Constraints Shape Movement

Broader scientific context

Why can two athletes perform the same exercise differently, and how does the environment shape the solution?

Body dimensions, task goals, instructions, and equipment interact to define the movement an athlete can perform. Understanding those constraints makes coaching and equipment research more precise.

Key Takeaways

Describe the relationship before judging the movement

Movement reflects an interaction among the athlete, the task and the environment; the same instruction can create different physical demands for different people.

Constraints can enable and restrict action. More freedom, more support or a guided path serves a purpose only in relation to the training objective.

Equipment geometry, contact, stability, traction and available space are separate variables that can be described and tested.

A cue changes the information given to an athlete. A physical change alters the available movement options; the two interventions can interact.

Pressing experiments show that different equipment and technique conditions can change acute performance, muscle activity and modeled loads.

Assistance and guidance have task-dependent learning effects. Equipment claims should be tested in the performance environment they are intended to support.

The same instructions can produce different movements

The same bench height, grip distance and rack position can create different demands for two athletes. One reaches the floor comfortably; another adjusts the pelvis or foot position. One unracks easily; another needs a handoff.

Before correcting a movement, check whether the athlete needs a clearer cue, a setup adjustment, more capacity or different equipment.

Karl Newell’s 1986 framework describes coordination as an interaction among organismic (athlete), environmental and task constraints. Movement reflects the athlete’s characteristics, the surrounding conditions, and the activity’s goal and requirements.[1]

For the Research Hub, the value is practical. Human-Centered Strength Equipment begins with users and tasks. The Athlete-to-Bench Interface locates the contacts through which the repetition occurs. Constraints connect those engineering questions to the organization of movement.

A constraint can enable as well as restrict

A constraint changes the possibilities of a task. A floor prevents the feet from passing downward through it and provides a surface against which force can be applied. A handle restricts where the hand contacts an implement and permits a usable grip. External support can reduce one demand while allowing an athlete to pursue another.

“Constraint” is therefore not a synonym for defect. A machine’s guided path may be exactly what a training objective requires. A competition rule deliberately narrows acceptable performances. An adjustment may expand the number of users who can reach a useful position. The appropriate judgment depends on the objective and the resulting behavior.

The constraints model · illustrated

One movement, three interacting conditions

Unoccupied competition bench inside a power rack, with a loaded bar resting on the rack hooks.

Athlete

Proportions, capacity and experience change the demands of reaching and controlling the setup.

Task

The load, goal and repetition rules define what the athlete must accomplish.

Environment

The bench, rack, floor and available space establish actual contacts and boundaries.

Figure 1. Athlete, task and environment interact to shape movement. Newell’s framework provides the foundation; the lifting examples are AMM applications.[1]
Equipment can be classified as part of the task or environment. Describe the changed feature consistently: bench height matters through its effect on the athlete’s foot contact and setup.[1]

Body proportions, capacity and experience shape the task

Body dimensions affect how an athlete fits a task: limb lengths, shoulder breadth, torso shape, hand size, and segment proportions can change reaches and contact relationships. These structural features are joined by changing capacities such as strength, mobility, fatigue, confidence and experience.

Athlete constraints therefore operate on different timescales. Skeletal proportions may be effectively constant within a study. Fatigue can change within a set. Familiarity can change across sessions. Pain or injury history may influence the strategy an athlete selects today without dictating every future repetition.

Study detail: body proportions and available actions

Warren’s classic stair-climbing experiments examined the relationship between step height and the climber’s leg length. The work showed why action possibilities can be better understood in relation to the person than from an environmental dimension alone. This is the idea of an affordance: an opportunity for action arising from a relationship between an individual and the environment. [2]

Judge fit through the movement it allows

In the weight room, assess fit through the action it allows. The same rack position can permit an independent unrack for one athlete and require a handoff for another.

Fit also changes with the objective. A grip that an athlete can reach may not permit the wrist and elbow relationship intended for a particular exercise. A support can accommodate the torso while leaving insufficient room for the dumbbells. A comfortable demonstration with no load may fail to represent the setup under working conditions.

A useful observation consequently names both the athlete characteristic and the relevant task response. “Shorter athlete” is less informative than “cannot maintain the specified foot contact at this loaded bench height.” The latter describes a relationship that can be altered and retested.

The same logic helps coaches avoid mistaking every difference for a technical error. First identify which features must meet the exercise standard. Then determine which differences reflect a workable individual solution and which undermine the intended outcome.

An exercise name does not define every task demand

“Bench press” leaves many decisions open. The task can demand a maximum valid repetition, repeated submaximal work, a prescribed pause, a selected range, an explosive intent, or a controlled return to loading. Each objective changes what a successful solution needs to accomplish.

Load, repetition target, grip, tempo, range of motion, resistance direction, and instructions belong in that description. So do the criteria used to stop a set. Completing a required number of repetitions and preserving a velocity threshold create different decision problems even when the first repetition is identical.

Define the task before comparing performance

Goal
What must the athlete accomplish?
Conditions
Specify load, grip, tempo, range and instructions.
Success
State what makes a repetition valid and when the set ends.

Renshaw and colleagues describe a constraints-led approach in which coaches change targets, rules, space or implements to encourage useful movement solutions.[3]

Applied to lifting, the first step is simply to name the task. If the objective is a reproducible paused repetition, the pause and valid contact conditions need to remain part of the assessment. If the objective is to expose the athlete to a new range or implement, the new condition is part of the intervention.

Relative load needs the same care. A percentage of a maximum measured in one condition may not represent the same relative challenge in another. Using equal absolute loads controls the external mass; using condition-specific percentages addresses relative intensity differently. Both designs can be sensible, but they ask different questions.

Load, Range of Motion, Eccentric Overload & Rep Quality explains how to specify these training variables.

The apparatus defines contacts, boundaries, and information

Equipment geometry determines where surfaces and handles are located. Its adjustment system determines which of those relationships can change. Stiffness, compliance and stability describe how it responds to force. Friction concerns resistance to sliding between contacting materials. These properties should be named separately.

Turn an equipment feature into an observable question
Condition Movement question Useful observation
Bench height and rack position Can this athlete reproduce the intended start and foot contact? Loaded height, reach, contact coordinates and assistance needed.
Width, contour and edges Which body regions contact the support, and where is clearance available? Contact map, regional pressure and body motion.
Support stability Does the apparatus move or deform during the task? Displacement, tilt and deformation under representative loading.
Surface traction Does the athlete move relative to the support? Material-pair friction and torso drift, measured separately.
Handle or implement geometry Which grip orientations and movement paths are available? Hand position, wrist and elbow motion, implement path.
Available space Can the intended movement occur without contact with surrounding structures? Clearance through the complete repetition and setup.

Select equipment features around the goal. A guided path may support targeted loading; a freer implement may increase the coordination demands of the task.

Contact gives the athlete information about torso position and sliding. Track those observations alongside movement and performance when comparing surfaces.

In practical use, keep the components separate long enough to understand them. If a new surface changes height, contour, traction and firmness simultaneously, the experiment tests the complete condition. Identifying the responsible feature requires comparisons that isolate or systematically vary those properties.

Different implements produce different pressing tasks

Pivotal study · Within-person equipment comparison

Maximum load and muscle activity depended on the implement

Participants and protocol

Saeterbakken, van den Tillaar and Fimland tested 12 resistance-trained men in counterbalanced Smith-machine, barbell and dumbbell chest presses. Tests were separated by 3–5 days. Investigators measured one-repetition maximum and surface EMG from four muscles.

Findings

Dumbbell load was 17% below barbell load and 14% below Smith-machine load. Pectoralis-major and anterior-deltoid activity did not differ significantly when averaged over the whole repetition; biceps activity was higher as stability requirements increased, while triceps activity was lower with dumbbells. [4]

Study result · Saeterbakken et al., 2011

The implement changed the maximum load

Reported reduction in dumbbell 1RM relative to each comparator · same 12 men

Lower 1RM load (%) · each bar uses its named comparator

View exact values and study context
Reported pairwise 1RM differences; counterbalanced equipment tests.
Comparison Dumbbell load P value
Vs barbell 17% lower ≤0.001
Vs Smith machine 14% lower ≤0.001
Figure 2. Acute 1RM differences between implements in the same 12 men. Bars show the reported percentages.[4]

The same athletes lifted different maximum loads across implements, with muscle-specific changes in activity.

Technique can interact with the apparatus even when the implement stays constant. Noteboom and colleagues recorded 10 experienced athletes across 21 bench-press conditions using a 16 kg bar. Grip, shoulder-abduction targets and scapular instructions were combined with motion and force measurements, then entered into a musculoskeletal model. Selected estimated shoulder loads changed across technique conditions. The condition using a pool noodle to release scapular support did not significantly differ from neutral in joint reaction forces. [5]

Observe movement and force direction alongside the support condition to understand the athlete’s response.

The existing articles on Bench Pad Width & Shoulder Mechanics and Stability, Traction & Setup Repeatability develop the equipment details. Here they support a motor-control question: what demands did the changed condition create, and which solution did the athlete use?

Instructions and equipment changes affect movement differently

An instruction tells the athlete something about the desired action. A physical change alters the setting in which the athlete attempts that action. They often work together, but they are not equivalent interventions.

Ask an athlete to keep a bar within a particular path and the athlete must interpret, select and regulate the requested behavior. Place the bar in a fixed guide and some paths are mechanically unavailable. Both conditions may produce a similar-looking repetition while requiring different control.

A second example involves a setup the athlete cannot reliably reach. Repeating the cue more loudly does not alter the distance. Changing the rack setting changes the physical demand; instruction can then help the athlete use the newly available setup. The original difficulty may have reflected the interaction, rather than a failure to understand.

Observe how the athlete uses a new setup. Familiar strategies can persist until the athlete explores the alternative or the task requires a change.

Research framework: perception, action and exploration

Ecological-dynamics approaches examine how athletes perceive and use opportunities for action in their surroundings. Coaches can apply that framework by designing tasks that invite relevant exploration.[3]

Test the cue and the equipment together

For an equipment comparison, consider crossing instruction with condition: the same cue on both surfaces, and different cues on the same surface, where feasible. That design can distinguish effects associated with the apparatus from those associated with the instruction. It also captures a possibility missed by a simple before-and-after demonstration: the cue and equipment may interact.

Physical help can change what gets practiced

Assisted-movement experiments change the forces experienced during practice, then remove the assistance for testing. Walking-balance and steering studies illustrate how to assess that learning.

Adjacent evidence · Walking balance

Fewer practice errors did not produce better independent balance

Protocol

Domingo and Ferris assigned 40 neurologically intact adults to four groups. Participants practiced for 30 minutes on treadmill-mounted beams 1.27 or 2.5 cm wide, with or without spring assistance at the pelvis. Everyone completed unassisted tests before and immediately after practice.

Finding

Assistance reduced practice failures, while unassisted practice produced greater improvement on the immediate unassisted walking-balance test.[6]

Guidance has produced a different result in another task. Marchal-Crespo and colleagues studied simulated vehicle steering with a robotic wheel. Assistance was adjusted to need and faded during training. In the young-adult group, guided practice improved unassisted retention a week later relative to practice without guidance; the overall older-adult comparison was not significant. Benefits depended on the participant and task component. [7]

Guidance can change error, timing information, stabilization demand and access to the target action. Assess it against what the learner must perform after practice.

For lifting, describe what support changes during practice, then test performance in the athlete’s intended setup.

Practice should represent the performance it is intended to support

Representative learning design asks whether important information and action relationships in practice resemble those needed for the target performance. Pinder and colleagues developed this argument for sport research and training, emphasizing functionality rather than superficial visual similarity. A drill can look like competition while removing the information an athlete normally uses to act. [8]

Match the test to the athlete’s goal

For strength training, the target needs to be explicit. Is the athlete preparing for a competition bench, a general strength test, a machine-specific task, or tolerance for a range of pressing activities? Those goals can justify different apparatus and different tests.

If a lifter trains on an assistance device but intends to perform without it, transfer is part of the claim. If the equipment itself is the intended performance environment, independent performance on that equipment may be the appropriate endpoint. There is no scientific requirement to remove a useful permanent support merely to make an exercise “more functional.” Function is defined relative to the task.

Measure what remains after practice

Nijmeijer and colleagues’ 2025 review included 25 randomized trials with 1,020 participants and kinematic or kinetic tests at least one night after practice. Implicit-learning approaches showed a small pooled advantage over control conditions across varied sports and interventions.[9]

Evidence synthesis · Nijmeijer et al., 2025

Learning comparisons at a delayed test

Pooled change in biomechanical outcomes · Hedges’ g with 95% confidence intervals

Implicit vs control0.45

95% CI 0.115 to 0.780

Explicit vs control0.377

95% CI −0.214 to 0.967

Implicit vs explicit0.162

95% CI −0.036 to 0.360

Standardized effect · zero = no between-group difference

View exact values and study context
Published pooled estimates from Sections 3.5–3.7; positive values favor the first-named approach.
Comparison Hedges’ g 95% CI P value
Implicit vs control 0.45 0.115 to 0.780 0.010
Explicit vs control 0.377 −0.214 to 0.967 0.211
Implicit vs explicit 0.162 −0.036 to 0.360 0.109
Figure 3. Retained biomechanical outcomes, expressed relative to each study’s movement goal. Lines show 95% confidence intervals around pooled effects.[9]

The design feature worth carrying into equipment research is the delayed movement assessment. Measure the way the athlete performs after practice, under a defined test condition. A same-session video can show a changed solution; a delayed assessment asks whether that solution remains available.

This is also a useful reason to preserve a few stable reference tasks in a program. They provide a comparable setting in which to observe progress while other exercises or conditions are deliberately varied.

Change a condition for a stated reason, then observe the response

A constraints audit begins with a question specific enough to answer. Instead of asking whether a bench is generally better, ask whether its configuration allows a defined population to reproduce a prescribed setup, achieve a target movement, or express force under comparable conditions.

Use this workflow to make successful setups reproducible and problematic setups specific enough to investigate.

Small changes can also have linked consequences. Raising a support changes its relationship to both feet and rack. Altering grip changes more than hand spacing. A new contour can change contact and effective height together. Record those consequences so the intervention remains identifiable.

Choose measurements that distinguish the likely explanations for the observed change.

An athlete’s available solutions can change during recovery

Injury, pain, weakness and rehabilitation can change the individual side of the relationship. A previously familiar task may now be solved with a different distribution of motion, effort or support. The environment still matters, but it interacts with a different athlete state.

Injury Is More Than Tissue Damage examines the physical and neuromuscular dimensions of that change. When Compensation Becomes the Strategy then asks how practiced alternatives can evolve as capacity returns.

Assess each movement solution by what it achieves, where it places demand and how well it serves the current goal.

Equipment changes the conditions that athletes practice in. Test what athletes retain after practice and what carries over to other tasks.

The athlete brings the capacity. The task supplies the goal. The environment supplies real contacts, forces and boundaries. The repetition is the solution produced by their interaction.

References

  1. Newell KM. (1986). Constraints on the development of coordination. In MG Wade & HTA Whiting (Eds.), Motor Development in Children: Aspects of Coordination and Control (pp. 341–360). Martinus Nijhoff. doi:10.1007/978-94-009-4460-2_19. ↩
  2. Warren WH Jr. (1984). Perceiving affordances: Visual guidance of stair climbing. Journal of Experimental Psychology: Human Perception and Performance, 10(5), 683–703. doi:10.1037/0096-1523.10.5.683. ↩
  3. Renshaw I, Chow JY, Davids K, Hammond J. (2010). A constraints-led perspective to understanding skill acquisition and game play: A basis for integration of motor learning theory and physical education praxis? Physical Education and Sport Pedagogy, 15(2), 117–137. doi:10.1080/17408980902791586. ↩
  4. Saeterbakken AH, van den Tillaar R, Fimland MS. (2011). A comparison of muscle activity and 1-RM strength of three chest-press exercises with different stability requirements. Journal of Sports Sciences, 29(5), 533–538. doi:10.1080/02640414.2010.543916. ↩
  5. Noteboom L, Belli I, Hoozemans MJM, Seth A, Veeger HEJ, van der Helm FCT. (2024). Effects of bench press technique variations on musculoskeletal shoulder loads and potential injury risk. Frontiers in Physiology, 15, 1393235. doi:10.3389/fphys.2024.1393235. ↩
  6. Domingo A, Ferris DP. (2009). Effects of physical guidance on short-term learning of walking on a narrow beam. Gait & Posture, 30(4), 464–468. doi:10.1016/j.gaitpost.2009.07.114. ↩
  7. Marchal-Crespo L, McHughen S, Cramer SC, Reinkensmeyer DJ. (2010; online 2009). The effect of haptic guidance, aging, and initial skill level on motor learning of a steering task. Experimental Brain Research, 201(2), 209–220. doi:10.1007/s00221-009-2026-8. ↩
  8. Pinder RA, Davids K, Renshaw I, Araújo D. (2011). Representative learning design and functionality of research and practice in sport. Journal of Sport and Exercise Psychology, 33(1), 146–155. doi:10.1123/jsep.33.1.146. ↩
  9. Nijmeijer EM, Brals FD, Kempe M, Elferink-Gemser MT, Benjaminse A. (2025). How are athletes trained to move? A systematic review exploring the effects of implicit and explicit learning on biomechanics of sport-specific tasks. Journal of Biomechanics, 184, 112671. doi:10.1016/j.jbiomech.2025.112671. ↩