Author(s): Diane Schretzman Mortimer, MD, Fiza Alam, BS, Lucas Dornan, BS
Originally published: August 4, 2017 Last updated: May 22, 2025
Jump to:

Introduction

Rehabilitation medicine utilizes the World Health Organization Functioning Model to effect positive changes in health, addressing not only body structure/ function, activity and participation, but also the contextual factors of environment and the person.1 Rehabilitation team members quantify functioning limitations, define targets, and implement interventions to bridge this gap. The past century has seen significant changes in these clinical strategies to improve functioning, reflecting a change in the underlying models of learning and behavior modification, in line with the practice of evidence-based-medicine.

“Learning” includes skill and knowledge acquisition as well as behavior modification.

Learning is broadly classified into

  • Explicit or Declarative – active process of learning
  • Implicit – mechanisms include
    • Practice resulting in automatic task performance (Procedural learning)
    • The result of stimuli application
      • Repeatedly (Associative: e.g., Habituation, Sensitization)
      • Associated with another stimulus or result (non-Associative: e.g., Classical and Operant Conditioning).

Overall, learning and behavior modification models attempt to explain improvement in functioning/ skill acquisition, either by recovery (of prior mechanism) or compensation (switch to alternative mechanism).2

Relevance to Clinical Practice

Motor learning principles underlie commonly used rehabilitation strategies.

From the medicine perspective, the motor system is comprised of the pyramidal and extrapyramidal systems. Consistent with the functioning approach to health is a relatively new concept (from the physical therapy perspective) that integrates the medical motor system with other anatomic/ physiologic systems. 

The movement system is a physical therapy construct that integrates multiple body structural systems, including the Neuro-musculoskeletal, Cardio-pulmonary, Endocrine, and Integumentary, into a functioning system that enables mobility, ADL, cognition and communication activities, in turn facilitating participation.3

The following sections briefly present basics of how the motor system is controlled (Motor control), how the motor system learns (Motor learning), and the relationship with neuroplasticity.

Motor control and learning

Motor control theories attempt to explain how the motor system is controlled.

Theories

The older theories of motor control were the basis for the historical Neurofacilitation approach (e.g., Bobath, Brunnstrom, Rood) for rehabilitation, while the newer motor control theories have driven the current utilization of the Task-Oriented Approach.2

The older theories of motor control include

  • Reflex Theory by Sherrington
  • Hierarchical Theory – top-down control in the nervous system
  • Motor Program Theory – the central pattern generator helps in movement

The newer theories of motor control include

  • Systems Theory – movement optimization in response to a combination of external forces, internal state and redundancy
  • Ecological theory – perceptual information guides goal directed action

Conceptualizing motor control

An overview of the concept of control systems is presented for the reader to understand the common framework for these theories. Broadly defined, control systems relate input to output in varying configurations.

  • Input system signals for the movement system can include vision, hearing, sensation, smell, and taste.
  • The output is motor task execution by the motor system.
  • The control systems can be thought of as the anatomic-physiologic systems that drive the intermediate steps in this process, using a combination of feedback (loop) and feed-forward (internal prediction) mechanisms.
  • Intermediate steps conceptualized from input (sensory) to output (motor) can include
    • perception / association (sensory cortex and association areas)
    • conceptualization (prefrontal cortex and association area)
    • plan / program (supplemental motor cortex, basal ganglia, cerebellum)
    • activation (primary motor cortex).2

A combination of these anatomic structures and physiologic functions driving the motor or movement system can be thought of as a Learning Network.4

Multiple non-mutually-exclusive learning networks have been proposed in literature for explaining the systems underlying distinct types of motor learning.5

Motor learning theories attempt to explain processes underlying movement being learned.

Theories

Older theories include

  • Closed-loop theory (sensory feedback guides skill acquisition)
  • Schema theory (open-loop process that allows improved learning with a variety of practice settings by formation of a generalized motor program).

Newer theories include

  • Ecological theory (optimization based upon perceptual as well as motor cues)
  • Stages of Learning theories2

Conceptualizing motor learning

There are several types of learning, that often overlap

  • Fast learning for routine tasks
  • Slow learning for more skilled tasks such as mastering an instrument
  • Offline consolidation for explicit learning
  • Long term retention of skills5

The process of learning, or skill acquisition can be conceptualized as being comprised of multiple sequential stages.2,5

  • Initial understanding and trying multiple strategies,
  • Identifying and consistently utilizing the optimal or efficient strategy (Adaptation),
  • Developing lasting Proficiency and/or Automaticity via motor programs.

From a neuromechanical perspective, for a given task many functionally equivalent motor behaviors (Motor abundance), and many movements by multiple combinations of muscles (Multifunctionality) are possible.

However, specific motor patterns, also referred to as Motor modules develops for that activity for that individual. This determination is guided by biomechanical task relevance (Motor structure), preciseness of movement required (Motor variability) and by Individuality.6

At the cerebral level, as skill is acquired, the activation of the Learning Network fades, except for the cerebellum. It is interesting to note that activation of multiple areas reemerges during stroke recovery.4

Examples of motor learning mechanisms and anatomic correlates of common task-oriented rehabilitation techniques are presented below

  • ADL training is accomplished by Learning through instruction and is hypothesized to be mediated by the prefrontal cortex.
  • Constraint induced movement therapy is based upon Reinforcement by reward or failure and is proposed to activate the cerebrum and basal ganglia.
  • Split-belt treadmill training is accomplished by Error-based adaptation and is considered to be mediated by the cerebellum and parietal cortex.6

Neuroplasticity

Neuroplasticity is defined as the nervous system’s ability to adapt its function and structure in response to stimuli and occurs at multiple levels.

  • Gene expression (e.g., Immediate Early Genes, IEGs in the motor M1 area).
  • Cellular plasticity (synaptic long-term potentiation and depression, or LTP and LTD, respectively).
  • Organ system level in the cortex, subcortical structures (e.g., cerebellum) and the spinal cord.7

Major principles of neurorehabilitation proposed to promote neuroplasticity include

  • Repetition to maintain and/or improve function,
  • Incorporating individual characteristics such as subject age and underlying stage of recovery, ensuring relevance and specificity of the therapy to the impairment
  • Increasing the challenge level iteratively
  • Utilizing therapy analogous to medicine in terms of optimal intensity, duration and frequency.
  • Carryover of learning from one activity to another can be generalized or inhibited, referred to as transference and impairment respectively.8

Current neurorehabilitation strategies can be broadly defined as compensatory, restorative, or adaptive.8 The examples below illustrate how neurorehabilitation principles underlie current rehabilitation approaches to individual impairments.

  • For aphasia, the principle of adding neuromodulation techniques such as transcranial direct current stimulation (tDCS) and fMRI related real-time neurofeedback to speech and language therapies (SLTs)is considered superior to SLTs alone. This approach underlies optimal intensity of therapy.9
  • For paresis, approaches include priming the central system, or augmenting the peripheral system, or a combination. Examples of priming include Sensory Stimulation including tactile, visual and kinesthetic, Guided Motor and Visual Imagery, and Action observation techniques such as mirror therapy. CNS stimulation techniques, including DCS (Transcranial Direct Current Stimulation) and TMS (Repetitive transcranial magnetic stimulation), as well as pharmacologic agents such as Amantadine and Methylphenidate can also be used to directly prime the nervous system. Examples of the augmentation include Functional electrical stimulation of nerves (TENS) and muscles with and without biofeedback.10
  • For apraxia, techniques to induce learning include internal (describing to self)/external (pictures) strategies, Errorless Completion (subject imitates examiner activity), and Action Observation and Execution.11
  • Constraint Induced Motor Therapy (CIMT) is proposed to help relearn from impairments related with Motor neglect, relying on mass repetition.8
  • For impaired mobility, as in spinal cord injury, neurorehabilitation techniques such as body weight supported treadmill training (BWSTT) have targeted the proposed Locomotor or Central Pattern Generator (CPG), based primarily upon animal experiments such as Sherrington’s cat. However, much of the literature today shows a move beyond the CPG toward incorporating kinesthetic feedback, lower limb kinetics and kinematics in the form of ankle robots for training and even on-the-ground training12 and increasing spinal circuit excitability by methods such as transcutaneous spinal direct current stimulation.13

Learning principles have been applied to musculoskeletal rehabilitation as well, with focus on goal-direct skilled training as opposed to strength training. As expected and evidenced by reduced cortical excitability, pain reduces skill acquisition and there is some evidence skilled training helps reduce and prevent chronic pain.8,14

Cutting Edge/Unique Concepts/Emerging Issues

Functional MRI (fMRI) has greatly improved our understanding of learning mechanisms and its anatomic correlates. For instance, emerging evidence using fMRI demonstrates human motor sequence learning occurring outside of the motor cortex, increasing the significance of premotor and parietal areas of the brain.15

Another development that is becoming instrumental is the field of Computational Neurorehabilitation. Along with rehabilitation robotics, computational modeling of motor tasks is also helping in development of evidence-based rehabilitation approaches and tools. Such computational models apply quantitative activity data (collected using sensors) and rehabilitation session data (collected from rehabilitation robots or practice sessions) to validated physiologic motor-sensory internal state conditions to predict functional outcomes in terms of movement and forces (kinematics and kinetics) over time.16

Virtual reality (VR) training has become another major motor learning tool, especially with the pandemic limiting physical interaction and subsequent availability of multiple accessible systems. Emerging research is demonstrating that at least some skills learned in VR training are transferrable to real-world situations.17,18

Another paradigm shift is modeling individual trajectories of change as opposed to just initial and final scores using mixed regression models to understand individual-level learning.19

Gaps in Knowledge/Evidence Base

It is important to acknowledge that the many models of learning and behavior modification are just that – models. They approximate our understanding of recovery from pathology such as stroke and SCI and form the basis of current rehabilitation.

Newer technology, such as functional MRIs, computational modeling, and rehabilitation robotics, is helping to refine these models, optimize rehabilitation approaches, and develop novel rehabilitation tools to improve functioning and quality of life. At the same time, there is no substitute for motivation, and subject engagement is crucial to recovery. Multidimensional approaches such as the Accelerated Skills Recovery Program (ASRP) that combines capacity, skill, and motivation could be utilized to optimize learning and then optimize rehabilitation and recovery.18,20

References

  1. World Health Organization. Towards a Common Language for Functioning, Disability and Health:  The International Classification of Functioning, Disability and Health, Geneva, 2002.
  2. Shumway-Cook A, Woollacott MH, Rachwani J, et al. Motor control: Translating Research into Clinical Practice. 6th ed. Philadelphia: Lippincott Williams & Wilkins; 2023.
  3. Sahrmann SA, The human movement system: our professional identity, Phys Ther. 2014;94(7):1034-42.
  4. Ween JE. Functional Imaging of stroke recovery: An ecological review from a neural network perspective with an emphasis on motor systems. J Neuroimaging 2008;18:227-236.
  5. Krakauer JW, Hadjiosif AM, Xu J, et al. Motor learning. Compr Physiol. 2019;9(2):613–663.
  6. Ting H, Chiel HJ, Trumbower RD, et al. Neuromechanical principles underlying movement modularity and their implications for rehabilitation. Neuron. 2015; 86(1): 38–54.
  7. Marzola P, Melzer T, Pavesi E, et al. Exploring the role of neuroplasticity in development, aging, and neurodegeneration. Brain Sci 2023; 13(12): 1610. Doi: 10.3390/brainsci13121610.
  8. Nilsen DM, Winterbottom L, Goldberg C. Fundamentals of neurorehabilitation. Rehabilitation Robots for Neurorehabilitation in High-, Low-, and Middle-Income Countries. Academic Press, Cambridge MA: 2023.  25–37.
  9. Shah-Basak P, Boukrina O, Li XR, et al. Targeted neurorehabilitation strategies in post-stroke aphasia. Restor Neurol Neurosci 2023. 41(3–4): 129–191. Doi: 10.3233/rnn-231344.
  10. Stolbkov YK, Gerasimenko YP. Observation of motor actions as a tool for motor rehabilitation. Neurosci Behav Phys 2021. 51(7): 1018–1026. Doi: 10.1007/s11055-021-01160-9.
  11. Stoykov ME, Corcos DM, Madhavan S. Movement-based priming: Clinical applications and neural mechanisms. J Mot Behav 2017. 49(1); 88–97. Doi: 10.1080/00222895.2016.1250716.
  12. Calabrò RS, Sorrentino G, Cassio A, et al. Robotic-assisted gait recovery following stroke: A systematic review of clinical guidelines and practical clinical recommendations. Eur J Rehabil Phys Med 2021. 57(3): 460-471. Doi: 10.23736/S1973-9087.21.06887-8.
  13. Comino-Suárez N, Moreno JC, Megia-Garcia A, et al. Transcutaneous spinal cord stimulation combined with robotic-assisted body weight-supported treadmill training enhances motor score and gait recovery in incomplete spinal cord injury: A double-blind randomized controlled clinical trial. J Neuroeng Rehabil 2025. 22(1). Doi: 10.1186/s12984-025-01545-8.
  14. Van Dillen LR., Lanier VM, Steger-may K, et al. Effect of motor skill training in functional activities vs strength and flexibility exercise on function in people with chronic low back pain. JAMA Neurol 2021. 78(4); 385. Doi: 10.1001/jamaneurol.2020.4821.
  15. Berlot E, Popp NJ, Diedrichsen J. A critical re-evaluation of fMRI signatures of motor sequence learning. eLife 2020. 9:e55241. doi:10.7554/eLife.55241.
  16. Reinkensmeyer DJ, Burdet E, Casadio M, et al. Computational neurorehabilitation: Modeling plasticity and learning to predict recovery. J Neuroeng Rehabil 2016. 13(1):42.
  17. Kim A, Schweighofer N, Finley JM. Locomotor skill acquisition in virtual reality shows sustained transfer to the real world. J Neuroeng Rehabil 2019. 16(1):113. doi: 10.1186/s12984-019-0584-y. PMID: 31521167; PMCID: PMC6744642.
  18. Levac DE, Huber ME, Sternad D. Learning and transfer of complex motor skills in virtual reality: a perspective review. J Neuroeng Rehabil 2019.16(1):121. doi: 10.1186/s12984-019-0587-8. PMID: 31627755; PMCID: PMC6798491.
  19. Anderson DI, Lohse KR, Lopes TCV, et al. Individual differences in motor skill learning: Past, present and future. Hum Mov Sci 2021. 78:102818. doi: 10.1016/ j.humov.2021. 102818. PMID 34049152.
  20. Winstein CJ, Kay DB, Translating the science into practice: shaping rehabilitation practice to enhance recovery after brain damage. Prog Brain Res. 2015; 218:331-60.

Original Version of the Topic

Prateek Grover, MD. Models of Learning and Behavioral Modification in Rehabilitation. 8/4/2017

Previous Revision(s) of the Topic

Prateek Grover, MD, MHA. Models of Learning and Behavioral Modification in Rehabilitation. 4/27/2022

Author Disclosure

Diane Schretzman Mortimer, MD
Nothing to Disclose

Fiza Alam, BS
Nothing to Disclose

Lucas Dornan, BS
Nothing to Disclose