A therapist monitors a stroke patient who is using the Aretech ZeroG-Lite treadmill-based gait training system and the Woodway split-belt treadmill as a platform for gait training with error augmentation.

A therapist monitors a stroke patient who is using the Aretech ZeroG-Lite treadmill-based gait training system and the Woodway split-belt treadmill as a platform for gait training with error augmentation.

By Jeff Berger, MPT

“A person who makes few mistakes makes little progress.” These words written by American author and aphorist Bryant McGill describe quite simply an emerging paradigm that is proving successful in the realm of neurorehabilitation—the paradigm of error augmentation.

The goal of improved motor function seems most attainable for neurorehab patients today when physical and occupational therapists employ interventions that are grounded in motor learning principles. Motor learning, or the process of developing or changing motor skills through practice and experience, can be enhanced if therapists create training scenarios that challenge patients to adapt their motor activity. Error augmentation, by increasing asymmetries and movement errors to which patients must then adapt, provides therapists inroads to connecting with a patient’s central nervous system, facilitating neuroplastic cortical change.

I had my first real taste of error augmentation at a continuing education course at The Rehabilitation Institute of Chicago (RIC) several years ago. It was at this course that I had the opportunity to interact with T. George Hornby, PT, PhD, and Jennifer Moore, MPT, NCS, DHS, both experts in the field of locomotor retraining. During the course, I was able to take a walk on a split-belt treadmill and experience true motor adaptation. When a person walks on a treadmill with each leg moving at a different speed, there is a quick accommodation that occurs as the faster leg spends less time in stance phase and the stride for that leg lengthens.

Once the belts are returned to the same speed, there is a period of time when the person continues to walk with the adapted pattern. When a patient with an asymmetric step length (characteristic of a hemiparetic gait pattern) is trained on the split-belt with the short-stride leg moving faster, the motor adaptation that occurs can help to correct the asymmetry. At the time I attended the course at RIC, most of the research linking split-belt treadmill training to locomotor adaptation for stroke survivors hadn’t been published, but the course sparked a new understanding for me of the potential that this technology could have for neurorehab patients. Thus, the vision for our own gait lab was born.

Not long after the course at RIC, St John Northshores Hospital—where I worked on the inpatient rehabilitation unit—was closed and the entire unit was moved to the health system’s medical center campus in Detroit. The move was an opportunity to design a rehab unit from the ground up and update equipment and rehab technologies. Development of a gait laboratory was an important element of the Cracchiolo Inpatient Rehabilitation Center project, and today the lab has been up and running for more than 2 years.

Therapist and patient using the Saebo ReJoyce with real-time error feedback and virtual reality interaction to assist with upper-extremity motor learning.

Therapist and patient using the Saebo ReJoyce with real-time error feedback and virtual reality interaction to assist with upper-extremity motor learning.

Equipped to Aggressively Challenge Patients

The two centerpieces of the St John Hospital & Medical Center (SJH&MC) Gait Laboratory are the Woodway split-belt treadmill from Woodway Inc, Waukesha, Wis, and Aretech ZeroG-Lite treadmill-based gait training system from Aretech, Ashburn, Va. The Woodway split-belt treadmill allows therapists to adapt patient walking patterns through error augmentation and the ZeroG-Lite provides dynamic body-weight support, allowing for early and intensive gait training interventions. The ZeroG-Lite system, which uses a harness suspended overhead to provide dynamic body weight support, enables the therapist to aggressively challenge patients with error-augmentation techniques with complete safety. The device can be used with patients up to 400 pounds, and is designed to have applications outside of stroke and neurological rehab that include amputee and orthopedic injury populations as well as individuals affected by multiple sclerosis, cerebral palsy, or spinal cord injury.

Several other manufacturers offer body weight support systems to the rehab market that are designed to provide utility in treating stroke patients and individuals affected by mobility impairments. Among them is the Vector Gait & Safety System from Valencia, Calif-based Bioness. The Vector uses a harness and overhead trolley system and is available in closed-loop or linear track designs. The Vector is engineered to accommodate 400 pounds static and 200 pounds dynamic body wight capacity. The SafeGait 360-degree Balance and Mobility Trainer from Gorbel Medical, Fishers, NY, is also a body weight support system patients use with a harness and overhead track. The SafeGait is designed to provide dynamic fall prevention and a barrier-free treatment zone.

As clinical application of error augmentation requires a therapist to first analyze a patient’s movement patterns to identify asymmetries and errors, the facility has recently added the Simi Aktisys Reality Motion System to the lab technologies. The Simi system software, capable of analyzing patient locomotion from different camera angles, allows therapists to detect even subtle gait asymmetries and then determine whether gait training interventions are having a genuine impact in terms of changing a patient’s walking pattern.

Product Resources

The following companies also provide products for stroke and neurological rehabilitation:







Clarke Health Care Products


GAITRite/CIR Systems Inc


Gorbel Medical/SafeGait


Motorika Medical Ltd




Saebo Inc






Vista Medical


Woodway USA


Tools to Treat (and Study) Asymmetries

Even before the SJH&MC Gait Laboratory opened its doors, error-augmentation techniques had been employed with success for neurorehab patients with a variety of diagnoses, including CVA, traumatic brain injury, incomplete spinal cord injury, multiple sclerosis, and Parkinson’s disease. I have even had success treating gait asymmetries resulting from orthopedic impairments. The ability to make dramatic changes in patient motor control over the past few years have fueled the team’s fire and compelled team members to take error augmentation to a completely new level.

Use of the split-belt feature on the Woodway treadmill to correct step-length asymmetries is only the first level of error augmentation therapists employ when working to adapt patient walking patterns. If patients have excessive internal or external rotation in one or both of their legs when walking, therapists will spiral resistance bands around the culprit legs to augment the impairment. If a patient walks with a “scissoring” pattern, bands are used to pull the limbs even further into an adducted “scissor” alignment. For patients who demonstrate asymmetry in their weight-bearing posture and lean to one side, bands are used to pull the trunk or pelvis even further to that side. When a patient walks with a flexed posture, a figure-8 strap is used around the patient’s shoulders and bands are attached to the front of the strap to accentuate the flexion posture. These error-augmentation strategies heighten patient knowledge of gait deficits and set the stage for motor adaptation and neuroplastic change.

An article published recently investigated patient perception of spatiotemporal gait asymmetries poststroke.1 The researchers concluded that there may be a threshold of error that patients must experience before they are consciously aware of their gait asymmetries, and that this awareness may be important for motor learning and long-term correction. This conclusion supports our use of multilayered error augmentation and has helped us to see that we may be on the right track in challenging patients with such intensity.

Gait training activities performed in the clinic with multilayered error augmentation.

Gait training activities performed in the clinic with multilayered error augmentation.

Drilling for Data to Improve Walking Outcomes

In the continued pursuit of optimizing patient walking outcomes, the facility is investigating adding technologies such as the Zeno Walkway from ProtoKinetics, Havertown, Pa, or the GAITRite Portable Gait Analysis Walkway System from CIR Systems Inc, Franklin, NJ. Likewise, we are exploring the possibility of adding wearable gait sensors such as the LEGSys system from Cambridge, Mass-based BioSensics, to the facility’s neurorehab arsenal. The walkway systems provide data about gait parameters such as step length, step width, stance time, and toe in/out angles, and would allow therapists to determine whether the outcomes of patient error-augmentation treadmill training are translating to overground walking. Wearable gait sensors can provide the same types of data about walking function, but would allow collection of real-time data and ability to perform gait and mobility analyses with patients outside of the hospital or clinic.

Along with gait rehabilitation, we have begun to employ error-augmentation techniques to improve patient motor control with functional activities such as sit-to-stand transfers and upper-extremity reaching. After attending the APTA 2015 Combined Sections Meeting, and hearing a presentation about hemispheric specialization and upper-extremity control mechanisms by Robert Sainburg, PhD, I now have an even better idea of how our rehab team can apply error-augmentation techniques to improve reaching ability and upper-extremity function.

We are also fortunate to have technologies such as the Motorika ReoGo system from Motorika, Mount Laurel, NJ, and the Saebo ReJoyce workstation from Saebo Inc, Charlotte, NC, that combine robotic technologies with virtual reality interaction. These technologies give patients real-time feedback when it comes to error, and allow therapists to design optimal motor learning training experiences for our neurorehab patients.

In looking toward the future for St John Hospital & Medical Center’s neurorehab program, we hope to throw our own hat into the clinical research ring and further investigate the use of error augmentation in optimizing motor learning outcomes. The technologies outlined in this article allow members of the therapy team to analyze patient movement patterns, maximize adaptable errors, and then objectively measure the effectiveness of treatment approaches. I would encourage therapists working in neurorehab to trial error-augmentation techniques in their own clinical settings and perhaps add to the clinical evidence base for this intervention strategy. Find the error in your patient’s ways, and accentuate the negative. PTP

Jeff Berger, MPT, is lead physical therapist at the Cracchiolo Inpatient Rehabilitation Center, St John Hospital & Medical Center, Detroit. For more information, contact PTProductsEditor@allied360.com.


  1. Wutzke CJ, Faldowski RA, Lewek MD. Individuals poststroke do not perceive their spatiotemporal gait asymmetries as abnormal. Phys Ther. 2015;95(9):1244-1253.