Videogames are emerging as a promising experimental paradigm in neuroimaging. Acquiring gameplay in a scanner remains challenging due to the lack of a scanner-compatible videogame controller that provides a similar experience to standard, commercial devices. In this paper, we introduce a videogame controller designed for use in the functional magnetic resonance imaging as well as magnetoencephalography. The controller is made exclusively of 3D-printed and commercially available parts. We evaluated the quality of our controller by comparing it to a non-MRI compatible controller that was kept outside the scanner. The comparison of response latencies showed reliable button press accuracies of adequate precision. Comparison of the subjects' motion during fMRI recordings of various tasks showed that the use of our controller did not increase the amount of motion produced compared to a regular MR compatible button press box. Motion levels during an ecological videogame task were of moderate amplitude. In addition, we found that the controller only had marginal effect on temporal SNR in fMRI, as well as on covariance between sensors in MEG, as expected due to the use of non-magnetic building materials. Finally, the reproducibility of the controller was demonstrated by having team members who were not involved in the design build a reproduction using only the documentation. This new videogame controller opens new avenues for ecological tasks in fMRI, including challenging videogames and more generally tasks with complex responses. The detailed controller documentation and build instructions are released under an Open Source Hardware license to increase accessibility, and reproducibility and enable the neuroimaging research community to improve or modify the controller for future experiments.
Videogames offer a wide range of rich and dynamic environments which open exciting avenues for research in cognitive neuroscience [[
Neuroimaging videogames studies have examined how brain networks interact to produce a variety of complex behaviors such as aggression [[
A commercial, fully MRI-compatible handheld videogame controller was recently proposed by Current Designs Inc., Philadelphia, PA, USA, requiring the separate acquisition of an electronics interface and a fiber optics cord extension, resulting in significant costs. An alternative approach to commercial solutions for research has however emerged: Open Source Hardware (OSH), made possible by the increasing accessibility of 3D printing, engineering parts, and modular micro-electronics [[
The aim of this study was to develop and validate a videogame controller that is both MRI- and MEG-compatible and can be reproduced by other neuroimaging research teams. This was achieved by making detailed documentation and design files accessible to allow for 3D printing and assembly. The controller was evaluated by comparing its responsiveness to a reference controller used in a mock scanner, in terms of button press and release latencies in a cued-response experiment. We also checked the impact of the CNeuroMod controller usage on subject motion, using frame displacement metrics derived from fMRI acquisitions. Finally, we acquired phantom MRI scans as well as MEG empty room data to determine if any change in background noise could be attributed to the controller's presence in the scanning environment.
The documentation for building the controller is provided in the Appendix of this article and can also be found at https://controller-doc.readthedocs.io/en/latest/. It contains a Bill of Materials (BOM) of the commercial parts necessary to build the controller, detailed assembly instructions, source code for the microcontroller, and a download section for the 3D models and Gerber files. The 3D model files are available either as.STL files, which can be readily sent to a 3D printer, or as. STEP files for manipulation and modification using Computer-Aided Design (CAD) software. Gerber files describe the different layers of the Printed Circuit Board (PCB) that need to be sent to a manufacturer for production. The buttons layout of the controller is presented in Fig 1.
Graph: Fig 1 Models and pictures of the different controller parts.A Split view of the various 3D-printed parts B 3D models of the assembled switch part. C 3D model of the PCB. D 3D model of the assembled controller. E-H Real pictures of the corresponding elements as assembled at our facility.
The design of handheld videogame controllers has been heavily influenced by the early model of the Super Nintendo Entertainment System (SNES
The controller was designed with the six following specific criteria in mind. Accordingly, the handheld part of the controller must:
- be built using a 3D printer and commercially available parts in order for it to be easily reproducible,
- be entirely void of electronics and ferro-magnetic components to ensure MRI- and MEG-compatibility,
- reliably register all button presses with a temporal precision comparable to that of the reference controller,
- mimic the feeling and comfort of a reference commercial videogame controller, the Super NES controller, a device held with both hands each manipulating a set of 4 buttons using the thumbs,
- allow users to comfortably play while lying down (position as imposed by MRI scanner and possible in some MEG scanners),
- be comfortable to use for individuals with different hand sizes.
The device is made up of two parts linked by a bundle of fiber optics (Fig 2): the first is the actual handheld controller, and the second is a printed circuit board (PCB) that allows communication between the controller and a stimulation computer. The handheld controller and fiber optics bundle are MRI and MEG-compatible, and as such can be used in the MRI room or MEG chamber while the PCB contains unshielded electronics and must remain in the control room. Our pilot controller featured a 10m long fiber optics cable, which was sufficient to connect the controller in the MRI room with the PCB in the control room.
Graph: Fig 2 Schematics of the two main parts of the controller.The PCB is installed in the control room, while the handheld part is MRI-compatible and can be used in the MRI scanning room.
A Teensy 3.5 (Teensy
The datasets used in this study come from the Courtois NeuroMod databank (https://
Four participants (2 female) were scanned while they performed a variety of tasks in the MEG and MRI scanners. All four subjects were right-handed and had normal hearing for their age, ages ranged between 31 and 47 at first data set acquisitions in 2018 (i.e. hcptrt dataset), and the timespan between first and last dataset acquisition (i.e. gamepad dataset, see Control datasets section) was approximately 2 years and 7 months. Subjects were recruited as part of the Courtois NeuroMod project (CNeuroMod), a dense fMRI database that is being collected on 6 subjects using both naturalistic and controlled stimuli from a broad range of cognitive domains and the main motivator for building the controller. Although the participants of the gamepad task do not properly reflect the population of videogame players, we expected their behavior to be illustrative of the controllers capability as the task didn't include videogame components per se.
The study was approved by the "Comité d'éthique de la recherche vieillissement neuroimagerie" of the CIUSSS du Centre-Sud-de-l'île-de-Montréal (CIUSSS-CSMTL) under the number CER VN 18-19-22 and all participants provided written consent.
A subjective evaluation of the controller in accordance with ISO-9241 on device comfort was assessed via a questionnaire of 8 items answered on a 5-points Likert scale, adapted from [[
fMRI data used in this paper were collected as part of the CNeuroMod project using a 3T Siemens Prima Fit MR scanner, a 64-channel head coil, and an accelerated simultaneous multi-slice imaging sequence [[
We set out to evaluate the reliability of button responses of our custom controller by comparing it to that of a commercially available, non-MRI-compatible controller. Thus, this experiment consisted of a basic cued-response task implemented in PsychoPy [[
In the MRI setup our MRI-compatible controller was installed in the MRI scanner and connected through a waveguide to the controller interface, located in the control room. The controller interface was in turn connected to the stimulation computer via USB to register and log key presses during the task. The remaining set up was exactly as described above in the section on fMRI acquisition (Fig 3).
Graph: Fig 3 Schematics of the MRI and Mock setups.The mock setup closely replicated the MRI setup while allowing the use of a commercial controller instead of our MRI-compatible controller. Icons obtained under CC license at https://thenounproject.com/.
In the mock setup we used an MRI simulator made by Psychology Software Tools (https://pstnet.com/products/mri-simulator/), and used an in-house 3D printed replica of the Siemens 64 channel head coil used in the MRI. An in-house built mirror mount replicating the Siemen's mirror attached to the head coil was used to present images to the participants. Instead of a projection, visual stimuli were presented using a small monitor placed within the mock bore, closely replicating the manner in which visual stimuli are presented in the MR scanner. A replica of the SNES
Participants were instructed they would have to perform a cued-response task, in which, using visual cues, they would have to press one of the 8 buttons present on either the MRI or SNES controllers, for a short or long period of time. Short presses were indicated by the presence of a yellow dot prior during a trial and long presses were indicated by a green bar during a trial (Fig 4). In short trials, a yellow dot preceding the cue instructed the participants to press and release the cued button as fast as possible (cue displayed for 300ms), while in long trials participants were instructed to hold the cued button as long as the green bar was present on the screen (cue was displayed for a duration between 1 and 3 seconds with a gaussian jitter). Participants performed the tasks in 4 sessions, and during each session performed the task in both the MRI and mock setups (2 runs each). In a run, the subject performed 5 repetitions of each of the 8 buttons in the two conditions (short and long) for a total of 80 trials per run (5*8*2). Each participant performed 8 runs per setup (Mock and MRI) split in 4 sessions, for a total of 1 280 trials (80*8*2).
Graph: Fig 4 Illustration of the unfolding of the simple cued-response task.
To illustrate the values of keypress durations that naturally occur when playing action platformer videogames, and use them as reference to interpret the results of our gamepad task, we computed the durations observed in the training phase of the shinobi dataset. The data for the shinobi_training dataset were obtained using the commercial controller (i.e non-MRI controller that was also part of our Mock scan condition of the gamepad task). This dataset comprised the same 4 subjects that performed the gamepad and shinobi tasks, and totals 1 165 trials across the 3 selected levels of Shinobi III: Return of the Ninja Master [[
In order to estimate the amount of motion induced by our controller both in controlled and naturalistic conditions, we compared the motion (quantified by Framewise Displacement, FD) observed during the fMRI acquisitions of the gamepad and the shinobi datasets respectively with other datasets obtained on the same 4 subjects, chosen as a reference because they did not involve the controller. One exception is the shinobi dataset as sub-03 did not participate, and thus we only included data from 3 of the 4 subjects. The control datasets (friends, hcptrt-gamble, hcptrt-motor and hcptrt-rest) were selected to reflect the variety of motion amounts one can expect during conventional active and passive tasks. We used around 38h of fMRI data from the friends dataset during which the subjects were watching seasons 1 up to 4 of the Friends TV show [[
We also used data from the hcptrt dataset, namely the resting state, gambling and motor tasks. All tasks in the hcptrt dataset were designed by the team at the Human Connectome Project [[
The motor task was repeated between 15 and 18 times per subject. In this task, participants were presented a visual cue and were asked move a body part (ie. tongue, fingers, toes, hand, feet) in response to that cue. We expected these three datasets to reflect relatively low, moderate and high levels of motion for the resting, gambling and motor tasks respectively.
Lastly we used fMRI data from 3 of the participants in the shinobi dataset, which consists of about 10h (approximately 56 runs) of fMRI data per subject. During the shinobi task, subjects played three selected levels of Shinobi III: Return Of The Ninja Master [[
For more information on each of the control datasets, see the Courtois NeuroMod's Project documentation page: (https://docs.cneuromod.ca/en/latest/).
Thirty runs of 60 volumes were acquired using a fBIRN gel phantom with the custom controller in the scanner, and 30 more without the controller, all following the imaging sequence described in the MRI acquisition section. Each run lasted about 2 minutes, and a total of 60 runs were acquired on a single day.
Empty room noise recordings were acquired as part of the routine procedure for MEG acquisitions. The data was collected on separate days with a MEG CTF-275 (with 270 functioning channels) while no participant was inside the MEG shielded room. In each recording, 35 seconds of signal were acquired at a sampling frequency of 1200Hz. No offset removal nor filters were used during these acquisitions. We used the empty room recordings acquired between 02/04/2019 and 09/06/2021 (N = 80), and split them into 4 subsets of 20 consecutive recordings. These subsets, named T1, T2, T3 and T4 include data acquired before the installation of our custom controller (T1, T2 and T3), and data acquired after the installation of our controller at the MEG facility (T4). Each of these subsets contained recordings spanning over between 1 and 3 and a half months.
In order to verify that our controller had the same performance capacity as the commercial controller, we tested the hypothesis of a greater response latency in the MRI (using our controller) than in the mock scanner (using the commercial reference controller). We removed: (a) trials for which no button press of the instructed key was recorded; (b) trials with negative reaction times and durations, which correspond to trials that were anticipated by the subject due to the task's pseudo-random sequence; and (c) trials for which the press timing was recorded but no timing for the release was recorded due to an overlap with the subsequent trial. Additionally, we removed (d) trials from the "long-press" condition for which the press duration was below 1s, as these trials might have been mistakenly performed as "short-press" trials. In total, we removed 71 trials of the mock condition and 50 trials of the MRI condition (Table 1). Statistical analyses were conducted on pooled reaction times, and on press duration for long and short trials separately. One-tailed Kolmogorov-Smirnov tests for independent samples with FDR (Benjamini-Hochberg) correction for multiple comparisons were conducted between the two setups. Effects size were estimated using Cohen's D.
Graph
Table 1 Number of removed trials at each step of the preprocessing pipeline.
A B X Y Up Down Left Right Total Mistakes Mock 3 3 8 4 9 7 6 7 47 MRI 6 3 0 3 6 3 3 2 26 Negative RTs Mock 0 2 1 0 0 0 0 0 3 MRI 0 1 0 0 0 1 0 0 2 Negative durations Mock 0 0 0 0 0 0 0 0 0 MRI 0 0 0 0 1 0 0 0 1 Delayed trials Mock 0 1 2 0 2 1 0 1 7 MRI 0 0 0 0 7 3 0 1 11 Missed trials Mock 3 2 1 2 3 2 0 1 14 MRI 1 3 1 1 3 0 1 0 10 Retained trials Mock 234 232 228 234 226 230 234 231 1849 MRI 213 213 219 216 203 213 216 217 1710 Total removed Mock 6 8 12 6 14 10 6 9 71 MRI 7 7 1 4 17 7 4 3 50
Framewise Displacement (FD) values were obtained from the confounds file generated by the fMRIprep preprocessing pipeline [[
fMRI temporal SNR (with versus without controller). Temporal Signal-to-Noise Ratio (SNR) maps were computed on each run then averaged per condition (data from the tasks was divided into one of two conditions; with or without the controller present in the scanner) using Nipype [[
Noise covariance matrices were computed on each empty room acquisition run using the MNE-python toolbox [[
The answers (N = 4) to each question of the utility ratings survey were averaged and inspected separately. All scales ranged from 1 to 5, although the interpretation of each score varies as summarized below and detailed in the methods section. For smoothness during operation, the average rating was 4.75 (±0.5 std, higher is better). For the force required for actuation and the operation speed, the average rating was 3.25 (±0.5 std, a score of 3 is best). The mental effort required for operation was rated at 3 (±0 std, a score of 3 is best). Both finger and wrist fatigue were rated below 2 (2±1.15 std and 1.25±0.5 std respectively, lower is better). Finally, both general comfort and overall useability were rated above 4 (4.25±0.5 std and 4.75±0.5 std respectively, higher is better).
Analysis of reaction times (Fig 5) showed no differences (all p > 0.05, see Table 2) between our MRI-compatible controller (MRI condition) and the commercial controller (Mock condition) when comparing each key separately. Analysis of release reaction times in long trials revealed significant (p < 0.05) differences for 6 of the 8 keys, showing that the MRI-compatible controller led to longer release reaction times (-0.417 < Cohen's D < -0.040 for all significant comparisons) than the reference controller. Analysis of press durations in short trials also showed significant differences (p < 0.05) for all keys, showing that the duration of short presses was longer using our MRI-compatible controller (-1.332 < Cohen's D < -0.300 for all comparisons).
Graph: Fig 5 Distributions of response latencies (pooled across subjects) in the MRI and Mock conditions.ns: p < = 1.00e+00; *: 1.00e-02 < p < = 5.00e-02; **: 1.00e-03 < p < = 1.00e-02; ***: 1.00e-04 < p < = 1.00e-03; ****: p < = 1.00e-04. A Comparison of reaction times (ms). B Comparison of release reaction times (ms) for long trials only. C Comparison of total key press durations (ms) for short trials only.
Graph
Table 2 Statistics table showing the comparisons between the MRI and Mock conditions.
Mock MRI Difference N Mean STD N Mean STD KS D Mean Press reaction times—All trials (ms) A 234 460 83 213 466 92 0.071 (ns) -0.071 6.251 B 232 444 96 213 436 95 0.022 (ns) 0.084 8.012 Down 230 501 155 213 461 121 0.000 (ns) 0.293 40.675 Left 234 514 170 216 457 108 0.000 (ns) 0.397 56.612 Right 231 488 132 217 455 113 0.026 (ns) 0.266 32.773 Up 226 508 141 203 477 156 0.015 (ns) 0.207 30.625 X 228 473 142 219 452 109 0.020 (ns) 0.164 20.772 Y 234 472 104 216 449 94 0.001 (ns) 0.233 23.132 Release reaction times—Long trials (ms) A 116 329 162 104 338 153 0.100 (ns) -0.058 9.118 B 117 325 101 106 344 182 0.283 -0.129 19.089 Down 113 330 113 106 360 158 0.312 -0.215 29.518 Left 114 307 117 107 357 87 0.449 -0.49 50.538 Right 112 295 152 107 325 127 0.284 -0.22 30.775 Up 109 309 107 97 331 154 0.266 -0.17 22.609 X 110 338 97 110 343 128 0.136 (ns) -0.047 5.318 Y 116 327 93 107 357 98 0.185 -0.314 30.006 Press durations—Short trials (ms) A 118 192 40 109 211 49 0.245 -0.439 19.633 B 115 195 36 107 222 74 0.198 -0.457 26.578 Down 117 189 46 107 213 57 0.293 -0.47 24.418 Left 120 180 41 109 208 61 0.287 -0.536 27.904 Right 119 176 48 110 236 62 0.580 -1.078 59.652 Up 117 186 41 106 231 72 0.321 -0.762 44.517 X 118 201 57 109 209 52 0.149 (ns) -0.152 8.34 Y 118 202 50 109 222 45 0.341 -0.412 19.595
1 One-tailed (MRI > Mock) Kolmogorov-Smirnov tests for independent samples, with FDR (Benjamini-Hocheberg) correction. KS: Kolmogorov-Smirnov statistics. D: Cohen's D. ns: p < = 1.00e+00;
- 2 *: 1.00e-02 < p < = 5.00e-02;
- 3 **: 1.00e-03 < p < = 1.00e-02;
- 4 ***: 1.00e-04 < p < = 1.00e-03;
- 5 ****: p < = 1.00e-04.
For reference, the naturalistic key durations observed during play in the shinobi_training dataset (Table 3) ranged from from an average of 164ms (for phasic actions that aren't impacted by keypress duration, such as hitting an enemy) to 1153ms for tonic actions that can be maintained over time, such as moving to the right.
Graph
Table 3 Descriptive statistics of keypress durations observed in the training phase of the shinobi dataset. These values reflect typical keypress durations when playing an action platformer videogame on a regular controller.
Duration (ms) Count Mean STD Min 25% 50% 75% Max Subject Key sub-01 B (Hit) 27939 134 45 16 116 133 150 1400 Y (Jump) 29645 350 187 16 200 317 467 1484 Down 24710 469 423 16 233 350 567 9367 Left 10683 669 670 16 200 450 883 15134 Right 29014 1320 2123 16 133 567 1600 38650 Up 6478 428 1055 16 33 66 250 16217 sub-02 B (Hit) 49597 235 169 16 117 183 316 3800 Y (Jump) 41653 426 263 16 250 366 533 2800 Down 17697 791 730 16 316 634 1000 8967 Left 11559 616 613 16 200 434 784 13950 Right 48203 1228 1606 16 100 750 1766 56317 Up 15443 490 511 16 150 367 584 7316 sub-03 B (Hit) 2155 180 214 16 133 150 183 4834 Y (Jump) 904 508 384 17 217 384 704 4700 Down 747 824 858 16 250 667 1075 13533 Left 398 575 517 16 233 434 817 4617 Right 1627 1159 1666 16 300 666 1358 22384 Up 328 446 533 16 83 316 621 6317 sub-06 B (Hit) 25961 61 53 16 50 50 67 5516 Y (Jump) 14786 266 260 16 66 150 433 1483 Down 6461 473 423 16 200 416 584 7833 Left 4699 369 370 16 117 250 508 3883 Right 17100 656 897 16 116 334 888 20017 Up 7052 361 427 16 96 250 483 6600 Total B (Hit) 105652 164 144 16 83 133 183 5516 Y (Jump) 86988 374 248 16 200 333 500 4700 Down 49615 589 581 16 234 433 766 13533 Left 27339 593 610 16 183 400 766 15134 Right 95944 1153 1704 16 117 567 1550 56317 Up 29301 445 658 16 67 283 534 16217
Analysis of run-median FD values (Fig 6 and Table 4) showed no significant differences of motion between the gamepad dataset and the other datasets that didn't use our MRI-compatible controller, namely the hcp-rest, hcp-motor and hcp-gamble datasets and the friends dataset (all p > 0.05), with the exception of 2 subjects that showed significant differences between the gamepad and friends datasets (p < 0.01). For all 3 subjects, the shinobi dataset had significantly higher run-median FD values than all the other datasets (all p < 0.05), except for one subject that showed no significant difference between the shinobi dataset and the hcptrt resting-state dataset.
Graph: Fig 6 Distributions of the median of FD values across runs in the datasets included in this study.Data were analyzed separately for each subject, which showed similar effects (A-D). E shows the pooled data of all subjects. ns: p < = 1.00e+00; *: 1.00e-02 < p < = 5.00e-02; **: 1.00e-03 < p < = 1.00e-02; ***: 1.00e-04 < p < = 1.00e-03; ****: p < = 1.00e-04.
Graph
Table 4 Descriptive statistics of the run-median Framewise Displacement (FD) values for the different datasets used in the motion analysis.
sub-01 sub-02 sub-03 sub-06 Total N Mean STD Dur. (hours) N Mean STD Dur. (hours) N Mean STD Dur. (hours) N Mean STD Dur. (hours) N Mean STD Dur. (hours) friends 194 0.098 0.027 37.89 194 0.139 0.043 37.89 196 0.039 0.022 38.308 194 0.143 0.033 37.89 778 0.105 0.031 151.978 gamepad 8 0.063 0.028 0.685 8 0.099 0.022 0.685 8 0.025 0.014 0.685 8 0.13 0.022 0.685 32 0.079 0.022 2.74 hcp-gamble 15 0.055 0.021 0.801 15 0.097 0.05 0.801 15 0.01 0.005 0.801 17 0.103 0.035 0.908 62 0.066 0.028 3.311 hcp-motor 15 0.053 0.019 0.901 15 0.138 0.041 0.894 15 0.027 0.029 0.894 18 0.156 0.051 1.073 63 0.094 0.035 3.762 hcp-rest 5 0.102 0.024 1.242 5 0.141 0.041 1.242 5 0.025 0.009 1.242 6 0.148 0.035 1.49 21 0.104 0.027 5.216 shinobi 57 0.12 0.018 10.572 56 0.158 0.029 10.735 0 NaN NaN 0 56 0.206 0.031 11.343 169 0.161 0.026 32.65
The temporal SNR maps showed similar distributions of means and standard deviations across the runs acquired with and without the presence of our controller in the MRI scanner (Fig 7). Comparing the average of voxels located within the EPI mask, the two distributions showed significant differences (both p < = 1.00e-04; M
Graph: Fig 7 Temporal SNR maps on runs acquired with and without the controller present in the scanner.A and B show the mean SNR across runs recorded with and without the controller respectively. C shows the distributions of the averaged SNR values for voxels within the EPI mask. D, E, and F show the same visualizations but for STD values instead. ****: p < = 1.00e-04.
Analysis of the difference between averaged covariance matrices acquired before (T3) and after (T4) the controller installation (Fig 8) showed no significant difference (Wilcoxon signed-rank test, p = 0.139). In contrast, although the controller wasn't present in the scanner during recordings of both T1 and T2, the comparison between these two subsets led to a significant difference (Wilcoxon signed-rank test, p = 0.004), showing that the impact of environmental fluctuations on noise levels could be detected by this method, while the controller's effect could not. In both cases (T1 vs T2 and T3 vs T4), the Pearson correlation between the two averaged matrices was fairly high (r = 0.901 and r = 0.953, respectively). Finally, the similarity of the distributions of pairwise correlations between individual recordings in the different subsets further suggests that the noise patterns observed after the addition of our custom controller weren't altered more than expected by normal environmental fluctuations.
Graph: Fig 8 Comparison of MEG sensors noise covariance over time.A—D Normalized average covariance across T1 and T2 subsets, and correlation between these two distributions. T1 and T2 both were acquired without the CNeuroMod controller present in the MEG scanner. E—H Normalized average covariance across T3 and T4 subsets, and correlation between these two distributions. T3 was acquired without the controller present in the scanner, while T4 was acquired with the controller. I—L Distributions of the pairwise correlations between each pair of individual recordings, for each pair of subsets.
A reproduction of the CNeuroMod controller was built by the engineering team at Neurospin and members of the MIND team at Inria (Paris-Saclay, France; Fig 9). The MIND team was not involved in the design process. The reproduced controller was built according to the design plans and documentation made available online and was fitted to their local MRI system. Minor modifications were implemented by the MIND team, for example the shutter and the cap were printed together instead of assembled after printing and the fiber optic cable was 12m long instead of 10m. Other minor assembly issues (buttons getting stuck) were reported and troubleshooted directly with AC, although other teams will also be able to obtain support via GitHub. The controller was tested by 9 different players during gameplay on Super Mario Bros [[
Graph: Fig 9 Pictures of the reproduced controller by team members at Inria/NeuroSpin.A shows the assembled PCB. B shows the bottom part of the controller, with the switch mechanism in the process of being assembled. C shows the optics fibers installed in the handheld part. D shows the fully assembled controller.
We built the CNeuroMod videogame controller to be compatible with both fMRI and MEG. We wrote the documentation, plans needed by other teams to reproduce the controller and its software and made it publically available under an Open Source Hardware license. A feasibility study showed that our controller consistently registers button presses and releases with only small differences in timing compared with a commercial controller. We showed that the CNeuroMod controller can be used without inducing abnormal motion levels. The presence of the CNeuroMod controller in the scanning environment had small effects on tSNR maps and noise covariance matrices with fMRI and MEG, respectively.
The design files, software and documentation required to build the controller are all released under a CERN-OHL-P license, an Open Hardware License that allows users to reproduce, modify and redistribute the covered material. This license has been chosen as it is the most permissive version of the CERN-OHL flavors, with the goal of allowing maximal freedom of reuse to anyone. The presented device requires moderate engineering and technical resources to be built, and the materials used were chosen on the basis of their availability and low costs. All the necessary parts can be obtained for an estimated price around 1000 CAD (Table 5). In comparison, the device proposed by Current Design Inc. (including cables and interface) are sold for a total price of 8390 USD at the time of writing. The provided documentation covers all the steps involved in the preparation, assembly and use of the controller. Through the release of this hardware under a permissive OHL, we anticipate that other research teams or commercial entities will be able to easily reuse and/or modify the controller design beyond purely financial considerations.
Graph
Table 5 Bill of material with estimated prices at the time of writing of this manuscript. More details about the suppliers and item numbers can be found in the documentation.
Part Supplier Qty Price (CAD) Fiber optic cable duplex 100m Digikey 1 279.71 Fiber optic transmitter Digikey 8 163.2 Fiber optic receiver Digikey 8 198.72 Teensy 3.5 microcontroller board Digikey 1 37.76 Gate driver Digikey 5 7 Hose 13mm ID, black PE 1 Digikey 1 111.55 Hose 19.48mm ID, black PE 1 Digikey 1 160.38 Voltage regulator Digikey 1 2.93 Resistor 330 Ω Digikey 8 3.67 Resistor 65 Ω Digikey 8 2.39 Capacitor ceramic 0.1 μF Digikey 8 2.49 Capacitor ceramic 10 μF Digikey 2 2.3 Connector header pin breakaway 24 pos Digikey 4 14.28 Connector receptacle 24 pos Digikey 2 9.18 Polycarbonate Pan Head Philips Screw 8–32 x 1" 2 McMaster-Carr 2 2 Polycarbonate Pan Head Philips Screw 8–32 x 1/2" 2 McMaster-Carr 3 3 Polycarbonate Pan Head Philips Screw 4–40 x 3/16" 3 McMaster-Carr 8 3.49 Nylon 6/6 Female Threaded Round Standoff 8–32 x 1/4" 2 McMaster-Carr 5 8.35 Nylon 6/6 Female Threaded Round Standoff 4–40 x 1/8" 3 McMaster-Carr 8 13.12 Total 1025.52
The study showed that the CNeuroMod controller consistently registers button presses as well as releases with a temporal precision comparable to a commercial controller. Although no differences were observed in button press reaction times, small significant differences were detected in release reaction times (for long presses) and in press duration (for short presses), showing longer release times with the prototype than with the reference controller. This result might indicate the presence of mechanical friction during the button release, an aspect that could be improved in further design iterations.
This difference in release reaction time could impact game play performance, particularly in games that require dense repetitions of button presses, but is likely negligible when playing most other games, as suggested by the successful collection of the shinobi dataset as well as the pilot implemented by the MIND team with Super Mario Bros [[
The analyses conducted on fMRI runs collected on a wide range of tasks showed similar amounts of motion during our gamepad task compared to control tasks that did not involve the use of a videogame controller. This suggests that the CNeuroMod controller does not increase the amount of motion in the scanner. Conversely, the comparison between the motion measured during the shinobi dataset to that of other control datasets indicated significantly higher levels of motion for the gaming task (shinobi), likely to be attributable to the videogame content and not to the controller itself. Even in the case of the shinobi dataset, many runs had a fairly low level of motion, with median FD levels below 0.2 or even 0.1. Overall, the usage of the Courtois NeuroMod controller seems compatible with the collection of usable fMRI data, and in some instances even high-quality low-motion data.
Analyses performed on fMRI data acquired on a gel phantom head showed a small impact of the CNeuroMod controller on temporal SNR. These differences are statistically significant but are relatively small compared to the amount of noise typically observed during such recordings. Because fMRI recordings are known to be sensitive to thermal noise [[
The MEG noise covariance matrices showed no significant impact of the CNeuroMod controller. Further analysis of empty room recordings collected at various periods (from April 2019 to June 2021) showed that the variation of noise levels induced by the controller stayed well within the range of changes that could be expected from normal environmental fluctuations.
The CNeuroMod MRI- and MEG-compatible videogame controller provides a versatile device to play most videogames. This versatility extends to any experimental task that requires the use of 8 or less buttons, representing a very large action space. For example, one can use this controller to answer complex questionnaires, e.g. featuring Likert scale, or perform neuropsychological tasks where users navigate many questions. The CNeuroMod controller is specifically geared towards retro gaming from the generation of the Super Nintendo Entertainment System (SNES
Future design iterations could reduce the latency of button releases, and enable the inclusion of additional buttons and analog joysticks that would expand the scope of playable videogames to modern titles.
The CNeuroMod controller is designed to be built using commonly accessible tools and materials, and released under an Open Hardware License for maximal reusability. A feasibility study showed response latencies comparable to a standard commercial controller. MRI and MEG experiments showed that the controller does not increase subject motion and has marginal effects on acquisition noise, if any. We hope that the CNeuroMod controller will assist researchers in leveraging the richness of videogame tasks in cognitive neuroscience. Our open design can also contribute to a growing ecosystem of open hardware for reproducible neuroimaging research.
S1 Appendix. Build instructions and documentation.
(PDF)
S1 Dataset. Cued-response task results.
(CSV)
S2 Dataset. Shinobi training behavioral data.
(GZ)
S3 Dataset. FD values of CNeuroMod datasets.
(CSV)
S4 Dataset. fMRI phantom data.
(CSV)
S5 Dataset. Noise covariances of MEG empty room data.
(GZ)
S6 Dataset. Utility ratings of the controller in both sites.
(ODS)
S1 Graphical abstract.
(TIF)
King Bradley R. Academic Editor
19 Jun 2023
PONE-D-23-14328Open design and validation of a reproducible videogame controller for MRI and MEGPLOS ONE
Dear Dr. Harel,
Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE's publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.
Please submit your revised manuscript by Aug 03 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://
Please include the following items when submitting your revised manuscript:
If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.
If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.
We look forward to receiving your revised manuscript.
Kind regards,
Bradley R. King
Academic Editor
PLOS ONE
Journal Requirements:
When submitting your revision, we need you to address these additional requirements.
1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at
https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne%5fformatting%5fsample%5fmain%5fbody.pdf and
https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf
2. We note that the grant information you provided in the 'Funding Information' and 'Financial Disclosure' sections do not match.
When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the 'Funding Information' section.
3. Thank you for stating the following in the Acknowledgments Section of your manuscript:
"The Courtois project on neural modeling was made possible by a generous donation from the Courtois foundation, administered by the Fondation Institut Gériatrie Montréal at CIUSSS du Centre-Sud-de-l'île-de-Montréal and University of Montreal. The Courtois NeuroMod team is based at "Centre de Recherche de l'Institut Universitaire de Gériatrie de Montréal", with several other institutions involved. See the CNeuromod documentation for an up-to-date list of contributors (https://docs.cneuromod.ca). The replication of the controller by the MIND team, Inria, France has received funding from the European Union's Horizon 2020 Framework Programme for Research and Innovation under the Specific Grant Agreement No. 945539 (Human Brain Project SGA3) and through the joint Inria "NeuroMind" team grant to PB, KJ, BT and Alexandre Gramfort. PB is a senior fellow ("chercheur boursier senior") of the "Fonds de recherche du Québec - Santé". KJ is supported by funding from the Canada Research Chairs program and a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada.
The authors would like to thank the MEG Imaging Center and Courtois-Neuromod scanning crews for acquiring the validation data. "
We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.
Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:
"The Courtois project on neural modeling was made possible by a generous donation from the Courtois foundation, administered by the Fondation Institut Gériatrie Montréal at CIUSSS du Centre-Sud-de-l'île-de-Montréal and University of Montreal. The Courtois NeuroMod team is based at "Centre de Recherche de l'Institut Universitaire de Gériatrie de Montréal", with several other institutions involved. See the CNeuromod documentation for an up-to-date list of contributors (https://docs.cneuromod.ca). The replication of the controller by the MIND team, Inria, France has received funding from the European Union's Horizon 2020 Framework Programme for Research and Innovation (https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-open-calls/horizon-2020%5fen) under the Specific Grant Agreement No. 945539 (Human Brain Project SGA3) and through the joint Inria "NeuroMind" team grant to PB, KJ, BT and Alexandre Gramfort. PB is a senior fellow ("chercheur boursier senior") of the "Fonds de recherche du Québec - Santé". KJ is supported by funding from the Canada Research Chairs (https://
Please include your amended statements within your cover letter; we will change the online submission form on your behalf.
4. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see
"Upon re-submitting your revised manuscript, please upload your study's minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see
Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data:
We will update your Data Availability statement to reflect the information you provide in your cover letter.
- 5. Please include your full ethics statement in the 'Methods' section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.
- 6. We note that Figures 1,2 and 3in your submission contain copyrighted images. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.
We require you to either (
a. You may seek permission from the original copyright holder of Figures 1, 2, and 3 to publish the content specifically under the CC BY 4.0 license.
We recommend that you contact the original copyright holder with the Content Permission Form (
"I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (
Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.
In the figure caption of the copyrighted figure, please include the following text: "Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year]."
b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder's requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.
7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information:
[Note: HTML markup is below. Please do not edit.]
Reviewers' comments:
Reviewer's Responses to Questions
1. Is the manuscript technically sound, and do the data support the conclusions?
The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.
Reviewer #1: Yes
Reviewer #2: Yes
***
2. Has the statistical analysis been performed appropriately and rigorously?
Reviewer #1: Yes
Reviewer #2: Yes
***
3. Have the authors made all data underlying the findings in their manuscript fully available?
The
Reviewer #1: Yes
Reviewer #2: Yes
***
4. Is the manuscript presented in an intelligible fashion and written in standard English?
PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.
Reviewer #1: Yes
Reviewer #2: Yes
***
5. Review Comments to the Author
Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)
Reviewer #1: The manuscript reports results from a project to build and test a prototype handheld videogame controller that is designed for full compatibility with the key technical limitations of brain imaging equipment. The report includes conceptual drawings, programming support and analysis of the prototype's response characteristics collected from a small sample of participants in the targeted neuroimaging environments.
The motivation, execution and testing analysis of the scanner-compatible controller are explained. The authors' interpretation of the success of the project is sound and supported by their data, and a confirming example from another research group is reported as well.
The report will interest some neuroimaging research teams that have the skills and mindset to build their own hardware rather than adopt commercially available products. A few concerns are described below, which could be addressed in a revision to make a stronger version of the manuscript.
1. The motivation to build a scanner-compatible videogame controller in-house and publish a thorough engineering analysis of the project is apparent in the Introduction, but it can be stated in more complete fashion.
Based on the prototype, the in-house device costs approximately CAD$1,000 in parts and an undefined amount of experienced labor to build. A bit of investigation reveals that the industry standard Current Designs scanner-compatible videogame controller costs approximately US$4,600, plus fees. The authors appear to justify their project by offering a less costly solution, but a detailed calculation is not presented to support this justification.
A further concern about the relative cost as a justification to build a controller in-house is that the comparability of the haptics and user satisfaction between the device alternatives is not addressed. The participant sample in the results is small and from a narrow age range. Although a larger cohort from a much broader age range would be desirable, some sort of utility rating (as collected from the MIND Team confirmation study) is really needed to fully support the conclusion that the CNeuroMod controller is a suitable substitute for the commercially proven and available alternative.
2. Release reaction times on long trials were significantly longer with the prototype than the mock condition commercial controller (non-scanner compatible). No explanation is offered in the Results or Discussion.
Why is there an apparent difference in the responsiveness of the prototype, depending upon duration of the button presses? How does this responsiveness relate to the poor rating from the MIND Team confirmation study participants (i.e., "a value of 2 for the following items: smoothness during operation and general comfort.")?
3. The title and text refer to a validation of the original test results, which had showed the comparability of the functional values from the prototype with a non-scanner compatible mock-up device. No true validation study is reported. In other words, there are no results from two or more cohorts performing the same tests as sequential experiments, and there is no analysis of validating results (i.e., an independent samples t-test, for example).
The MIND Team "study" is relevant, but not well described. The measures in the main results do not appear in the MIND Team section (3.4), and the Mind Team user ratings do not appear in the main results. This is not a replication, and referring to the MIND Team information as a replication needs to corrected throughout.
- 4. The manuscript reads clearly enough, although the voice or writing style changes noticeably between sections of the text. Conforming the entire report to one writing style would improve its readability.
- 5. Figure 8 includes panels A through L. The legend for Figure 8 only explains panels A through J.
Panels K and L should either be explained in the legend and text, or eliminated from the figure.
Reviewer #2: Overall the paper was quite straightforward and clearly addresses the core question the team started out with. I have one methodological question with regard to how the number of participants and their demographics was chosen. 4 participants isn't very many obviously and the age range of the participants isn't necessarily a good match for what I'd expect of most future work on gaming/neuroimaging. So it'd be useful for the authors to speak to how they determined the sample size and what types of criteria for participants were considered. I'd note though that because I would call this a "feasibility" illustration, the small sample size & potential non-representativeness of the participants aren't "fatal flaws"; but it would be valuable for readers to know how these factors were chosen.
The other issue that I think would improve the paper is a deeper dive into what I would call the exceptionally dated configuration of the controller. The authors indicate that, "Key features of the SNESTM controller can still be found in modern systems such as the XBoxTM (Microsoft) and PlayStationTM (Sony)" which is true, but still reasonably misleading. For instance, it's a bit like if the team created a new car that was basically a Model T Ford, but said that it "still contains many features of modern cars such as the 2023 Honda Accord." It might be true (e.g., it would have 4 wheels, a steering wheel, a windshield), it would still be super dated. Video game controllers have used a dual stick setup since the late 1990s. There are essentially no modern games that I can think of that someone would be able to play with the controller developed here - certainly none of the games used in the research that the authors reference in the introduction (i.e., most of that work on video games/neuroscience has been after the year 2000). So the authors should spend some time on that particular limitation (or else on why it isn't a limitation).
***
6. PLOS authors have the option to publish the peer review history of their article (https://journals.plos.org/plosone/s/editorial-and-peer-review-process#loc-peer-review-history). If published, this will include your full peer review and any attached files.
If you choose "no", your identity will remain anonymous but your review may still be made public.
Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our https://
Reviewer #1: No
Reviewer #2: No
***
[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]
While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.
29 Jul 2023
Response to reviewers
We thank the editor for providing us the opportunity to revise our work, and the reviewers for the insightful comments which have helped improve our study. A point-by-point response to reviewers can be found below.
Formatting :
https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne%5fformatting%5fsample%5fmain%5fbody.pdf and
https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf
Authors answer : We reformatted the manuscript to follow the PLOS ONE.
2. We note that the grant information you provided in the 'Funding Information' and 'Financial Disclosure' sections do not match.
When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the 'Funding Information' section.
Authors answer : The information provided in the Financial Disclosure section of the submission has been re-verified and is correct.
3. Thank you for stating the following in the Acknowledgments Section of your manuscript:
"The Courtois project on neural modeling was made possible by a generous donation from the Courtois foundation, administered by the Fondation Institut Gériatrie Montréal at CIUSSS du Centre-Sud-de-l'île-de-Montréal and University of Montreal. The Courtois NeuroMod team is based at "Centre de Recherche de l'Institut Universitaire de Gériatrie de Montréal", with several other institutions involved. See the CNeuromod documentation for an up-to-date list of contributors (https://docs.cneuromod.ca). The replication of the controller by the MIND team, Inria, France has received funding from the European Union's Horizon 2020 Framework Programme for Research and Innovation under the Specific Grant Agreement No. 945539 (Human Brain Project SGA3) and through the joint Inria "NeuroMind" team grant to PB, KJ, BT and Alexandre Gramfort. PB is a senior fellow ("chercheur boursier senior") of the "Fonds de recherche du Québec - Santé". KJ is supported by funding from the Canada Research Chairs program and a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada.
The authors would like to thank the MEG Imaging Center and Courtois-Neuromod scanning crews for acquiring the validation data. "
We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.
Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:
"The Courtois project on neural modeling was made possible by a generous donation from the Courtois foundation, administered by the Fondation Institut Gériatrie Montréal at CIUSSS du Centre-Sud-de-l'île-de-Montréal and University of Montreal. The Courtois NeuroMod team is based at "Centre de Recherche de l'Institut Universitaire de Gériatrie de Montréal", with several other institutions involved. See the CNeuromod documentation for an up-to-date list of contributors (https://docs.cneuromod.ca). The replication of the controller by the MIND team, Inria, France has received funding from the European Union's Horizon 2020 Framework Programme for Research and Innovation (https://research-and-innovation.ec.europa.eu/funding/funding-opportunities/funding-programmes-and-open-calls/horizon-2020%5fen) under the Specific Grant Agreement No. 945539 (Human Brain Project SGA3) and through the joint Inria "NeuroMind" team grant to PB, KJ, BT and Alexandre Gramfort. PB is a senior fellow ("chercheur boursier senior") of the "Fonds de recherche du Québec - Santé". KJ is supported by funding from the Canada Research Chairs (https://
Please include your amended statements within your cover letter; we will change the online submission form on your behalf.
Authors answer : We removed the paragraph related to funding information in the "Acknowledgement" section. The information ("Financial Disclosure") provided in the submission form were verified and are correct.
4. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see
"Upon re-submitting your revised manuscript, please upload your study's minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see
Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data:
We will update your Data Availability statement to reflect the information you provide in your cover letter.
Authors answer : We saved the cleaned versions of the datasets used to generate the figures as.csv files (and a.pkl file for MEG covariance matrices) and added them to Supporting Information.
Due to restrictions from the ethics committee in charge of the study, human neuroimaging data is only accessible through a registered access model in a databank hosted at our institution. The instructions to apply for access to the data can be found here: https://
5. Please include your full ethics statement in the 'Methods' section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.
Authors answer : We modified the manuscript to add a section named "2.2.2 Ethics statement", containing the information provided in the submission form.
6. We note that Figures 1,2 and 3in your submission contain copyrighted images. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For more information, see our copyright guidelines:
We require you to either (
a. You may seek permission from the original copyright holder of Figures 1, 2, and 3 to publish the content specifically under the CC BY 4.0 license.
We recommend that you contact the original copyright holder with the Content Permission Form (
"I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (
Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.
In the figure caption of the copyrighted figure, please include the following text: "Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year]."
b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder's requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.
Authors answer : The figure 1 contains only images we created ourselves and for which we own the full rights (both pictures and 3d models). The figures 2 and 3 have been re-created from scratch using only original images for which we own the full rights.
7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.imagesorg/plosone/s/supporting-information.
Authors answer : We added a section named Supporting Information that mentions the documentation and dataset files.
Content :
1a. The motivation to build a scanner-compatible videogame controller in-house and publish a thorough engineering analysis of the project is apparent in the Introduction, but it can be stated in more complete fashion.
Based on the prototype, the in-house device costs approximately CAD$1,000 in parts and an undefined amount of experienced labor to build. A bit of investigation reveals that the industry standard Current Designs scanner-compatible videogame controller costs approximately US$4,600, plus fees. The authors appear to justify their project by offering a less costly solution, but a detailed calculation is not presented to support this justification.
Authors answer : We thank the reviewer for pointing out the need for a more thorough price comparison. We modified the paragraph named "Open design and reusability" to add a table (Table 5) that breaks down the prices of the bill of materials. The Discussion has been modified to also discuss the price of a commercial vs open solution.
1b. A further concern about the relative cost as a justification to build a controller in-house is that the comparability of the haptics and user satisfaction between the device alternatives is not addressed. The participant sample in the results is small and from a narrow age range. Although a larger cohort from a much broader age range would be desirable, some sort of utility rating (as collected from the MIND Team confirmation study) is really needed to fully support the conclusion that the CNeuroMod controller is a suitable substitute for the commercially proven and available alternative.
Authors answer : The utility ratings of the CNeuroMod sample (i.e. the subjects who performed the cued-response task) were added to the paper as well as the answers of the Inria/Neurospin participants. These additional data are now presented in the Utility ratings sections of the Methods and Results. The Discussion has been modified to relate utility ratings to other behavioral results, and the ratings dataset has been added to Supporting Information.
2. Release reaction times on long trials were significantly longer with the prototype than the mock condition commercial controller (non-scanner compatible). No explanation is offered in the Results or Discussion.
Why is there an apparent difference in the responsiveness of the prototype, depending upon duration of the button presses? How does this responsiveness relate to the poor rating from the MIND Team confirmation study participants (i.e., "a value of 2 for the following items: smoothness during operation and general comfort.")?
Authors answer : We discuss this effect under the section named "Reactivity of the CNeuroMod videogame controller". We agree that the wording did not make clear that we were addressing specifically the longer release times of our prototype, so we modified the text to make it more explicit. We also added a sentence stating that this effect would likely contribute to the poor utility ratings.
3. The title and text refer to a validation of the original test results, which had showed the comparability of the functional values from the prototype with a non-scanner compatible mock-up device. No true validation study is reported. In other words, there are no results from two or more cohorts performing the same tests as sequential experiments, and there is no analysis of validating results (i.e., an independent samples t-test, for example).
The MIND Team "study" is relevant, but not well described. The measures in the main results do not appear in the MIND Team section (3.4), and the Mind Team user ratings do not appear in the main results. This is not a replication, and referring to the MIND Team information as a replication needs to corrected throughout.
Authors answer : We agree with the reviewer that the terms "replication" and "validation" were used improperly. We removed the term validation from the title. We also replaced references to this study as "validation" for other terms, such as "feasibility study". We replaced references to "replication" by "reproduction" of the controller device, because the study wasn't replicated by the MIND team, but they produced a second controller unit.
4. The manuscript reads clearly enough, although the voice or writing style changes noticeably between sections of the text. Conforming the entire report to one writing style would improve its readability.
Authors answer : The style and voice of the manuscript have been reviewed and corrected by one of the authors who is a native English speaker.
5. Figure 8 includes panels A through L. The legend for Figure 8 only explains panels A through J.
Panels K and L should either be explained in the legend and text, or eliminated from the figure.
Authors answer : The legend was corrected to cover panels A through L.
6. Overall the paper was quite straightforward and clearly addresses the core question the team started out with. I have one methodological question with regard to how the number of participants and their demographics was chosen. 4 participants isn't very many obviously and the age range of the participants isn't necessarily a good match for what I'd expect of most future work on gaming/neuroimaging. So it'd be useful for the authors to speak to how they determined the sample size and what types of criteria for participants were considered. I'd note though that because I would call this a "feasibility" illustration, the small sample size & potential non-representativeness of the participants aren't "fatal flaws"; but it would be valuable for readers to know how these factors were chosen.
Authors answer : We modified the "Participants" section to explain that the subjects were recruited as part of a larger, longitudinal dataset acquisition for which the controller has been developed. We also included a sentence to state that we expected the task to reflect the controllers capability irrespective of videogame experience.
7. The other issue that I think would improve the paper is a deeper dive into what I would call the exceptionally dated configuration of the controller. The authors indicate that, "Key features of the SNESTM controller can still be found in modern systems such as the XBoxTM (Microsoft) and PlayStationTM (Sony)" which is true, but still reasonably misleading. For instance, it's a bit like if the team created a new car that was basically a Model T Ford, but said that it "still contains many features of modern cars such as the 2023 Honda Accord." It might be true (e.g., it would have 4 wheels, a steering wheel, a windshield), it would still be super dated. Video game controllers have used a dual stick setup since the late 1990s. There are essentially no modern games that I can think of that someone would be able to play with the controller developed here - certainly none of the games used in the research that the authors reference in the introduction (i.e., most of that work on video games/neuroscience has been after the year 2000). So the authors should spend some time on that particular limitation (or else on why it isn't a limitation).
Authors answer : We agree that our description of the controllers main features could be misleading. To clarify, we adjusted paragraph 2.1.1. to remove the reference to modern titles and state that the goal behind the design of this controller was to enable retro videogame play, and not necessarily modern titles. We additionally modified the discussion section called "Implications and future work" to explicitly state that future improvements of the design are going to be required to play modern titles.
Attachment.
Submitted filename: response_letter.docx
King Bradley R. Academic Editor
3 Aug 2023
Open design of a reproducible videogame controller for MRI and MEG
PONE-D-23-14328R1
Dear Dr. Harel,
We're pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.
Within one week, you'll receive an e-mail detailing the required amendments. When these have been addressed, you'll receive a formal acceptance letter and your manuscript will be scheduled for publication.
An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at
If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they'll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.
Kind regards,
Bradley R. King
Academic Editor
PLOS ONE
Additional Editor Comments (optional):
Reviewers' comments:
Reviewer's Responses to Questions
1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the "Comments to the Author" section, enter your conflict of interest statement in the "Confidential to Editor" section, and submit your "Accept" recommendation.
Reviewer #1: All comments have been addressed
Reviewer #2: All comments have been addressed
***
2. Is the manuscript technically sound, and do the data support the conclusions?
The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.
Reviewer #1: Yes
Reviewer #2: (No Response)
***
3. Has the statistical analysis been performed appropriately and rigorously?
Reviewer #1: Yes
Reviewer #2: (No Response)
***
4. Have the authors made all data underlying the findings in their manuscript fully available?
The
Reviewer #1: Yes
Reviewer #2: (No Response)
***
5. Is the manuscript presented in an intelligible fashion and written in standard English?
PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.
Reviewer #1: Yes
Reviewer #2: (No Response)
***
6. Review Comments to the Author
Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)
Reviewer #1: (No Response)
Reviewer #2: (No Response)
***
7. PLOS authors have the option to publish the peer review history of their article (https://journals.plos.org/plosone/s/editorial-and-peer-review-process#loc-peer-review-history). If published, this will include your full peer review and any attached files.
If you choose "no", your identity will remain anonymous but your review may still be made public.
Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our https://
Reviewer #1: Yes: Peter E Wais
Reviewer #2: No
***
King Bradley R. Academic Editor
15 Aug 2023
PONE-D-23-14328R1
Open design of a reproducible videogame controller for MRI and MEG
Dear Dr. Harel:
I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.
If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.
If we can help with anything else, please email us at plosone@plos.org.
Thank you for submitting your work to PLOS ONE and supporting open access.
Kind regards,
PLOS ONE Editorial Office Staff
on behalf of
Dr. Bradley R. King
Academic Editor
PLOS ONE
The authors would like to thank the MEG Imaging Center, the Functional Neuroimaging Unit at CRIUGM and the Courtois NeuroMod scanning crews for acquiring the data.
By Yann Harel; André Cyr; Julie Boyle; Basile Pinsard; Jeremy Bernard; Marie-France Fourcade; Himanshu Aggarwal; Ana Fernanda Ponce; Bertrand Thirion; Karim Jerbi and Pierre Bellec
Reported by Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author
