{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Modeling the pupillometric signal\n", "\n", "**WARNING**: This functionality is **experimental**. Some of the provided algorithms are unpublished (or in the process of being published) and may not work well.\n", "\n", "The idea behind the algorithms is detailed in this notebook (slides from a symposium talk):\n", "\n", "- [Estimation of tonic and phasic pupillometric signals](symp_talk_uit2019.html)\n", "\n", "This notebooks also includes results from a simulation study showing the superiority of the algorithm to more traditional approaches." ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import sys\n", "sys.path.insert(0,\"..\") ## not necessary if pypillometry is installed on your system\n", "import pypillometry as pp\n", "import pylab as plt \n", "import numpy as np\n", "import scipy" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "It can sometimes be useful to think about the pupillometric signal as being composed of different components. One comon assumption, based on the finding that the pupil reflects activity in the norepinephrinergic system, consists of slow, tonic (baseline) and faster, phasic (response) fluctuations. \n", "\n", "`pypillometry` comes with functions to create artificial data. These functions are built on such a model where stimulus- or event-induced responses are superimposed on a slow, baseline-like component:" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "faked=pp.create_fake_pupildata(ntrials=20)\n", "plt.figure(figsize=(15,5))\n", "faked.plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The orange line is the tonic, baseline-component. At the timing of each event (grey lines), a scaled version of a \"response-kernel\" (Hoeks & Levelt, 1993) is added. Finally, random noise is added on top of the modeled data. The challenge is to extract both the size of the response as well as the baseline value at each stimulus when only using the raw pupillometric data (the blue line)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Traditional approaches for disentangling tonic and phasic components\n", "\n", "One common way to solve this problem and analyse pupillometric data on the trial-by-trial level is therefore to extract the average pupillometric signal just before a stimulus (as a measure of the baseline signal) and just after the stimulus (as a measure of the pupil's response. In `pypillometry` this functionality is implemented by `PupilData.stat_per_event()` which allows to extract a summary of the signal relative to the events in the dataset.\n", "\n", "For example, the following code extracts\n", "- the average signal in the time-window from 200 ms before each event until the timing of the event itself (`0`) as baseline\n", "- the average signal in the time-window from 800 ms after each event until 1200 ms after the event as a measure of the response \n", "\n", "often, the baseline is subtracted from the response:" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(array([5497.22338969, 5505.84230613, 5504.95766177, 5502.83389561,\n", " 5519.26795734, 5500.93556967, 5521.47146776, 5501.94256506,\n", " 5508.15286254, 5502.7746573 , 5507.29647592, 5516.96014844,\n", " 5510.85424104, 5515.54653477, 5518.08930806, 5504.37508904,\n", " 5526.7414329 , 5507.19514677, 5507.87170839, 5509.50337549,\n", " 5502.12178611, 5505.54231241, 5502.98219096, 5500.63723808,\n", " 5500.5047791 , 5501.95173174, 5501.98036015, 5499.75653658,\n", " 5505.69011558, 5507.55745466, 5516.96108655, 5510.85644785,\n", " 5511.06295341, 5506.1358251 , 5504.84753547, 5506.00476406,\n", " 5519.09558608, 5504.34079763, 5510.49458483, 5502.97984779]),\n", " array([ 7.87399393, -2.33877461, -3.16018652, 13.57307738,\n", " -18.91961534, 18.66346135, -20.29346888, 4.90085667,\n", " -5.43644688, 3.31891129, 8.1205739 , -6.9479222 ,\n", " 3.82577502, 1.20349654, -14.09887951, 21.19385231,\n", " -20.06858291, 0.25081461, 0.72956157, -4.55229805,\n", " 3.19235136, -2.93086936, 14.50920921, 19.22592926,\n", " 16.11029615, 5.47806116, 5.20489091, 6.21127191,\n", " 5.57229406, 8.38191287, -7.08207666, 3.79996363,\n", " -1.66852599, 15.08431094, 20.87224058, 6.45305353,\n", " -15.01720952, 6.72919367, -0.8283899 , 2.7633361 ]))" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "baseline=faked.stat_per_event( (-200, 0 ) )\n", "response=faked.stat_per_event( ( 800, 1200) )-baseline\n", "baseline,response" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Note that `PupilData.stat_per_event()` supports selecting specific events, any summary-function (default is `numpy.mean`) and has functionality for handling missing data in the used time-windows, e.g.:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(array([5497.23314426, 5505.74146081, 5504.96472822, 5502.80653328,\n", " 5519.34855095, 5500.9466973 , 5521.61377533, 5501.9995312 ,\n", " 5508.18492429, 5502.80102799, 5507.32622705, 5516.99380643,\n", " 5510.86912668, 5515.56038628, 5518.10239105, 5504.39980492,\n", " 5526.80368604, 5507.20711887, 5507.81666783, 5509.55677362,\n", " 5502.05091102, 5505.53396316, 5502.99002577, 5500.60428626,\n", " 5500.49712934, 5501.99230174, 5502.01976366, 5499.76123856,\n", " 5505.68953437, 5507.56270336, 5517.00246874, 5510.86912668,\n", " 5511.03050428, 5506.11351642, 5504.79115636, 5505.76330355,\n", " 5519.40998699, 5504.28362604, 5510.45780027, 5503.00292442]),\n", " array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n", " 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n", " 0., 0., 0., 0., 0., 0.]))" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "faked.stat_per_event( (-200,0), event_select=\"event\", statfct=np.median, return_missing=\"nmiss\")" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Advance methods for tonic/phasic component estimation" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The simple method detailed above is appealing for its simplicity but has severe limitations. Most importantly, multiple overlapping pupillary responses can \"look like\" baseline-fluctuations when added together, thereby artificially inflating baseline-estimates particularly in cases where events are spaced closely in time (\"fast paradigms\"). For that reason, we developed specialized algorithms to disentangle tonic and phasic components of the pupillometric signal.\n", "\n", "This algorithm uses an iterative procedure to remove an initial estimate of the responses from the signal to continue to estimate the underlying baseline. Details about how this algorithm works and which parameters it supports are available in [this notebook](symp_talk_uit2019.html) and will be available in a forthcoming publication.\n", "\n", "In practice, the functionality is implemented in `PupilData.estimate_baseline()` and `PupilData.estimate_response()`. The response-estimation depends on the estimated baseline, hence the `estimate_baseline()` function should always be called first. In order to increase speed, we filter the data and downsample it to 50 Hz before running the baseline- and response-estimation functions" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "WARNING:pystan:Automatic Differentiation Variational Inference (ADVI) is an EXPERIMENTAL ALGORITHM.\n", "WARNING:pystan:ADVI samples may be found on the filesystem in the file `/var/folders/28/_ftmv1_n41n48znrymwflm940000gp/T/tmp0twejg2w/output.csv`\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "Using cached StanModel\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "WARNING:pystan:Automatic Differentiation Variational Inference (ADVI) is an EXPERIMENTAL ALGORITHM.\n", "WARNING:pystan:ADVI samples may be found on the filesystem in the file `/var/folders/28/_ftmv1_n41n48znrymwflm940000gp/T/tmpu2rv_yv0/output.csv`\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "MSG: optimizing both npar and tmax, might take a while...\n", "......................................................................................................................................................................................................................................................" ] } ], "source": [ "d=faked.lowpass_filter(cutoff=5)\\\n", " .downsample(fsd=50)\\\n", " .estimate_baseline()\\\n", " .estimate_response()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "Here, we allowed the two shape parameters `npar` and `tmax` to vary freely together with the amplitude of the responses. This allows an individualized shape of the pupillary response for each subject (which appears reasonable given the results in Hoeks & Levelt, 1993) but may also take a long time to optimize and potentially results in pathological solutions. In that case, one or both of the parameters can be fixed, for example to reasonable group-level values.\n", "\n", "After running these methods, the baseline is stored in the `PupilData.baseline` variable:\n" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "array([5498.57714444, 5498.6606226 , 5498.7422532 , ..., 5491.61364603,\n", " 5491.29801098, 5490.99052775])" ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" } ], "source": [ "d.baseline" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "and the estimated response in `PupilData.response_pars`:" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "data": { "text/plain": [ "{'npar': 11.872666808731674,\n", " 'npar_free': True,\n", " 'tmax': 899.6180651831687,\n", " 'tmax_free': True,\n", " 'coef': array([ 7.69816302, 4.28928702, 3.8821806 , 9.52004979, 3.60587198,\n", " 15.93467607, 4.31018922, 2.38161115, 3.5997683 , 5.8555845 ,\n", " 12.65606094, 7.67183007, 8.4533453 , 10.55249302, 1.67096352,\n", " 23.38023839, 5.53552921, 4.1318766 , 4.97613487, 6.11872951,\n", " 3.74244261, 2.41996112, 13.48671774, 4.22908343, 5.0185794 ,\n", " 0.14989147, 9.72097211, 3.30870393, 0. , 3.11073058,\n", " 0. , 0. , 0.17421374, 6.88414573, 0. ,\n", " 8.97973124, 2.15049561, 4.66517684, 8.81778523, 9.65322884]),\n", " 'bounds': {'npar': (1, 20), 'tmax': (100, 2000)}}" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "d.response_pars" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The resulting baseline-estimation and the estimated full model (baseline+response) can be plotted:\n" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": { "needs_background": "light" }, "output_type": "display_data" } ], "source": [ "plt.figure(figsize=(15,5))\n", "d.plot()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "The orange curve (estimated baseline) is still more wiggly than the real baseline (green) but give a considerable better estimate than a traditional method. The overall fit of the model (red) is excellent (which is to be expected, as this is simulated data). The misfit comes from randomly interspersed \"spurious\" events in the randomly generated data.\n", "\n", "We can quantify how well the novel baseline-estimation works relative to the traditional method by comparing it to the \"ground-truth\" which is available for artificial data." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We calculate the ground-truth and the traditional and novel estimates for each event-onset (=trial):" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [], "source": [ "real_baseline=pp.stat_event_interval(d.tx, d.sim_baseline, d.event_onsets, [0,0])\n", "real_response=d.sim_response_coef\n", "\n", "traditional_baseline=d.stat_per_event( (-200,0) )\n", "traditional_response=d.stat_per_event( ( 800,1200) )-traditional_baseline\n", "\n", "novel_baseline=pp.stat_event_interval(d.tx, d.baseline, d.event_onsets, [0,0])\n", "novel_response=d.response_pars[\"coef\"]" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "And compare them by means of the correlation of the estimated and mean values:" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Traditional method:\n", "Baseline: Corr(trad,real)= 0.37434828350870303\n", "Response: Corr(trad,real)= 0.38109869050417233\n", "\n", "\n", "Novel method:\n", "Baseline: Corr(nov, real)= 0.7623237267947978\n", "Response: Corr(nov, real)= 0.6876359509086555\n" ] } ], "source": [ "print(\"Traditional method:\")\n", "print(\"Baseline: Corr(trad,real)=\",scipy.stats.pearsonr(traditional_baseline, real_baseline)[0])\n", "print(\"Response: Corr(trad,real)=\",scipy.stats.pearsonr(traditional_response, real_response)[0])\n", "print(\"\\n\")\n", "print(\"Novel method:\")\n", "print(\"Baseline: Corr(nov, real)=\",scipy.stats.pearsonr(novel_baseline, real_baseline)[0])\n", "print(\"Response: Corr(nov, real)=\",scipy.stats.pearsonr(novel_response, real_response)[0])" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "We see that the correlations are much higher for the novel method when compared to the traditional methods. More sophisticated simulation studies are reported in [this notebook](symp_talk_uit2019.html)." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "\n", "The parameters for the baseline-estimation function are described in the API-documentation for `pypillometry.baseline.baseline_envelope_iter_bspline()` and, in more detail, in [this notebook](symp_talk_uit2019.html).\n", "\n", "The parameters for the response-estimation function are described in the API-documentation for `pypillometry.pupil.pupil_response()`." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.6" } }, "nbformat": 4, "nbformat_minor": 4 }