MindModeling@Home: Difference between revisions
first light |
first light |
||
| Line 7: | Line 7: | ||
| description = MindModeling@Home is an inactive volunteer computing project that used computational cognitive process modeling to study the mechanisms and processes of human performance and learning, hosted by Wright State University and the University of Dayton. | | description = MindModeling@Home is an inactive volunteer computing project that used computational cognitive process modeling to study the mechanisms and processes of human performance and learning, hosted by Wright State University and the University of Dayton. | ||
| status = | | status = Completed | ||
| category = Cognitive Science | | category = Cognitive Science | ||
| compute = CPU | | compute = CPU | ||
| Line 25: | Line 25: | ||
| size = | | size = | ||
| stats as of = | | stats as of = {{Start date and age|2016|08|01}} | ||
| average performance = | | average performance = 5555.82 GigaFLOPS | ||
| active users = | | active users = 2729 | ||
| total users = | | total users = 18702 | ||
| active hosts = | | active hosts = 4334 | ||
| total hosts = | | total hosts = 44271 | ||
| rac = | | rac = | ||
| Line 41: | Line 41: | ||
}} | }} | ||
'''MindModeling@Home''' is an inactive, non-profit [[wikipedia:Volunteer computing|volunteer computing]] research project for the advancement of [[wikipedia:Cognitive science|cognitive science]].<ref name="wikipedia">{{Cite web |title=MindModeling@Home |url=https://en.wikipedia.org/wiki/MindModeling@Home |website=Wikipedia |access-date=2026-06-28}}</ref> The project was hosted by [[wikipedia:Wright State University|Wright State University]] and the [[wikipedia:University of Dayton|University of Dayton]] in Dayton, Ohio, and used idle processing time donated by volunteers to run computational cognitive process models intended to improve scientific understanding of the mechanisms and processes that enable and moderate human performance and learning.<ref name="wikipedia" /><ref name="official">{{Cite web |title=MindModeling@Home |url=https://mindmodeling.org/ |website=mindmodeling.org |access-date=2026-06-28}}</ref> | '''[https://web.archive.org/web/20200421212107/https://mindmodeling.org/ MindModeling@Home]''' is an inactive, non-profit [[wikipedia:Volunteer computing|volunteer computing]] research project for the advancement of [[wikipedia:Cognitive science|cognitive science]].<ref name="wikipedia">{{Cite web |title=MindModeling@Home |url=https://en.wikipedia.org/wiki/MindModeling@Home |website=Wikipedia |access-date=2026-06-28}}</ref> The project was hosted by [[wikipedia:Wright State University|Wright State University]] and the [[wikipedia:University of Dayton|University of Dayton]] in Dayton, Ohio, and used idle processing time donated by volunteers to run computational cognitive process models intended to improve scientific understanding of the mechanisms and processes that enable and moderate human performance and learning.<ref name="wikipedia" /><ref name="official">{{Cite web |title=MindModeling@Home |url=https://mindmodeling.org/ |website=mindmodeling.org |access-date=2026-06-28}}</ref> | ||
The project ran on the [[wikipedia:BOINC|BOINC]] platform and was listed in BOINC's Cognitive Science category.<ref name="boincwiki">{{Cite web |title=MindModeling@Home |url=https://boinc.berkeley.edu/wiki/MindModeling@Home |website=boinc.berkeley.edu |access-date=2026-06-28}}</ref> It required a 64-bit operating system, was best suited to multi-core computers, and supported Microsoft Windows, Mac OS X, and Linux; unlike many other BOINC projects, it was not compatible with mobile devices.<ref name="wikipedia" /> | The project ran on the [[wikipedia:BOINC|BOINC]] platform and was listed in BOINC's Cognitive Science category.<ref name="boincwiki">{{Cite web |title=MindModeling@Home |url=https://boinc.berkeley.edu/wiki/MindModeling@Home |website=boinc.berkeley.edu |access-date=2026-06-28}}</ref> It required a 64-bit operating system, was best suited to multi-core computers, and supported Microsoft Windows, Mac OS X, and Linux; unlike many other BOINC projects, it was not compatible with mobile devices.<ref name="wikipedia" /> | ||
| Line 65: | Line 65: | ||
* '''Sequential and simultaneous task performance''', studying how the brain carries out tasks in sequence versus in parallel by measuring blood flow. | * '''Sequential and simultaneous task performance''', studying how the brain carries out tasks in sequence versus in parallel by measuring blood flow. | ||
Much of this work made use of computationally intensive [[wikipedia:Cognitive architecture|cognitive architectures]] and Monte Carlo-style search over model parameter spaces. For a cognitive model with a parameter vector <math>\theta \in \mathbb{R}^n</math> and an objective | Much of this work made use of computationally intensive [[wikipedia:Cognitive architecture|cognitive architectures]] and Monte Carlo-style search over model parameter spaces. For a cognitive model with a parameter vector <math>\theta \in \mathbb{R}^n</math> and an objective functi<math>f(\theta)</math>on measuring the discrepancy between simulated and observed human performance, the project's volunteer computing infrastructure was used to evaluate <math>f(\theta)</math> at very large numbers of candidate parameter settings in parallel across donated hosts, an approach described in the project's published methodology as simultaneous performance exploration and optimized search.<ref name="hpdc2010">{{Cite journal |last=Moore |first=L. Richard |last2=Kopala |first2=Matthew |last3=Mielke |first3=Thomas |last4=Krusmark |first4=Michael |last5=Gluck |first5=Kevin A. |title=Simultaneous performance exploration and optimized search with volunteer computing |url=http://portal.acm.org/citation.cfm?doid=1851476.1851518 |journal=Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing |year=2010 |doi=10.1145/1851476.1851518}}</ref> | ||
== Technical operation == | == Technical operation == | ||
Latest revision as of 18:08, 28 June 2026
MindModeling@Home is an inactive, non-profit volunteer computing research project for the advancement of cognitive science.[1] The project was hosted by Wright State University and the University of Dayton in Dayton, Ohio, and used idle processing time donated by volunteers to run computational cognitive process models intended to improve scientific understanding of the mechanisms and processes that enable and moderate human performance and learning.[1][2]
The project ran on the BOINC platform and was listed in BOINC's Cognitive Science category.[3] It required a 64-bit operating system, was best suited to multi-core computers, and supported Microsoft Windows, Mac OS X, and Linux; unlike many other BOINC projects, it was not compatible with mobile devices.[1]
History
MindModeling@Home was released to the public on March 17, 2007.[1] The project began life under the name MindModeling@Home (Beta) and was based at the Cognitive Engineering Research Institute (CERI) in Mesa, Arizona, an independent, not-for-profit research institute formed by partners from private industry, academia, and government laboratories with the goal of applying human-centered design and problem-oriented research to socio-technical systems.[4] Project leadership and development were later carried out by researchers associated with the Air Force Research Laboratory (AFRL) at Wright-Patterson Air Force Base, together with collaborators at Wright State University and the University of Dayton Research Institute.[2][5]
In its early years, MindModeling@Home also drew on institutional computing resources beyond ordinary volunteer hosts, incorporating two United States Air Force Department of Defense LSF clusters and an additional Beowulf cluster into the project alongside donations from individual volunteers.[6]
Over time, project operations were handed off between successive staff members. During the project's later years, several long-serving staff departed and operational responsibilities passed to a smaller remaining team, who continued routine maintenance of the servers and forums even as the submission of new scientific workloads became intermittent, since work availability depended on the funding and research phase of the sponsoring AFRL studies rather than on the volunteer infrastructure itself.[2] The project's official forum announcements describe its status during this period as "not down or closed," with project servers continuing to run even when new work units were not being issued for extended stretches.[1]
As of 2026, MindModeling@Home is listed as inactive,[1] with both BOINCstats and Wikipedia recording zero active users, hosts, and average performance for the project.[1]
Research areas
Publicly described research areas explored using MindModeling@Home's volunteer computing resources included:[1]
- N-2 repetition, studying why people have more difficulty returning to a task after performing two intervening tasks rather than one.
- Eye movement during reading, observing how people read with the goal of reducing eye strain and improving reading speed and comprehension.
- Visual decision-making, modeling decisions arising from visual processing, including focus and filtering of visual information.
- Integrated Learning Models (ILM), developing algorithms based on how people learn and make decisions.
- Sequential and simultaneous task performance, studying how the brain carries out tasks in sequence versus in parallel by measuring blood flow.
Much of this work made use of computationally intensive cognitive architectures and Monte Carlo-style search over model parameter spaces. For a cognitive model with a parameter vector and an objective function measuring the discrepancy between simulated and observed human performance, the project's volunteer computing infrastructure was used to evaluate at very large numbers of candidate parameter settings in parallel across donated hosts, an approach described in the project's published methodology as simultaneous performance exploration and optimized search.[7]
Technical operation
Like other BOINC-based projects, MindModeling@Home distributed independent units of work to volunteers' computers, which processed them using idle CPU time and returned results to the project's servers.[3] Because the cognitive models being explored were computationally intensive, runs could be lengthy, and the project's own usage guidance noted that prolonged periods of computation could cause some host computers to run hot; volunteers experiencing this were advised to pause work on the project until their computer cooled down.[1] The project was also affected by ordinary infrastructure issues common to BOINC projects, including at least one recorded power outage affecting the servers in October 2018.[1] The project's public-facing site remained in beta for the bulk of its operational life, with project staff stating in forum discussions that there was no fixed date for when it would formally exit beta status.[1]
Scientific publications
MindModeling@Home's computing resources contributed, directly or indirectly, to a number of peer-reviewed papers and conference proceedings in cognitive science and distributed computing. The BOINC project maintains a consolidated list of Publications by BOINC Projects, which includes a section for MindModeling@Home.[8]
Godwin, Hayward J..(2015).Faster than the speed of rejection: Object identification processes during visual search for multiple targets. Journal of Experimental Psychology: Human Perception and Performance. pp. 1007-1020. DOI: 10.1037/xhp0000036.
Moore, L. Richard.(2014).An interpolation approach for fitting computationally intensive models. Cognitive Systems Research. pp. 53-65. DOI: 10.1016/j.cogsys.2013.09.001.
Moore, L. Richard.(2011).Cognitive model exploration and optimization: a new challenge for computational science. Computational and Mathematical Organization Theory. pp. 296-313. DOI: 10.1007/s10588-011-9092-8.
Moore, L. Richard.(2010).Simultaneous performance exploration and optimized search with volunteer computing. Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing. pp. 312-315. DOI: 10.1145/1851476.1851518.
Harris, Jack.MindModeling@Home. . . and Anywhere Else You Have Idle Processors. Defense Technical Information Center.
Gluck, Kevin.Combinatorics meets processing power: Large-scale computational resources for BRIMS. 16th Conference on Behavior Representation in Modeling and Simulation, BRIMS.
See also
References
- ↑ 1.00 1.01 1.02 1.03 1.04 1.05 1.06 1.07 1.08 1.09 1.10 MindModeling@Home. Wikipedia. Retrieved 2026-06-28.
- ↑ 2.0 2.1 2.2 MindModeling@Home. mindmodeling.org. Retrieved 2026-06-28.
- ↑ 3.0 3.1 MindModeling@Home. boinc.berkeley.edu. Retrieved 2026-06-28.
- ↑ Forum::Chatter::Team BOINCstats: MindModeling@Home (Beta). BOINCstats/BAM!. Retrieved 2026-06-28.
- ↑ (2009).Mindmodeling@Home. . . and Anywhere Else You Have Idle Processors. Defense Technical Information Center. Retrieved 2026-06-28.
- ↑ Combining BOINC with Grids and clusters. boinc.berkeley.edu. Retrieved 2026-06-28.
- ↑ Moore, L. Richard.(2010).Simultaneous performance exploration and optimized search with volunteer computing. Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing. DOI: 10.1145/1851476.1851518.
- ↑ Publications by BOINC Projects. boinc.berkeley.edu. Retrieved 2026-06-28.