{ "cells": [ { "cell_type": "markdown", "id": "ae1413fa", "metadata": {}, "source": [ "# Specifying Multivariate Effect Size\n", "\n", "In basic statistics, **Cohen’s d′** tells you how far apart two class means sit in units of the data’s standard deviation:\n", "\n", "$$\n", "d' = \\frac{\\mu_{1} - \\mu_{2}}{\\sigma}\n", "$$\n", "\n", "In a **multivariate** setting—when you have many channels or features—the same intuition holds, but now “distance” depends not only on the raw mean difference and noise level, but also on:\n", "\n", "- **How many features** carry the signal \n", "- **How noisy or correlated** those features are \n", "- **How the signal is spatially distributed** (uniformly across channels or concentrated in a subset) \n", "\n", "Our simulator takes your requested $d'$ and **automatically boosts or scales down** the raw amplitude of the injected pattern so that, no matter what noise level, covariance structure, or channel count you choose, the true multivariate Cohen’s $d'$ equals what you asked for—and a Bayes-optimal classifier would achieve the corresponding theoretical accuracy: \n", "\n", "$$\\Phi(d'/2)$$\n" ] }, { "cell_type": "markdown", "id": "b02c9d80", "metadata": {}, "source": [ "## Trials per subject\n", "\n", "It’s well known that **more trials** generally improve decoding accuracy—but only **up to a point**. For any fixed effect size $d'$, there is a **ceiling** on accuracy (e.g. $d'=1$ → ~69 %). \n", "\n", "To illustrate this, we can simulate data with 20, 40, 80 or 160 trials per condition:" ] }, { "cell_type": "code", "execution_count": 17, "id": "5ac358ea", "metadata": {}, "outputs": [], "source": [ "# Simulation parameters\n", "import numpy as np\n", "import pandas as pd\n", "import matplotlib.pyplot as plt\n", "from multisim import Simulator\n", "import matplotlib.pyplot as plt\n", "from sklearn.svm import SVC\n", "from sklearn.pipeline import make_pipeline\n", "from sklearn.preprocessing import StandardScaler\n", "from mne.decoding import SlidingEstimator, cross_val_multiscore\n", "from scipy.ndimage import gaussian_filter1d\n", "from mne.stats import permutation_cluster_1samp_test, bootstrap_confidence_interval\n", "from scipy.stats import ttest_1samp\n", "\n", "# Add descriptors:\n", "cond_names = [\"category\", \"attention\"]\n", "mapping = {\n", " \"category\": {1: \"face\", -1: \"object\"},\n", " \"attention\": {1: \"attended\", -1: \"unattended\"},\n", "}\n", "\n", "n_channels = 32 # EEG system with 32 electrodes\n", "n_subjects = 20 # Recording from 20 subjects\n", "noise_std = 1 / 8 # Variance of the data\n", "ch_cov = None # Assuming that the data of each sensor are independent\n", "sfreq = 50 # Simulating data at 50Hz\n", "tmin = -0.25\n", "tmax = 1.0\n", "t = np.arange(0, 1, 1 / sfreq) # time vector (in seconds)\n", "kernel = None\n", "intersub_noise_std = 0\n", "\n", "# Specifying the effects:\n", "effects = [\n", " {\"condition\": 'category', \"windows\": [0.1, 0.2], \"effect_size\": 1}, \n", " {\"condition\": 'attention', \"windows\": [0.3, 0.4], \"effect_size\": 1}\n", " ] # Packaging them in a list to pass to the simulator class\n" ] }, { "cell_type": "code", "execution_count": 18, "id": "b1f3be64", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Text(0.5, 1.0, 'Sensor space decoding')" ] }, "execution_count": 18, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Creating the design matrix of our 2 by two balanced design:\n", "n_trials = [40, 80, 120, 160, 320]\n", "\n", "# Create the classifier:\n", "clf = make_pipeline(StandardScaler(), SVC())\n", "\n", "# Time resolved\n", "time_decod = SlidingEstimator(clf, n_jobs=None, scoring=\"roc_auc\", verbose=True)\n", "\n", "fig, ax = plt.subplots()\n", "for n in n_trials:\n", " # Design matrix:\n", " X = pd.DataFrame(np.array([[1, 1, -1, -1] * int(n / 4), [1, -1] * int(n / 2)]).T, \n", " columns=cond_names)\n", " # Simulate the data:\n", " sims = Simulator(\n", " X, # Design matrix\n", " effects, # Effects to simulate\n", " noise_std, # Observation noise\n", " n_channels, # Number of channelss\n", " 1, # Number of subjects\n", " tmin,\n", " tmax, # Start and end of epochs\n", " sfreq, # Sampling frequency of the data\n", " ch_cov=ch_cov, # Spatial covariance of the data\n", " kern=kernel, # Temporal kernel,\n", " intersub_noise_std=intersub_noise_std, # Intersubject variability\n", " )\n", " epochs = sims.export_to_mne(X=X.copy(), mapping=mapping)\n", "\n", " # Perform decoding:\n", " # Extract the data:\n", " data = epochs[0].get_data()\n", " # Decode faces vs. objects:\n", " cate_lbl = np.array([mapping[\"category\"][val] for val in X.to_numpy()[:, 0]])\n", " scores_category = np.mean(\n", " cross_val_multiscore(\n", " time_decod, data, cate_lbl, cv=5, n_jobs=-1, verbose=\"WARNING\"\n", " ),\n", " axis=0,\n", " )\n", "\n", " # Plot the results:\n", " ax.plot(\n", " epochs[0].times,\n", " gaussian_filter1d(scores_category, 1),\n", " label=f\"N trials = {n}\",\n", " alpha=0.8,\n", " )\n", "\n", "ax.axhline(0.5, color=\"k\", linestyle=\"--\", label=\"chance\")\n", "ax.set_xlim([epochs[0].times[0], epochs[0].times[-1]])\n", "ax.set_xlabel(\"Times\")\n", "ax.set_ylabel(\"AUC\") # Area Under the Curve\n", "ax.legend()\n", "ax.axvline(0.0, color=\"k\", linestyle=\"-\")\n", "ax.set_title(\"Sensor space decoding\")" ] }, { "cell_type": "markdown", "id": "b0967877", "metadata": {}, "source": [ "As we can see, increasing the number of trials only helps until a certain point. There is not so big of a difference between 160 and 320 trials. \n", "\n", "This has important implications. You may think that if there is a pattern in your data, if you have an absurdly high amount of trials, you will be able to get the decoding accuracy to a 100%. That's not true, the max decoding accuracy is capped by the effect size. Even if you had 2000 trials, you wouldn't get higher in the previous simulation. Let's simulate that data set just to prove the point:" ] }, { "cell_type": "code", "execution_count": 19, "id": "3910e08c", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Text(0.5, 1.0, 'Sensor space decoding')" ] }, "execution_count": 19, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Create the classifier:\n", "clf = make_pipeline(StandardScaler(), SVC())\n", "\n", "# Time resolved\n", "time_decod = SlidingEstimator(clf, n_jobs=None, scoring=\"roc_auc\", verbose=True)\n", "\n", "# Design matrix:\n", "X = pd.DataFrame(np.array([[1, 1, -1, -1] * int(2000 / 4), [1, -1] * int(2000 / 2)]).T,\n", " columns=cond_names)\n", "# Simulate the data:\n", "sims = Simulator(\n", " X, # Design matrix\n", " effects, # Effects to simulate\n", " noise_std, # Observation noise\n", " n_channels, # Number of channelss\n", " 1, # Number of subjects\n", " tmin,\n", " tmax, # Start and end of epochs\n", " sfreq, # Sampling frequency of the data\n", " ch_cov=ch_cov, # Spatial covariance of the data\n", " kern=kernel, # Temporal kernel,\n", " intersub_noise_std=intersub_noise_std, # Intersubject variability\n", ")\n", "epochs = sims.export_to_mne(X=X.copy(), mapping=mapping)\n", "\n", "# Perform decoding:\n", "# Extract the data:\n", "data = epochs[0].get_data()\n", "# Decode faces vs. objects:\n", "cate_lbl = np.array([mapping[\"category\"][val] for val in X.to_numpy()[:, 0]])\n", "scores_category = np.mean(\n", " cross_val_multiscore(\n", " time_decod, data, cate_lbl, cv=5, n_jobs=-1, verbose=\"WARNING\"\n", " ),\n", " axis=0,\n", ")\n", "\n", "# Plot the results:\n", "fig, ax = plt.subplots()\n", "ax.plot(\n", " epochs[0].times,\n", " gaussian_filter1d(scores_category, 1),\n", " label=f\"N trials = {2000}\",\n", " alpha=0.8,\n", ")\n", "\n", "ax.axhline(0.5, color=\"k\", linestyle=\"--\", label=\"chance\")\n", "ax.set_xlim([epochs[0].times[0], epochs[0].times[-1]])\n", "ax.set_xlabel(\"Times\")\n", "ax.set_ylabel(\"AUC\") # Area Under the Curve\n", "ax.legend()\n", "ax.axvline(0.0, color=\"k\", linestyle=\"-\")\n", "ax.set_title(\"Sensor space decoding\")" ] }, { "cell_type": "markdown", "id": "b657e29b", "metadata": {}, "source": [ "## Number of channels\n", "\n", "While the number of trials has an impact on decoding accuracy, the number of channels shouldn't. That is because under the hood, we are normalizing the multivariate $d'$ by the number of features (and to be precise, the covariance matrix of the data). Let's simulate data set with different number of channels:" ] }, { "cell_type": "code", "execution_count": 20, "id": "ee3fc15e", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Text(0.5, 1.0, 'Sensor space decoding')" ] }, "execution_count": 20, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Creating the design matrix of our 2 by two balanced design:\n", "n_channels = [32, 64, 128, 256]\n", "\n", "# Create the classifier:\n", "clf = make_pipeline(StandardScaler(), SVC())\n", "\n", "# Time resolved\n", "time_decod = SlidingEstimator(clf, n_jobs=None, scoring=\"roc_auc\", verbose=True)\n", "\n", "fig, ax = plt.subplots()\n", "for n in n_channels:\n", " # Design matrix:\n", " X = pd.DataFrame(np.array([[1, 1, -1, -1] * 40, [1, -1] * 80]).T,\n", " columns=cond_names)\n", " # Simulate the data:\n", " sims = Simulator(\n", " X, # Design matrix\n", " effects, # Effects to simulate\n", " noise_std, # Observation noise\n", " n, # Number of channelss\n", " 1, # Number of subjects\n", " tmin,\n", " tmax, # Start and end of epochs\n", " sfreq, # Sampling frequency of the data\n", " ch_cov=ch_cov, # Spatial covariance of the data\n", " kern=kernel, # Temporal kernel,\n", " intersub_noise_std=intersub_noise_std, # Intersubject variability\n", " )\n", " epochs = sims.export_to_mne(X=X.copy(), mapping=mapping)\n", "\n", " # Perform decoding:\n", " # Extract the data:\n", " data = epochs[0].get_data()\n", " # Decode faces vs. objects:\n", " cate_lbl = np.array([mapping[\"category\"][val] for val in X.to_numpy()[:, 0]])\n", " scores_category = np.mean(\n", " cross_val_multiscore(\n", " time_decod, data, cate_lbl, cv=5, n_jobs=-1, verbose=\"WARNING\"\n", " ),\n", " axis=0,\n", " )\n", "\n", " # Plot the results:\n", " ax.plot(\n", " epochs[0].times,\n", " gaussian_filter1d(scores_category, 1),\n", " label=f\"N channels = {n}\",\n", " alpha=0.8,\n", " )\n", "\n", "ax.axhline(0.5, color=\"k\", linestyle=\"--\", label=\"chance\")\n", "ax.set_xlim([epochs[0].times[0], epochs[0].times[-1]])\n", "ax.set_xlabel(\"Times\")\n", "ax.set_ylabel(\"AUC\") # Area Under the Curve\n", "ax.legend()\n", "ax.axvline(0.0, color=\"k\", linestyle=\"-\")\n", "ax.set_title(\"Sensor space decoding\")" ] }, { "cell_type": "markdown", "id": "431255e3", "metadata": {}, "source": [ "Surprisingly, the decoding accuracy decreases with increased number of channels, which shouldn't happen based on our normalization. However, this behavior is expected. When we increase the number of features, we need to decrease the magnitude of the activation in each channel that are part of the pattern to yield a similar theoretical effect size. If we kept the magnitude the same, we would increase the effect size when increasing the number of channels. However, by decreasing the magnitude in each channel, we make it harder to discriminate from the noise of the signal, explaining the decrease in decoding accuracy. \n", "\n", "- **More channels** → the simulator spreads the same total signal across more sensors → **lower per-channel amplitude** \n", "- Lower per-channel SNR makes each feature harder to decode \n", "- As a result, accuracy **drops** with channel count unless you also increase trials or concentrate the signal in fewer channels \n", "\n", "This highlights a key trade-off: if your true brain signal lives in many sensors, you need **enough trials** (or stronger regularization/dimensionality reduction) to recover that tiny per-channel effect.\n" ] }, { "cell_type": "markdown", "id": "d3b56074", "metadata": {}, "source": [ "## Effect size, subjects & trials\n", "\n", "When planning a study you must decide:\n", "\n", "1. **How many trials per subject?** \n", "2. **How many subjects overall?**\n", "\n", "\n", "There is an indirect trade off at play between these two, but one doesn't compensate for the other. \n", "- Too few trials per subject → you can’t reliably estimate decoding weights, and within-subject noise remains high. \n", "- Too few subjects → group-level statistics lack power, even if individual accuracies are above chance.\n", "\n", "Crucially, **effect size** $d'$ sets a hard **upper limit** on decoding accuracy. Even with thousands of trials, you cannot exceed $\\Phi(d'/2)$. And so for small effect sizes, you need many subjects to detect an effect. Let's simulate a data set with 8 subjects with an effect size of 0.2 to illustrate that point:" ] }, { "cell_type": "code", "execution_count": 21, "id": "4a5b3ffa", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Using a threshold of 1.894579\n", "stat_fun(H1): min=-3.498215304828258 max=3.0838380071597933\n", "Running initial clustering …\n", "Found 4 clusters\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "c544d1a8e9e14d25962a7715a0119eee", "version_major": 2, "version_minor": 0 }, "text/plain": [ " 0%| | Permuting (exact test) : 0/255 [00:00" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Design matrix:\n", "X = pd.DataFrame(np.array([[1, 1, -1, -1] * int(320 / 4), [1, -1] * int(320 / 2)]).T,\n", " columns=cond_names)\n", "# Specifying the effects:\n", "effects = [\n", " {\"condition\": 'category', \"windows\": [0.1, 0.2], \"effect_size\": 0.2}, \n", " {\"condition\": 'attention', \"windows\": [0.3, 0.4], \"effect_size\": 0.2}\n", " ] # Packaging them in a list to pass to the simulator class\n", "\n", "# Simulate the data:\n", "sims = Simulator(\n", " X, # Design matrix\n", " effects, # Effects to simulate\n", " noise_std, # Observation noise\n", " 32, # Number of channelss\n", " 8, # Number of subjects\n", " tmin,\n", " tmax, # Start and end of epochs\n", " sfreq, # Sampling frequency of the data\n", " ch_cov=ch_cov, # Spatial covariance of the data\n", " kern=kernel, # Temporal kernel,\n", " intersub_noise_std=1 / 32, # Intersubject variability\n", ")\n", "epochs = sims.export_to_mne(X=X.copy(), mapping=mapping)\n", "\n", "# Extract labels:\n", "cate_lbl = np.array([mapping[\"category\"][val] for val in X.to_numpy()[:, 0]])\n", "\n", "# Loop through each subject:\n", "scores_category = []\n", "for epo in epochs:\n", " # Extract the data:\n", " data = epo.get_data()\n", " # Classification of category\n", " scores_category.append(\n", " np.mean(\n", " cross_val_multiscore(\n", " time_decod, data, cate_lbl, cv=5, n_jobs=-1, verbose=\"WARNING\"\n", " ),\n", " axis=0,\n", " )\n", " )\n", "\n", "scores_category = np.array(scores_category)\n", "\n", "# Group level statistics:\n", "# Cluster based permutation test for the category:\n", "T_obs, clusters, cluster_p_values, H0 = permutation_cluster_1samp_test(\n", " scores_category - 0.5,\n", " n_permutations=1024,\n", " tail=1,\n", " out_type=\"mask\",\n", " verbose=True,\n", ")\n", "sig_mask_cate = np.zeros(len(epochs[0].times), dtype=bool)\n", "for c, p_val in enumerate(cluster_p_values):\n", " if p_val < 0.05:\n", " sig_mask_cate[clusters[c]] = True\n", "# Compute the confidence intervals:\n", "ci_low_cate, ci_up_cate = bootstrap_confidence_interval(scores_category)\n", "\n", "fig, ax = plt.subplots()\n", "ax.plot(epochs[0].times, np.mean(scores_category, axis=0), label=\"category\", color=\"b\")\n", "ax.fill_between(epochs[0].times, ci_low_cate, ci_up_cate, alpha=0.3, color=\"b\")\n", "ax.plot(\n", " epochs[0].times[sig_mask_cate],\n", " np.ones(np.sum(sig_mask_cate)) * 0.4,\n", " marker=\"o\",\n", " linestyle=\"None\",\n", " color=\"b\",\n", ")\n", "ax.axhline(0.5, color=\"k\", linestyle=\"--\", label=\"chance\")\n", "ax.set_xlim([epochs[0].times[0], epochs[0].times[-1]])\n", "ax.set_xlabel(\"Times\")\n", "ax.legend()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "81c522aa", "metadata": {}, "source": [ "Crucially, you should decide on the number of trials per subject to make sure you approach the ceiling decoding accuracy based on your effect size, and you should select the number of subjects to be able to detect an effect of that magnitude in your sample. \n" ] }, { "cell_type": "markdown", "id": "9eaf2392", "metadata": {}, "source": [ "## Putting it all together\n", "To see how all these things play together, we will illustrate how to modulate number of subjects and trials at a given effect size (0.2) to show our toolbox can be applied to determine the optimal experimental design:" ] }, { "cell_type": "code", "execution_count": 22, "id": "e2884a6d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "Text(0, 0.5, 'Number of subjects')" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "n_trials = [20, 40, 80, 120, 140, 160, 180, 200]\n", "n_subjects = [5, 10, 15, 20, 25, 30, 35, 40]\n", "\n", "# Specifying the effects:\n", "effects = [\n", " {\"condition\": 'category', \"windows\": [0.1, 0.2], \"effect_size\": 0.2}, \n", " {\"condition\": 'attention', \"windows\": [0.3, 0.4], \"effect_size\": 0.2}\n", " ] # Packaging them in a list to pass to the simulator class\n", "\n", "\n", "tstat = np.zeros((len(n_subjects), len(n_trials)))\n", "for i_s, ns in enumerate(n_subjects):\n", " for i_t, nt in enumerate(n_trials):\n", " # Design matrix:\n", " X = pd.DataFrame(np.array([[1, 1, -1, -1] * int(nt / 4), [1, -1] * int(nt / 2)]).T,\n", " columns=cond_names)\n", " # Simulate the data:\n", " sims = Simulator(\n", " X, # Design matrix\n", " effects, # Effects to simulate\n", " noise_std, # Observation noise\n", " 32, # Number of channels\n", " ns, # Number of subjects\n", " tmin,\n", " tmax, # Start and end of epochs\n", " sfreq, # Sampling frequency of the data\n", " ch_cov=ch_cov, # Spatial covariance of the data\n", " kern=kernel, # Temporal kernel,\n", " intersub_noise_std=intersub_noise_std, # Intersubject variability\n", " )\n", " epochs = sims.export_to_mne(X=X.copy(), mapping=mapping)\n", "\n", " # Extract labels:\n", " cate_lbl = np.array([mapping[\"category\"][val] for val in X.to_numpy()[:, 0]])\n", "\n", " # Loop through each subject:\n", " scores_category = []\n", " for epo in epochs:\n", " # Extract the data:\n", " data = epo.get_data()\n", " # Classification of category\n", " scores_category.append(\n", " np.mean(\n", " cross_val_multiscore(\n", " time_decod, data, cate_lbl, cv=5, n_jobs=-1, verbose=\"WARNING\"\n", " ),\n", " axis=0,\n", " )\n", " )\n", " scores_category = np.array(scores_category)\n", "\n", " tstat[i_s, i_t] = np.max(ttest_1samp(scores_category, 0.5).statistic)\n", "\n", "fig, ax = plt.subplots()\n", "im = ax.imshow(\n", " tstat,\n", " aspect=\"auto\",\n", " cmap=\"RdYlBu_r\",\n", " extent=[n_trials[0], n_trials[-1], n_subjects[0], n_subjects[-1]],\n", ")\n", "cb = plt.colorbar(im, label=\"t-stat\")\n", "ax.set_title(\"Statistical difference at effect size of 0.2\")\n", "ax.set_xlabel(\"Number of trials\")\n", "ax.set_ylabel(\"Number of subjects\")" ] }, { "cell_type": "markdown", "id": "d6dc048d", "metadata": {}, "source": [ "## Conclusion and Practical Recommendations\n", "\n", "The toy examples above highlight how **effect size**, **trials per subject**, and **number of subjects** jointly determine your ability to detect a multivariate pattern:\n", "\n", "- A larger effect size $d'$ raises the theoretical accuracy ceiling. \n", "- More trials reduce the variability of your within-subject decoder and help you approach that ceiling. \n", "- More subjects reduce the uncertainty of the group-level mean and increase statistical power.\n", "\n", "Importantly, these illustrations assumed **no between-subject variability**, an arbitrary noise level, and statistically independent channels. Real EEG/MEG data exhibit spatial correlations, nonstationary noise, and individual differences that will shift the precise numbers.\n", "\n", "Use these simulations as **guiding examples**, not rigid rules. Tailor the parameters—noise level, spatial covariance, effect size, trial counts, subject counts—to match your planned experiment. By exploring your specific design _in silico_, you can make **principled decisions** about how many trials and participants you truly need to achieve reliable decoding at your desired effect size.\n", "\n", "Our toolbox makes these trade-offs transparent and interactive, so you can optimize your study design before collecting a single real data point.\n", "\n", "In the next tutorial, we provide all the mathematical details behind our simulation." ] } ], "metadata": { "kernelspec": { "display_name": ".venv", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.13.0" } }, "nbformat": 4, "nbformat_minor": 5 }