptive Models (ExGaussian, LogNormal, Wald)
+## The Data
-### Modelling RT with a Bayesian Linear Model
-
-#### The Data
+Data from @wagenmakers2008diffusion - Experiment 1.
+We excluded all trials with uninterpretable response time [see @theriault2024check] such as too fast response (<180 ms) and too slow response (>2 sec).
```{julia}
+#| code-fold: false
+
using Downloads, CSV, DataFrames
-using Turing, Distributions
+using Turing, Distributions, SequentialSamplingModels
using CairoMakie
-df = CSV.read(Downloads.download("https://raw.githubusercontent.com/RealityBending/DoggoNogo/main/study1/data/data_game.csv"), DataFrame)
+df = CSV.read(Downloads.download("https://raw.githubusercontent.com/DominiqueMakowski/CognitiveModels/main/data/wagenmakers2008.csv"), DataFrame)
first(df, 10)
```
+We create a new column, `Accuracy`, which is the "binarization" of the `Condition` column, and is equal to 1 when the condition is `"Accuracy"` and 0 when it is `"Speed"`.
+
+```{julia}
+#| output: false
+
+df = df[df.Error .== 0, :]
+df.Accuracy = df.Condition .== "Accuracy"
+```
+
+
+::: {.callout-tip title="Code Tip"}
+Note the usage of *vectorization* `.==` as we want to compare each element of the `Condition` vector to the target `"Accuracy"`.
+:::
+
+```{julia}
+function plot_distribution(df, title="Empirical Distribution of Data from Wagenmakers et al. (2018)")
+ fig = Figure()
+ ax = Axis(fig[1, 1], title=title,
+ xlabel="RT (s)",
+ ylabel="Distribution",
+ yticksvisible=false,
+ xticksvisible=false,
+ yticklabelsvisible=false)
+ CairoMakie.density!(df[df.Condition .== "Speed", :RT], color=("#EF5350", 0.7), label = "Speed")
+ CairoMakie.density!(df[df.Condition .== "Accuracy", :RT], color=("#66BB6A", 0.7), label = "Accuracy")
+ CairoMakie.axislegend("Condition"; position=:rt)
+ CairoMakie.ylims!(ax, (0, nothing))
+ return fig
+end
+
+plot_distribution(df, "Empirical Distribution of Data from Wagenmakers et al. (2018)")
+```
+
+
+## Descriptive Models (ExGaussian, LogNormal, Wald)
+
+### Modelling RT with a Bayesian Linear Model
+
+#### The Data
+
#### The Model
```{julia}
#| code-fold: false
-@model function model_linear(rt; isi=nothing)
+@model function model_linear(rt; condition=nothing)
# Set priors on variance, intercept and effect of ISI
σ² ~ truncated(Normal(0, 1); lower=0)
intercept ~ truncated(Normal(0, 1); lower=0)
- slope_isi ~ Normal(0, 0.5)
+ slope_accuracy ~ Normal(0, 0.5)
for i in 1:length(rt)
- μ = intercept + slope_isi * isi[i]
+ μ = intercept + slope_accuracy * condition[i]
rt[i] ~ Normal(μ, σ²)
end
end
-model = model_linear(df.RT, isi=df.ISI)
+
+model = model_linear(df.RT, condition=df.Accuracy)
chain_linear = sample(model, NUTS(), 200)
# Summary (95% CI)
quantile(chain_linear; q=[0.025, 0.975])
```
-::: {.callout-tip title="Code Tip"}
-We first initialize the model by passing the `RT` and `ISI` columns.
-:::
+The effect of Condition is significant, people are on average slower (higher RT) when condition is `"Accuracy"`.
+But is our model good?
#### Posterior Predictive Check
```{julia}
#| output: false
-pred = predict(model_linear([(missing) for i in 1:length(df.RT)]; isi=df.ISI), chain_linear)
+pred = predict(model_linear([(missing) for i in 1:length(df.RT)], condition=df.Accuracy), chain_linear)
pred = Array(pred)
```
@@ -64,23 +105,14 @@ pred = Array(pred)
#| fig-width: 10
#| fig-height: 7
-f = Figure()
-ax = Axis(f[1, 1], title="Predicted Data by Linear Model",
- xlabel="RT (s)",
- ylabel="Distribution",
- yticksvisible=false,
- xticksvisible=false,
- yticklabelsvisible=false)
-
-CairoMakie.density!(df.RT, color="grey")
+fig = plot_distribution(df, "Predictions made by Linear Model")
for i in 1:length(chain_linear)
- lines!(ax, Makie.KernelDensity.kde(pred[:, i]), color="orange", alpha=0.1)
+ lines!(Makie.KernelDensity.kde(pred[:, i]), color=ifelse(df.Accuracy[i] == 1, "#388E3C", "#D32F2F"), alpha=0.1)
end
-CairoMakie.ylims!(ax, (0, nothing))
-f
+fig
```
-### The Problem with Linear Models
+#### The Problem with Linear Models
Reaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the "best fitting" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\mu$ and standard deviation $\sigma$) are not good descriptors of the data.
@@ -93,20 +125,22 @@ However, this practice should be avoided as it leads to various issues, includin
Instead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.
-### Shifted LogNormal Models
+### Shifted LogNormal Model
One of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.
+### ExGaussian Model
+
-### Wald
+### Wald Model
Moe from statistical models that *describe* to models that *generate* RT-like data.
-### Generative Models (DDM)
+## Generative Models (DDM)
Use DDM as a case study to introduce generative models
-### Other Models (LBA, LNR)
+## Other Models (LBA, LNR)
## Additional Resources
diff --git a/content/_freeze/4_rt/execute-results/html.json b/content/_freeze/4_rt/execute-results/html.json
index 6969ef0..fddac1f 100644
--- a/content/_freeze/4_rt/execute-results/html.json
+++ b/content/_freeze/4_rt/execute-results/html.json
@@ -1,10 +1,10 @@
{
- "hash": "26c45ef47fe7e81fdb051963e24f2bf1",
+ "hash": "80e397fa7a24ec91634f00cf896c2e90",
"result": {
"engine": "jupyter",
- "markdown": "# Reaction Times\n\n\nThis repository contain the following vignettes:\n\n- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)\n- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)\n\n## Descriptive Models (ExGaussian, LogNormal, Wald)\n\n### Modelling RT with a Bayesian Linear Model\n\n#### The Data\n\n::: {#718393b9 .cell execution_count=1}\n``` {.julia .cell-code}\nusing Downloads, CSV, DataFrames\nusing Turing, Distributions\nusing CairoMakie\n\ndf = CSV.read(Downloads.download(\"https://raw.githubusercontent.com/RealityBending/DoggoNogo/main/study1/data/data_game.csv\"), DataFrame)\nfirst(df, 10)\n```\n\n::: {.cell-output .cell-output-display execution_count=2}\n```{=html}\n1 0.517 1.638 1 S002 S1 NA Positive 0 0 2 0.433 2.282 2 S002 S1 0.475 Positive 1 0 3 0.533 0.739 3 S002 S1 0.494 Negative 2 0 4 0.35 0.602 4 S002 S1 0.458 Positive 0 1 5 0.517 1.951 5 S002 S1 0.47 Negative 1 0 6 0.383 0.952 6 S002 S1 0.456 Positive 0 1 7 0.383 0.875 7 S002 S1 0.445 Positive 1 0 8 0.417 2.662 8 S002 S1 0.442 Positive 2 0 9 0.367 2.8 9 S002 S1 0.433 Positive 3 0 10 0.417 0.258 10 S002 S1 0.432 Positive 4 0
Quantiles\n parameters 2.5% 97.5% \n Symbol Float64 Float64 \n σ² 0.0662 0.0741\n intercept 0.3432 0.3485\n slope_isi -0.0215 -0.0188\n \n```\n:::\n\n:::\n:::\n\n\n::: {.callout-tip title=\"Code Tip\"}\nWe first initialize the model by passing the `RT` and `ISI` columns.\n:::\n\n#### Posterior Predictive Check\n\n::: {#c0d701ef .cell execution_count=3}\n``` {.julia .cell-code}\npred = predict(model_linear([(missing) for i in 1:length(df.RT)]; isi=df.ISI), chain_linear)\npred = Array(pred)\n```\n:::\n\n\n::: {#5a9bdace .cell fig-height='7' fig-width='10' execution_count=4}\n``` {.julia .cell-code}\nf = Figure()\nax = Axis(f[1, 1], title=\"Predicted Data by Linear Model\",\n xlabel=\"RT (s)\",\n ylabel=\"Distribution\",\n yticksvisible=false,\n xticksvisible=false,\n yticklabelsvisible=false)\n\nCairoMakie.density!(df.RT, color=\"grey\")\nfor i in 1:length(chain_linear)\n lines!(ax, Makie.KernelDensity.kde(pred[:, i]), color=\"orange\", alpha=0.1)\nend\nCairoMakie.ylims!(ax, (0, nothing))\nf\n```\n\n::: {.cell-output .cell-output-stderr}\n```\n┌ Warning: Found `resolution` in the theme when creating a `Scene`. The `resolution` keyword for `Scene`s and `Figure`s has been deprecated. Use `Figure(; size = ...` or `Scene(; size = ...)` instead, which better reflects that this is a unitless size and not a pixel resolution. The key could also come from `set_theme!` calls or related theming functions.\n└ @ Makie C:\\Users\\domma\\.julia\\packages\\Makie\\VRavR\\src\\scenes.jl:220\n```\n:::\n\n::: {.cell-output .cell-output-display execution_count=5}\n{}\n:::\n:::\n\n\n### The Problem with Linear Models\n\nReaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the \"best fitting\" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\\mu$ and standard deviation $\\sigma$) are not good descriptors of the data.\n\n\n\n> Linear models try to find the best fitting Normal distribution for the data. However, for reaction times, even the best fitting Normal distribution (in red) does not capture well the actual data (in grey).\n\nA popular mitigation method to account for the non-normality of RTs is to transform the data, using for instance the popular *log-transform*. \nHowever, this practice should be avoided as it leads to various issues, including loss of power and distorted results interpretation [@lo2015transform; @schramm2019reaction].\nInstead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.\n\n\n### Shifted LogNormal Models\n\nOne of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.\n\n\n### Wald\n\nMoe from statistical models that *describe* to models that *generate* RT-like data.\n\n### Generative Models (DDM)\n\nUse DDM as a case study to introduce generative models\n\n### Other Models (LBA, LNR)\n\n\n## Additional Resources\n\n- [**Lindelov's overview of RT models**](https://lindeloev.github.io/shiny-rt/): An absolute must-read.\n- [**De Boeck & Jeon (2019)**](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00102/full): A paper providing an overview of RT models.\n- [https://github.com/vasishth/bayescogsci](https://github.com/vasishth/bayescogsci)\n\n",
+ "markdown": "# Reaction Times\n\n\nThis repository contain the following vignettes:\n\n- [**Drift Diffusion Model (DDM) in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/ddm.html)\n- [**Ex-Gaussian models in R: A Tutorial**](https://dominiquemakowski.github.io/easyRT/articles/exgaussian.html)\n\n## The Data\n\nData from @wagenmakers2008diffusion - Experiment 1.\nWe excluded all trials with uninterpretable response time [see @theriault2024check] such as too fast response (<180 ms) and too slow response (>2 sec).\n\n::: {#0832bae0 .cell execution_count=1}\n``` {.julia .cell-code code-fold=\"false\"}\nusing Downloads, CSV, DataFrames\nusing Turing, Distributions, SequentialSamplingModels\nusing CairoMakie\n\ndf = CSV.read(Downloads.download(\"https://raw.githubusercontent.com/DominiqueMakowski/CognitiveModels/main/data/wagenmakers2008.csv\"), DataFrame)\nfirst(df, 10)\n```\n\n::: {.cell-output .cell-output-display execution_count=2}\n```{=html}\n1 1 Speed 0.7 false Low 2 1 Speed 0.392 true Very Low 3 1 Speed 0.46 false Very Low 4 1 Speed 0.455 false Very Low 5 1 Speed 0.505 true Low 6 1 Speed 0.773 false High 7 1 Speed 0.39 false High 8 1 Speed 0.587 true Low 9 1 Speed 0.603 false Low 10 1 Speed 0.435 false High
Quantiles\n parameters 2.5% 97.5% \n Symbol Float64 Float64 \n σ² 0.1651 0.1699\n intercept 0.5072 0.5166\n slope_accuracy 0.1327 0.1451\n \n```\n:::\n\n:::\n:::\n\n\nThe effect of Condition is significant, people are on average slower (higher RT) when condition is `\"Accuracy\"`.\nBut is our model good?\n\n#### Posterior Predictive Check\n\n::: {#be300190 .cell execution_count=5}\n``` {.julia .cell-code}\npred = predict(model_linear([(missing) for i in 1:length(df.RT)], condition=df.Accuracy), chain_linear)\npred = Array(pred)\n```\n:::\n\n\n::: {#07909a8b .cell fig-height='7' fig-width='10' execution_count=6}\n``` {.julia .cell-code}\nfig = plot_distribution(df, \"Predictions made by Linear Model\")\nfor i in 1:length(chain_linear)\n lines!(Makie.KernelDensity.kde(pred[:, i]), color=ifelse(df.Accuracy[i] == 1, \"#388E3C\", \"#D32F2F\"), alpha=0.1)\nend\nfig\n```\n\n::: {.cell-output .cell-output-stderr}\n```\n┌ Warning: Found `resolution` in the theme when creating a `Scene`. The `resolution` keyword for `Scene`s and `Figure`s has been deprecated. Use `Figure(; size = ...` or `Scene(; size = ...)` instead, which better reflects that this is a unitless size and not a pixel resolution. The key could also come from `set_theme!` calls or related theming functions.\n└ @ Makie C:\\Users\\domma\\.julia\\packages\\Makie\\VRavR\\src\\scenes.jl:220\n```\n:::\n\n::: {.cell-output .cell-output-display execution_count=7}\n{}\n:::\n:::\n\n\n#### The Problem with Linear Models\n\nReaction time (RTs) have been traditionally modeled using traditional linear models and their derived statistical tests such as *t*-test and ANOVAs. Importantly, linear models - by definition - will try to predict the *mean* of the outcome variable by estimating the \"best fitting\" *Normal* distribution. In the context of reaction times (RTs), this is not ideal, as RTs typically exhibit a non-normal distribution, skewed towards the left with a long tail towards the right. This means that the parameters of a Normal distribution (mean $\\mu$ and standard deviation $\\sigma$) are not good descriptors of the data.\n\n\n\n> Linear models try to find the best fitting Normal distribution for the data. However, for reaction times, even the best fitting Normal distribution (in red) does not capture well the actual data (in grey).\n\nA popular mitigation method to account for the non-normality of RTs is to transform the data, using for instance the popular *log-transform*. \nHowever, this practice should be avoided as it leads to various issues, including loss of power and distorted results interpretation [@lo2015transform; @schramm2019reaction].\nInstead, rather than applying arbitrary data transformation, it would be better to swap the Normal distribution used by the model for a more appropriate one that can better capture the characteristics of a RT distribution.\n\n\n### Shifted LogNormal Model\n\nOne of the obvious candidate alternative to the log-transformation would be to use a model with a Log-transformed Normal distribution.\n\n### ExGaussian Model\n\n\n### Wald Model\n\nMoe from statistical models that *describe* to models that *generate* RT-like data.\n\n## Generative Models (DDM)\n\nUse DDM as a case study to introduce generative models\n\n## Other Models (LBA, LNR)\n\n\n## Additional Resources\n\n- [**Lindelov's overview of RT models**](https://lindeloev.github.io/shiny-rt/): An absolute must-read.\n- [**De Boeck & Jeon (2019)**](https://www.frontiersin.org/articles/10.3389/fpsyg.2019.00102/full): A paper providing an overview of RT models.\n- [https://github.com/vasishth/bayescogsci](https://github.com/vasishth/bayescogsci)\n\n",
"supporting": [
- "4_rt_files"
+ "4_rt_files\\figure-html"
],
"filters": [],
"includes": {
diff --git a/content/_freeze/4_rt/figure-html/cell-4-output-2.svg b/content/_freeze/4_rt/figure-html/cell-4-output-2.svg
new file mode 100644
index 0000000..3bf0846
--- /dev/null
+++ b/content/_freeze/4_rt/figure-html/cell-4-output-2.svg
@@ -0,0 +1,647 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/content/_freeze/4_rt/figure-html/cell-7-output-2.svg b/content/_freeze/4_rt/figure-html/cell-7-output-2.svg
new file mode 100644
index 0000000..c5d19fb
--- /dev/null
+++ b/content/_freeze/4_rt/figure-html/cell-7-output-2.svg
@@ -0,0 +1,1695 @@
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/data/make_data.R b/data/make_data.R
index bd794c8..b418d9e 100644
--- a/data/make_data.R
+++ b/data/make_data.R
@@ -21,6 +21,6 @@ df <- rtdists::speed_acc |>
.keep = "none") |>
filter(Participant %in% c(1:6))
-write.csv(df, "wagenmakers2008_speedaccuracy.csv", row.names = FALSE)
+write.csv(df, "wagenmakers2008.csv", row.names = FALSE)
# summary(df)