Spontaneous transitions in the coordination of a whole body task
Introduction
Spontaneous transitions between different patterns of coordinated movement have the potential to provide important information about the role of neuromuscular constraints in the coordination and control of movement. Such transitions have been very well described and modeled in cyclic movements of peripheral limbs and limb segments (fingers, hands, feet, and arms). However, with the exception of gait transitions (Abernethy, Burgess-Limerick, Engstrom, Hanna, & Neal, 1995; Diedrich & Warren, 1995; Hanna, Abernethy, Neal, & Burgess-Limerick, 2000) whole body movements have not been similarly addressed. This research explores the boundaries of applicability of this approach by investigating the characteristics of spontaneous transitions in another whole body movement – one involving repetitive lifting and lowering of a load. Lifting and lowering is a particularly attractive task for study because of the amenability of the task constraints (height, mass, frequency) to independent manipulation and the relatively detailed biomechanical modeling of the movement which has been undertaken.
The first example of spontaneous transitions in human movement to be systematically examined was bimanual finger tapping (Kelso, 1981; Yaminishi, Kawato, & Suzuki, 1980). If a person is asked to tap both index fingers at a common frequency, but 180° out of phase (i.e., one finger reaches maximum flexion at the moment the other reaches maximum extension), this mode of coordination is stable and can be maintained accurately, at least at low tapping frequencies. If the frequency of finger tapping is gradually increased, the 180° mode can be maintained for some time. However at a critical value of the tapping frequency the coordination of the fingers suddenly (within one cycle) and spontaneously (unintentionally) changes so that flexion and extension of the two fingers occurs synchronously. In-phase coordination is now the only stable mode and this mode of coordination remains stable if the tapping frequency continues to increase (Kelso, 1984).
This phenomenon has been examined in some detail (e.g., Kelso, Scholz, & Schöner, 1986; Schöner & Kelso, 1988). If tapping frequency is decreased after the transition the system remains in the in-phase mode and no second transition occurs (i.e., the system exhibits hysteresis). If the system starts at low frequency in the in-phase mode, and then frequency is increased, no spontaneous transition occurs. The in-phase mode is more stable than the out-of-phase mode (indicated by lower variability). Intentional production of intermediate relative phase modes is difficult and the resulting coordination is much more variable than the in-phase or out-of-phase modes.
Similar observations have been made of transitions between patterns of intra-limb coordination. Kelso, Buchanan, and Wallace (1991) described spontaneous transitions between in-phase and out-of-phase coordination patterns of wrist and elbow flexion and extension as a function of frequency and whether the forearm was pronated or supinated. Buchanan and Kelso (1993) extended this observation, describing spontaneous transitions as a function of changes in forearm position while frequency remained constant.
This paper concerns an example of spontaneous transitions between different patterns of lower limb coordination as function of lift height during repetitive lifting and lowering. This example of spontaneous transitions in a whole body task provides an experimental paradigm which has potential to provide considerable insight into the role of musculoskeletal constraints in determining inter-joint coordination.
Manual lifting techniques are commonly differentiated on the basis of the posture adopted to lift a load. Distinctions are typically drawn between techniques involving a posture at the start of the lift in which the knee joints are flexed only slightly, if at all, and the trunk inclined substantially (a stooped posture), and a technique involving substantial knee flexion at the start of the lift and less trunk inclination (a squat or semi-squat posture) (e.g., Grieve, 1974; Kumar, 1984; Toussaint, van Baar, van Langen, de Looze, & van Dieën, 1992). Although the biomechanical consequences of different lifting techniques have been reported (e.g., Anderson & Chaffin, 1986; Gagnon & Smyth, 1992; Kumar, 1984; see Van Dieën, Hoozemans, & Toussaint, 1999, for a recent review), and an attempt has been made to define the different techniques quantitatively (Burgess-Limerick & Abernethy, 1997a), it is not known whether these different techniques represent qualitatively different modes of manual lifting, or points on a continuum. If the former were true then spontaneous transitions between different techniques might be anticipated in appropriate circumstances. In addition to providing a fruitful paradigm for investigating mechanisms of control and coordination, the observation of spontaneous transitions would also provide a justification for the previously assumed distinction in lifting techniques, and provide a more principled basis for subsequent biomechanical investigations.
Manual lifting, whether performed discretely or repetitively, involves cycles of flexion and extension movements of limbs and trunk. The pattern of self-selected, or preferred, lifting most frequently observed involves a semi-squat posture and simultaneous knee extension, hip extension and ankle plantar-flexion (Burgess-Limerick, Abernethy, Neal, & Kippers, 1995). Stooped postures were observed less frequently, and analysis of data from individual subjects gathered in previous experiments involving discrete lifting revealed some evidence for the existence of qualitatively different modes of discrete manual lifting (bi-modal distributions in parameters describing the postures adopted to lift loads; Burgess-Limerick and Abernethy, 1997a, Burgess-Limerick and Abernethy, 1997b; see also Van Dieën, van der Burg, Raaijmakers, & Toussaint, 1988). An influence of starting height on the relative frequency of adoption of these modes was noted (stooped postures are more frequently adopted to lift loads from relatively high starting heights; Burgess-Limerick & Abernethy, 1997b; and semi-squat postures were almost always adopted to lift loads at floor height). However, transitions between different modes were not observed because, in these experiments, the lifting was either performed as discrete trials; or when repetitive lifting and lowering was examined, the experimental manipulations involved separate trials in different height conditions. The observation of transitions between different modes is impossible in the first case, and unlikely in the second. This paper describes an experimental paradigm in which such transitions were reliably observed and describes the fundamental characteristics of the phenomenon.
Section snippets
Method
Eleven participants (6 male, 5 female; aged 19–44 years) performed 12 trials of a task in which a 1 kg load was continuously raised and lowered. The lowest height of the load was indicated by a shelf connected to a computer controlled torque motor. The task involved repetitively lifting the load from this shelf to a constant height (waist level) then lowering the load back to the shelf. The height of the shelf was adjusted throughout each trial. A ramp protocol was employed for each trial such
Results
Two qualitatively different patterns of coordination were spontaneously adopted when the lifting technique was self-selected: a squat technique and a stoop technique. The squat technique is characterised by a moderate range of motion at ankle, knee, and hip joints, and in phase coordination (simultaneous ankle plantar-flexion, knee extension and hip extension). This pattern of coordination is illustrated in Fig. 1(a) in an ascending trial in which the pattern of coordination was maintained
Discussion
The observation of abrupt transitions between qualitatively different patterns of manual lifting provides a justification for the distinction which has traditionally been made between squat and stoop methods of lifting. Further, the identification of the change in phase relationships between the ankle and hip provides a means of solving the problem of how to define the techniques independent of the absolute range of motion of the knee (see Burgess-Limerick & Abernethy, 1997a).
The observation
References (22)
- et al.
Temporal coordination of human gait
- et al.
A biomechanical evaluation of five lifting techniques
Applied Ergonomics
(1986) - et al.
Qualitatively different modes of manual lifting
International Journal of Industrial Ergonomics
(1997) - et al.
Technical note: Relative phase quantifies interjoint coordination
Journal of Biomechanics
(1993) - et al.
Joint moments and muscle activity in the lower extremities and lower back in lifting and lowering tasks
Journal of Biomechanics
(1993) - et al.
Nonequilibrium phase transitions in coordinated biological motion: Critical fluctuations
Physics Letters A
(1986) - et al.
Coordination of the leg muscles in leglift and backlift
Journal of Biomechanics
(1992) - et al.
Stoop or squat: A review of biomechanical studies on lifting technique
Clinical Biomechanics
(1999) - et al.
Posturally induced transitions in rhythmic multijoint limb movements
Experimental Brain Research
(1993) - et al.
Towards a definition of manual lifting technique
Human Factors
(1997)
Self-selected manual lifting technique: Functional consequences of interjoint coordination
Human Factors
Cited by (21)
Evaluation and comparison of lift styles for an ideal lift among individuals with different levels of training
2019, Applied ErgonomicsCitation Excerpt :Participants from both conferences were pooled as they represented workers in North America. Squat and stoop lift styles are considered distinct lift styles (Burgess-Limerick et al., 2001) that represent two ends of a continuum with modified styles fitting somewhere in between. With squat lifting, the knees are in deep flexion with approximately 45° between the thigh and lower leg while the trunk remains at a near vertical orientation; this position is reversed in the stoop style with the knees being relatively straight (approximately 135° between the thigh and lower leg) and the trunk in a fully flexed position (Straker, 2003a,b).
Grasping an object at floor-level: Is movement strategy a matter of age?
2018, Applied ErgonomicsCitation Excerpt :Several studies have compared the respective benefits and advantages of these different techniques, and various factors have been found to promote use of one or the other. Some of these factors are linked to the object carried – such as its weight (Hoozemans et al., 2008), how cumbersome it is, the height from which it must be collected and that at which it must be deposited (Burgess-Limerick et al., 2001), the speed of transfer (Lin et al., 1999) or its trajectory (de Looze et al., 1998) – while others correspond to the functional capacities of the carrier – such as their gender (Lindbeck and Kjellberg, 2001; Plamondon et al., 2014b; Sheppard et al., 2016) or age (Van Dieën et al., 1994), the muscles solicited (Trafimow et al., 1993), the physiological cost (Kumar 1984), the oxygen consumption during the exercise (Hagen and Harms-Ringdahl, 1994) or how much experience the carrier has with transporting heavy loads (Plamondon et al., 2010, 2014a). This last point illustrates how expertise causes carriers to adapt their posture depending on the context, as determined by the object (weight, dimensions, fragility) or the trajectory to be covered between the initial point and the destination.
Motor adaptation capacity as a function of age in carrying out a repetitive assembly task at imposed work paces
2017, Applied ErgonomicsCitation Excerpt :However, these habits or changes were independent of the subjects’ back flexibility, height or weight. Fluctuations between squatting and stooping were also observed by Burgess-Limerick et al. (2001) in young subjects. For these authors, the transition from one technique to the other appeared to be part of a compromise between the biomechanical advantages of the two techniques, and the influence of lifting height on this compromise.
Multi-dimensional coordination in cross-country skiing analyzed using self-organizing maps
2014, Human Movement ScienceMoving beyond quiet stance: Applicability of the inverted pendulum model to stooping and crouching postures
2014, Journal of BiomechanicsArtificial neural networks for analyzing inter-limb coordination: The golf chip shot
2011, Human Movement ScienceCitation Excerpt :Therefore, treating the SOM trajectory as a collective variable, as described in coordination dynamics (see Kelso (1995) for a review), seems appropriate. Attractor diagrams have been used to describe the stability of coordination at specified control parameter values (Burgess-Limerick, Shemmell, Barry, Carson, & Abernethy, 2001; Haken et al., 1985) and theoretically for golf (Knight, 2004). Here, constructing an attractor diagram represents a second stage in the analysis.