NCSU Human Factors and Ergonomics (HFE) Area
Home Intro NIOSH TPG Area Tour People Publications Courses Equipment Links Contact Us



- Michael Clamann

Research Overview

Michael is the lead for the project, “Haptic Simulation Design for Motor Rehabilitation and Fine Motor Skill Training.”

        The purpose of this work is to design and investigate novel features and strategies that incorporate virtual reality (VR)-based haptic simulations to retrain impaired motor functions and train new fine motor skills in veterans with traumatic brain injury (TBI). Consideration of critical applications, like motor skill training, in haptic simulation design is expected to promote effective design from a human performance perspective.

        For this study, we will design and prototype advanced VR haptic simulators requiring use of fine motor skills, such as drawing and simulated surgical tasks. Through human factors experimentation with the simulations, we will also assess interventional strategies for motor development, including virtual haptic aids (e.g., force boundaries and potentials). Finally, we propose to validate the effect of the haptic simulations on neurocognitive and motor performance using behavioral indices and functional magnetic resonance imaging (fMRI).

        We will conduct a series of motor training sessions using the haptic simulators. We will expose subjects to the various settings of the simulation design parameters along with the virtual haptic aids and gradual reduction of force feedback, relative to nominal forces, across sessions. Post-therapy motor and simulator tests, as well as follow-up fMRI scanning will be conducted to provide evidence of the effectiveness of the simulation design and rehabilitation efficacy.
We hypothesize that experience with a haptic simulation design based on motor skill training demands and human performance metrics will accelerate skill development relative to a fidelity-centered approach to design. We also expect that haptic-simulator experience will improve fine-motor control and motor planning, and brain blood flow will increase in regions mediating motor control.

        If our hypotheses are supported, the broader impacts will include: identification of VR-based rehabilitation treatment regimens that may have real-world implications for various populations suffering from brain injuries (e.g., stroke patients); and identification of improvements in existing rehabilitation strategies for motor and praxis impairment in individuals suffering from TBI.


January 2010

Michael is working on a workstation for the project to investigate the effects of haptic interfaces on patients with traumatic brain injuries.

Our team is currently conducting a Cognitive Tasks Analysis to identify the aspects of occupational therapy techniques that can be applied to a VR interface with haptic feedback and developing the hardware and software that will be used by the participants.

(Back to top)

February 2010

In February, the team met with subject matter experts in occupational therapy and diagnosis to identify specific task characteristics that will help refine the VR simulations currently under development.

Meeting with subject matter experts supplements our current effort to reproduce in VR real-world  tasks  used in therapeutic sessions.

We also began constructing two custom workstations that will be used to present the VR tasks during data collection.

(Back to top)

March 2010

In March, Michaels group aggregated what they had learned so far and began producing paper prototypes of the VR software. They also developed a draft experimental design, which they submitted to the IRB for review. This latter task also required them to identify real-world tests of psychomotor ability that they will be integrating with the upcoming study.

Work on the custom workstations continued and the first visual tests of the new VR platforms they built at the end of last year were a success.

(Back to top)

April 2010

In April we continued refining the VR task software prototypes and began work on 3D object rendering and animation. Progress at this point has allowed us to create interactive prototypes in the VR environment. We also produced additional design concepts for both the VR and the real-world tasks.

Construction of a workstation that can be used for VR drawing tasks was completed.

(Back to top)

August 2010

Over the summer, the team continued development of the VR task. After reviewing numerous prototypes internally, we decided to replicate a real-world task using the VR. We chose to replicate the Block Design task from the Wechsler Adult Intelligence Scale, third edition (WAIS-III). It was chosen for its ability to measure cognitive and motor skills and for its availability as a physical task to be presented to our control groups. VR development following this selection has focused on replicating the Block Design task in VR.

We also updated the study plan to include data collection using unimpaired participants. In this second year, we will run this group through our experimental procedures to identify performance differences as a result of VR “therapy” in the Block Design task.

These updates, as well as the team’s other accomplishments made in the prior year, were delivered in an annual progress report to the NSF along with a summary of the ongoing expectations for the project.

Finally, we purchased an additional haptic device, a Phantom Omni (pictured below), to be used as part of a second test platform.

(Back to top)

September 2010

In September, the team gained experience running participants through the Block Design task. The team was trained on the Block Design protocol and ran 8 pilot participants to assess learning speed for the task. The results were used to determine the number of times each participant will need to perform the Block Design task before reaching asymptotic performance.

The year 2 study plan was updated to include the results of the pilot study and the incorporation of pre and post testing using the fMRI at the Duke University Medical Center. This will help us identify physiological differences that result from repeated therapy sessions with the Block Design task.

The updated design will include pre-testing using the fMRI and two tests of psychomotor skills. After the pre-test, participants will return to the lab for multiple “therapy” sessions where they will perform the Block Design task. This will be either the physical (native) task, the basic VR task, or an augmented VR task that includes some form of haptic aiding. The exact form of haptic aiding is still under review.

(Back to top)

October 2010

This month we completed the beta version of the Block Design VR and began beta testing. Six beta testers performed multiple rounds of testing to identify modifications that needed to be made before a final release and a formal pilot test of the VR.

(Back to top)

November 2010

The team completed a literature review to identify the form of haptic aiding that will be used in the augmented VR condition. The team identified several possible forms of haptic aiding and decided to implement two compatible forms of aiding in the augmented version of the VR. These are currently under development.

One form of haptic aiding will be assistance. This enhancement will help guide or “pull” the cursor toward the intended target. This aiding will begin at a long distance from the intended target and will be a source of positive feedback for the user. The other form of haptic aiding will be interference, or negative feedback. This enhancement will reject or “push” the cursor back at a short distance when the user tries to perform an incorrect operation. The two forms of haptic aiding will be implemented at the same time.

(Back to top)

December 2010

In December, the haptic team completed additional pilot tests for the Matrix Reasoning task, the second test that will be used during data collection to measure differences in psychomotor performance as a result of the virtual reality therapy conditions. The three tests will include the VR Rey Osterreith Complex Figure task (VR-ROCF), Matrix Reasoning and WAIS-III Block Design tasks. A pilot test for the VR-ROCF will take place in January.

The team also identified lab space to be used during data collection and set up the experimental apparatus in preparation for a pilot test the first week in January. Two areas have been designated, one for testing and one for therapy session. Having two separate spaces will allow the team to meet with different participants simultaneously. A second therapy workstation is also planned to be used for data collection. Four team members will be trained to perform testing and therapy sessions during data collection.

The team purchased a high-definition television for presenting virtual reality task. The larger display will allow the VR to be presented using more immersive, flexible display configurations, such as real-world (i.e., native) dimensions, during therapy sessions.

(Back to top)

January 2011

In January the haptic team completed its first pilot test integrating the three psychomotor tests (VR Rey-Osterreith Complex figure, Matrix Reasoning and Block Design) and the Basic VR Block Design to validate the experimenter instructions and provide additional training to all team members using all parts of the equipment and procedures. The results of the pilot informed additional changes, which were made to the lab space, experimenter instructions and the information provided to the experiment participant. Staff was also selected for executing test and therapy portions of the data collection.

The experimental apparatus is being expanded to include one testing workstation and two VR workstations to add flexibility to data collection. A more detailed pilot featuring the updated workstations and the augmented haptic condition is expected to follow.

(Back to top)

February 2011

In February, a software release was finalized to allow for scheduling of a preliminary study. The study includes 18 unimpaired right-handed student participants who will complete pre and post-test trials featuring the VR version of the complex figure task, the Matrix Reasoning test and the Block Design test. In-between pre and post testing the participants will complete three days of therapy sessions. Participants will perform either the native Block Design task, Basic VR Block Design or Augmented VR Block Design. The results of this preliminary test will be used to inform the next phase of the study, which will incorporate testing using the fMRI at Duke University Medical Center.

Data collection for the preliminary study is expected to conclude in March.

Michael is also helping locate equipment for the EMG system.

(Back to top)

March 2011

In March, the Haptic team collected data from 21 participants, exceeding the 6 participants planned for each condition (Native Task, Basic VR and Augmented VR). Three additional participants are scheduled for April, after which data collection for this first part of the study will be complete.

A preliminary data analysis suggests that similar learning patterns (i.e., curves) are occurring for all three conditions, although the performance levels for the Native Block Design exceed those for Basic and Augmented VR. This was expected as the VR tasks have the added challenge of interacting with the interface using a stylus, which adds to the required learning.

Additional detailed data analysis to include pre and post test performance differences as a result of the training conditions are planned immediately following data collection. In addition, drafting of the research portion of a year-end report and journal manuscript are planned for April.

(Back to top)

April 2011

Data collection for the reduced experiment (i.e., with unimpaired subjects) concluded in April. Preliminary data analysis reflected significant differences in block design scores among the three conditions, as expected. This was not surprising as the Native, Basic VR and Augmented VR training tasks had notable differences in their administration, in particular when comparing the Native and VR interfaces. Learning rates were also significantly different among the conditions, contrary to the preliminary analysis.

Analysis of the test scores was not quite as conclusive. Block Design and Matrix Reasoning testing did not reveal significant differences among the three conditions. This suggests that the nature of the training conditions may not have had an effect on Block Design or Matrix Reasoning performance. Analysis of the Complex Figure results is still under review, as the analysis requires additional detail. After a preliminary analysis, the team decided to analyze the complex figure in terms of its 18 separate components (i.e., individual shapes) to identify the type of drawing that might be influenced by the VR and Haptic training conditions. This analysis is expected to continue in May.

Composition of the year-end report of the experiment and results continued in April with Michael coordinating the writing efforts. Writing is expected to continue in May.

(Back to top)

August 2011

Data analysis of the Year 1 experiment concluded in May. The team found evidence of improved test performance following VR training with augmented haptic features with insignificant findings for the native task and VR with basic haptic features. The results provide support for using high-fidelity VR-based haptic simulations of standardized psychomotor tests for motor skill training, including visual and haptic enhancements for effective pattern recognition and discrete movement of objects. The results also provide insight for design of future haptic VR features. The detailed results were delivered in an annual progress report to the NSF along with a summary of the ongoing accomplishments resulting from the project. Working with postdoctoral research engineer, Dr. Guk-Ho Gil, Michael also helped produce a manuscript summarizing the study for submission to the journal, Assistive Technology.

The results of the Year 1 study and observations made by the research team during data collection informed numerous changes that are expected to improve the design of the haptic technologies. These changes include removing visual aiding that may not have been helpful to the participants, adjusted gain controls for manipulating the blocks with the haptic device to decrease performance times and reduce fatigue, and redundant visual assistance to recommend block placement options. Over the summer the team continued to work designing, developing, implementing and testing the new features. The Year 2 study data collection is scheduled to begin in the fall.

(Back to top)

September 2011

In September, the Haptic Team continued to prepare for the upcoming Year 2 experiment. In the current study we will look at differences in training effectiveness among three VR conditions: haptic assistance, visual assistance and a combination of haptic and visual. Forms of haptic assistance include attractive forces that "pull" correctly oriented blocks toward the work surface and repulsive forces that "push" incorrectly placed blocks away from the work surface. Visual assistance includes placing a grid overlay on the stimulus design when a block is placed incorrectly and highlights that recommend correct block placement. Like last year, we will meet with unimpaired participants from the student population. Data collection is scheduled to begin in October.

(Back to top)

October 2011

Michael is lead author on the paper, "Comparison Of Infant Car Seat Grip Orientations and Lift Strategies," which was accepted for publication in October. Working with fellow NIOSH trainee Kinley Taylor and coauthors Biwen Zhu, Leah Beaver and Dr. David Kaber, Michael tested the effectiveness of two types of infant car seat handles to determine if a handle designed to reduce wrist deviations could increase grip stability and decrease lifting effort. The results of the experiment showed that an angled handle (designed by the research team) enabled the user to lift the seat with a steadier grip and less effort. The research was conducted as part of a research practicum to fulfill NIOSH-TPG curriculum requirements. The paper will appear in an upcoming issue of Applied Ergonomics.

(Back to top)

November 2011

In November, Michael continued working with the Haptic team to collect performance data from 24 participants. Data collection is expected to conclude in December.

Michael is also working with fellow NIOSH trainee Kinley Taylor to draft formal written and video procedures for measuring joint angles using the Ergonomics Lab's electrogoniometer.

Michael and Kinley were also featured in press release about the Infant Car Seat handle they designed for ISE 796, a course they completed as part of their NIOSH requirements.

(Back to top)

January 2012

An abstract summarizing the Year 1 study, year 2 of data collection for the Haptic study concluded in December, and Michael worked with the project team in January developing a data analysis plan and identifying relevant results. In general, the results supported the results of the Year 1 study and continued to validate the usefulness of combining virtual reality with haptic control for psychomotor skill training.

Michael's primary contribution to the data analysis effort was to perform data mining of the simulation output files to identify whether the results could be used to objectively identify strategies implemented by the participants. According to current literature on Block Design performance, participants tend to implement one of two strategies: global or analytic. Participants employing a global strategy tend to view the design as a whole, as opposed to individual squares. These individuals tend to reorient each block through trial and error until they match the design. Participants who employ an analytic strategy mentally segment each design into squares corresponding to block faces and place each block directly to match the squares. The results of the Year 2 study are consistent with the literature. Participants employing an analytic strategy tend to perform better on the block design task. However, participants presented with visual aiding tend to rely on the assistance instead of their own strategies. An abstract summarizing the Year 1 study was accepted for presentation at the upcoming Applied Human Factors and Ergonomics conference under the title, "Assessment of a Virtual Reality-based Haptic Simulation for Motor Skill Training." Michael is lead author. Michael is also coauthor of the paper, "Usability Evaluation of a Virtual Reality System for Motor Rehabilitation" scheduled to be presented at the same conference by Woo Ram Jeon.

Michael's paper, "Comparison of Infant Car Seat Grip Orientations and Lift Strategies" was featured in a January edition of the Technician, NCSU's newspaper.

(Back to top)

February 2012

Michael Clamann, Kinley Taylor and Biwen Zhu completed the goniometer tutorial, now available at the NCSU HFE Lab Website:

Michael is also assisting with the data analysis effort, developing the year-end report summarizing the study, and preparing a manuscript for publication of the results in a refereed journal.

(Back to top)

March 2012

In March, Michael submitted a proposal to present research at the 2012 Meeting of the Human Factors and Ergonomics Society. The manuscript, "The Effects of Haptic and Visual Aiding on Psychomotor Task Strategy Development," is a summary of the data analysis effort he completed in January assessing trends in participant strategy and their relationships with perceptual style (i.e., field dependence and independence). He will also be demonstrating the block design software at the meeting with other human factors professionals during a session titled, "Me and My VE."

Michael and Dr. David Kaber were interviewed for IE magazine about the publication in Applied Ergonomics, "Comparison of Infant Car Seat Grip Orientations and Lift Strategies." The interview appeared in the March edition.

(Back to top)

April 2012

Analysis of the Year 2 data collection results continued in April. A draft manuscript of the results, including an updated literature review and methods sections, was also started in April to serve as a basis for the annual NSF report due in June. Michael made contributions to all the report sections.

The manuscript titled, "Investigating Human Performance in a Virtual Reality Haptic Simulator as Influenced by Fidelity and System Latency" was accepted as a Technical Correspondence for the IEEE Transactions on Systems, Man, and Cybernetics. Michael is a coauthor.

Michael also attended training sessions for the new faceLAB equipment acquired by the Cognitive Ergonomics Lab. This hardware tracks eye movements in high resolution and provides an additional means for objective data collection during experiments.

(Back to top)

August 2012

The report summarizing the second (Year 2) experiment was submitted to the NSF in May, with Michael making substantial contributions to the report. Work also began on the design of a new control to be used in future Haptic data collection. Currently, participants use a pen-like stylus to manipulate blocks in a virtual environment, which is a different experience compared to manipulating a block directly. Because of this difference, some members of the research team are designing a block identical in size to the cubes used in the native Block Design task. This block will be fit on the Phantom Omni haptic device and used for data collection. It is expected that use of this improved control device will make the virtual reality task more similar to the native task.

Michael's manuscript, "The Effects of Haptic and Visual Aiding on Psychomotor Task Strategy Development," was accepted for presentation at the 2012 Meeting of the Human Factors and Ergonomics Society. He will present his paper to the Virtual Reality Technical Group in October. This will be the second conference where Michael will present results of the Haptic study. He also presented the results of the Year 1 study at the Applied Human Factors and Ergonomics conference in July.

In August Michael also helped draft a manuscript to publish the results of the Year 2 study.

(Back to top)

September to October 2012

In September, Michael received IRB approval for his dissertation experiment design. His dissertation research involves comparing the effectiveness of different types of haptic aiding for training in a writing task. The topic was motivated by the haptic research he has been performing in the Ergonomics Lab.

In October, Michael traveled to Boston to present a paper at the 56th Annual Meeting of the Human Factors and Ergonomics Society. His paper, "Effects of Haptic and Visual Aiding on Psychomotor Task Strategy Development" was an analysis of how the various types of visual and haptic aiding used in the NSF Haptic research influenced individual strategy development in the block design task. The results showed that participants receiving visual aiding tended to rely on the assistance rather than form their own strategies. Furthermore, participants receiving combinations of visual and haptic aiding performed worse than those receiving visual only assistance, which was counter to expectations. Michael was also coauthored a paper, "Me & My VE," with Dr. Kaber, which demonstrated the haptic apparatus used for the current NSF Haptic research and was also presented at HFES 2012.

In addition to HFES, Michael is also preparing an abstract for the HCII 2013 conference and worked with Dr. Guk-Ho Gil to complete a manuscript of the most recent NSF Haptic study for the journal Presence.

(Back to top)

November to December 2012

In November and December, the NSF Haptic project team worked to finalize the design of the Year 4 experiment, which is expected to include participants with a history of minor traumatic brain injury. The final experiment will include a mixed population of unimpaired and impaired participants with baseline and post-testing to include fMRI tests as an objective measure of changes in processing due to exposure to the VR-based haptic simulations. The project team also designed a new haptic control grip. The new grip replaces the pen-like stylus with a block to better simulate the tactile sensation of manipulating blocks in the block design task.

This month Michael also completed a manuscript, "Effects of Laptop Touchpad Texturing on User Performance," for submission to the International Journal of Human-Computer Interaction. This paper is an investigation of how texture and roughness of laptop touchpads could affect performance in a computer-based pointing task. He also coauthored a poster with Wooram Jeon for the KSEA symposium at NC State.

In November, Michael was also inducted into Alpha Pi Mu, the Industrial Engineering honor society.

(Back to top)

January to February 2013

In January Michael submitted his final draft of his dissertation proposal to his committee. In February, he received an unconditional pass for the defense of his proposal titled, "Comparison of adaptive haptic forces in a virtual environment for fine motor skill training." He expects to complete his dissertation in the Summer of 2013.

In January and February, Michael presented the NSF Haptic research to two tour groups visiting the Ergonomics Lab, Including EPA staff, students at the Ergonomics Center, and visiting faculty from Georgia Tech.

(Back to top)

March 2013

In March Michael presented current research on Virtual Reality and Haptic control to the ISE Advisory Board, visiting faculty from University College London, and faculty from the NCSU Vet School. for the latter prestentaion, Michael prepared a poster with Wooram Jeon and Wenqi Ma for presentation at a Biomedical Engineering open house at NC State.

Michael’s abstract to the HCII 2013 conference was accepted for presentation, and he subsequently submitted the full paper, "Comparison of Enhanced Visual and Haptic Features in a Virtual Reality-Based Haptic Simulation," which summarizes the results of the Year 3 NSF Haptic study. The conference will take place in July.

(Back to top)

April 2013

In April the NSF haptic team continued to prepare the experiment materials for the Year 4 NSF Haptic study incorporating healthy participants and veterans with a history of mTBI.

Michael also prepared and submitted a manuscript for the IEEE International Conference on Systems, Man and Cybernetics (IEEE SMC 2013) titled, "Evaluation of a Virtual Reality and Haptic-Based Block Design Test." This paper further investigates the VR Block Design task developed for the NSF Haptic research as a tool for psychomotor testing comparable to traditional block design. The results show that although there are differences between the fine motor aspects of the two types of tasks, the cognitive requirements are similar.

In April, a manuscript detailing the Year 2 NSF Haptic study was officially accepted for publication by Assistive Technology. Michael was primarily responsible for revising the paper through two review cycles and completing the final approved version. The paper, "Evaluation of an Augmented Virtual Reality and Haptic Control Interface for Psychomotor Training," is expected to appear in an upcoming AT publication.

(Back to top)

Summer 2013

This summer, Michael led the effort to complete the last year of data collection for the Year 4 NSF Haptic study NSF. For this final phase of the NSF work, the study compared two different types of haptic controls. In previous years, experiments were designed to compare the learning effects of various types of haptic and visual displays to native forms of fine motor tasks. For the last study, the team compared the stylus control used in previous years to a new block-shaped control designed to represent the blocks used in the native form of the Block Design test. The VR simulation of the Block Design test used for training was identical to the simulation used in previous years. The goal of the work was to determine whether the stylus or the block control would be more effective for Block Design training and learning. It was expected that the block control would be more effective, as it better replicated the feel of the blocks used in the native form of the task, which, in previous years, exhibited speed advantages over the stylus. A new task, representing an occupational assembly task, was added to the test battery. The intent was to determine whether the block-shaped control would also improve assembly task performance as a result of training. The assembly task used for the study required building a model of a helicopter using Legos. Data collection was performed in July and August.

The experiment design included objective measurement of neuro-cognitive responses using fMRI at Duke Medical Center in Durham, North Carolina; however, participant attrition prevented the collection of a representative sample size.

In June, Michael completed Phase 1 of data collection for his dissertation. Nine participants visited the lab and traced characters from a foreign alphabet. The goal of Phase 1 was to identify patterns of motor learning based on speed and accuracy. The resulting models were consistent with expectations and showed that participants have consistent speed improvements when leaning a novel writing task. The models will be used to control adaptive haptic aiding designed to accelerate learning and will be implemented in Phase 2 of the experiment, scheduled for the fall following additional software updates.

Michael submitted the manuscript, "Effects of Laptop Touchpad Texturing on User Performance" to the International Journal on Human-Computer Interaction (IJHCI). This paper summarizes an experiment that compares various types of laptop touchpad texturing on speed and accuracy.

The manuscript "Evaluation of a Virtual Reality and Haptic-Based Block Design Test" submitted in April was accepted for presentation at the SMC 2013. Michael will travel to Manchester, UK in October to present the work.

(Back to top)

September 2013

Data collection was completed for Year 4 of the NSF Haptic Study. 21 participants trained for approximately 10 hours each spread across 8 sessions. The results will now be analyzed and Michael will prepare the final report to the NSF summarizing the work.

In September a new study utilizing the VR Block Design simulation was started. A group of undergraduate researchers began developing an experiment (as part of a class project) using EMG measurement to compare levels of exertion between the stylus and prototype block controls used in the Year 4 NSF Study. Michael will help the team prepare the materials and train in using the EMG equipment.

(Back to top)

October to November 2013

In October, Michael was interviewed along with other members of the Ergonomics lab by INDYWeek magazine. The article featured current trends in virtual reality research in the Raleigh/Durham/Chapel Hill area. Michael presented the VR Block Design simulation used for the NSF Haptic project. The article is available online:

Michael traveled to Manchester UK for the IEEE SMC 2013 conference where he presented the paper, "Evaluation of a Virtual Reality and Haptic-Based Block Design Test." A paper Michael coauthored, "Validation of a Haptic-Based Simulation to Test Complex Figure Reproduction Capability," was accepted for publication in the IEEE Transactions on Systems, Man & Cybernetics.

Throughout October and November, Michael finished testing 32 participants in Phase 2 of his dissertation research. He expects to analyze the data in December and defend the research to his dissertation committee in January.

(Back to top)

December 2013

In December, Michael won an $8,000 NC OSHERC Pilot Project Award for the project “Electromyography Study of Native Block Design vs. Virtual Reality Simulation.” This award will fund new EMG equipment and experiment apparatus for the project started by the undergraduate team in September. The award will support additional data collection by a graduate research team in 2014.

The paper, "Effects of Laptop Touchpad Texturing on User Performance," submitted in June to IJHCI, was accepted for publication and will appear in an upcoming issue.

(Back to top)

January 2014

In January, Michael successfully defended his doctoral dissertation, “Adaptive haptic forces in a virtual environment improve fine motor skill training” under the direction of Dr. David Kaber. The work compared the training effectiveness different types of haptic aiding, including virtual fixtures and error amplification, as well of adaptive forms of aiding that were modified over time based on participant learning. The results showed that adaptive forms of aiding were more effective for improving speed, while error amplification forms of haptic aiding were more effective for training accuracy. This also means that adaptive forms of error amplification were effective for training in speed and accuracy.

Dr. Clamann will be starting a research position in industry in February.

(Back to top)