Changes in intra-abdominal pressure,trunk muscle activation and force during isokinetic lifting and lowering

Intra-abdominal pressure (IAP), force and electromyographic (EMG) activity from the abdominal (intra-muscular) and trunk extensor (surface) muscles were measured in seven male subjects during maximal and sub-maximal sagittal lifting and lowering with straight arms and legs. An isokinetic dynamometer was used to provide five constant velocities (0.12–0.96 m·s^–1) of lifting (pulling against the resistance of the motor) and lowering (resisting the downward pull of the motor). For the maximal efforts, position-specific lowering force was greater than lifting force at each respective velocity. In contrast, corresponding IAPs during lowering were less than those during lifting. Highest mean force occurred during slow lowering (1547 N at 0.24 m·s^–1) while highest IAP occurred during the fastest lifts (17.8 kPa at 0.48–0.96 m·s^–1). Among the abdominal muscles, the highest level of activity and the best correlation to variations in IAP (r=0.970 over velocities) was demonstrated by the transversus abdominis muscle. At each velocity the EMG activity of the primary trunk and hip extensors was less during lowering (eccentric muscle action) than lifting (concentric muscle action) despite higher levels of force (r between –0.896 and –0.851). Sub-maximal efforts resulted in IAP increasing linearly with increasing lifting or lowering force (r=0.918 and 0.882, respectively). However, at any given force IAP was less during lowering than lifting. This difference was negated if force and IAP were expressed relative to their respective lifting and lowering maxima. It appears that the IAP increase primarily accomplished by the activation of the transversus abdominis muscle can have the dual function of stabilising the trunk and reducing compression forces in the lumbar spine via its extensor moment. The neural mechanisms involved in sensing and regulating both IAP and trunk extensor activity in relation to the type of muscle action, velocity and effort during the maximal and sub-maximal loading tasks are unknown.     

Qualitative discrepancies between trunk muscle activity and dynamic postural requirements at the initiation of reaching movements performed while sitting

Reaching movements are associated with widespread, nonfocal muscle activity. That activity is often assumed to play a postural role. We tested this assumption for the trunk muscles at the initiation of reaching movements with the following question. Does initial trunk muscle activity play a dynamic postural role by resisting the segmental interactive effects of the arm movement on the trunk? Seated subjects performed bilateral reaching movements while target direction was systematically varied. Muscle activity was recorded from flexors and extensors of the trunk and shoulder. Trunk muscle activity was compared with trunk torques calculated from simulations of reaching movements in which the trunk was modeled to stay still. Recorded trunk muscle activity was in qualitative agreement with torque predictions for only some target directions, suggesting that the nervous system does not activate trunk muscles across all target directions to counteract postural disturbances at the initiation of reaching movements.     

The influence of natural body sway on neuromuscular responses to an unpredictable surface translation

Previous research has shown that the postural configuration adopted by a subject, such as active leaning, influences the postural response to an unpredictable support surface translation. While those studies have examined large differences in postural conditions, it is of additional interest to examine the effects of naturally occurring changes in standing posture. Thus, it was hypothesized that the normal postural sway observed during quiet standing would affect the responses to an unpredictable support surface translation. Seventeen young adults stood quietly on a moveable platform and were perturbed in either the forward or backward direction when the location of the center of pressure (COP) was either 1.5 standard deviations anterior or posterior to the mean baseline COP signal. Postural responses, in the form of electromyographic (EMG) latencies and amplitudes, were recorded from lower limb and trunk muscles. When the location of the COP at the time of the translation was in the opposite, as compared to the same, direction as the upcoming translation, there was a significantly earlier onset of the antagonists (10-23%, i.e. 15-45 ms) and a greater EMG amplitude (14-39%) in four of the six recorded muscles. Stepping responses were most frequently observed during trials where the position of the COP was opposite to the direction of the translation. The results support the hypothesis that postural responses to unpredictable support surface translations are influenced by the normal movements of postural sway. The results may help to explain the large variability of postural responses found between past studies.     

Scaling of plantarflexor muscle activity and postural time-to-contact in response to upper-body perturbations in young and older adults

In this study, we describe and compare the compensatory responses of healthy young and older adults to sequentially increasing upper-body perturbations. The scaling of plantarflexor muscular activity and minimum time-to-contact (TtC_MIN) was examined, and we determined whether TtC_MIN predictions of instability (stepping transitions) for the older subjects were similar to those we previously reported for younger subjects (Hasson et al. in J Biomech 41:2121–2129, 2008). We found that the older subjects stepped at a lower perturbation level than the younger subjects; however, this response was appropriate based on their greater center of mass (CoM) accelerations, which may have been caused by differences in pre-perturbation states between the age groups. Although the CoM acceleration increased linearly with perturbation magnitude, the amount of gastrocnemius and soleus muscular activity increased nonlinearly in both age groups. There were no differences in the maximum plantarflexor torque responses, suggesting that the maximum torque capabilities of the older subjects were not limiting factors. As previously demonstrated in the younger subjects, the older subjects showed a quadratic decrease in TtC_MIN with increasing perturbation magnitude. The vertices of the quadratics gave accurate predictions of stepping transitions in both age groups, even though the older subjects stepped at lower perturbation magnitudes. By probing the postural system’s behavior through sequentially increasing upper-body perturbations, we observed a complementary nonlinear scaling of muscle activity and TtC_MIN, which suggests that subjects could use TtC or a correlate as an informational variable to help determine whether a step is necessary.

Neck,trunk and limb muscle responses during postural perturbations in humans

Summary This study examined the EMG onsets of leg, trunk, and neck muscles in 10 standing human subjects in response to support surface anterior and posterior translations, and to plantar and dorsiflexion rotations. The objective of the study was to test the hypothesis that the responses radiating upward from distal leg muscles represent part of a large ascending synergy encompassing axial muscles along the entire length of the body. If these responses are not ascending, then the muscles of the neck, and possibly the trunk, can be independently activated by vestibular, proprioceptive or visual inputs. We analysed the timing of postural muscle responses within and between body segments in order to determine whether they maintained a consistent temporal relationship under translational and rotational platform movement paradigms. Our results did not strongly support an ascending pattern of activation in all directions of platform perturbation. Temporal differences between activation patterns to platform perturbations in the forward or backward directions were revealed. In response to posterior platform translations we observed an ascending pattern of muscle responses along the extensor surface of the body. In addition, responses elicited in the neck flexor and abdominal muscles occurred as early as those of the stretched ankle muscles. This pattern of upward radiation from stretched ankle muscles was not as clear for anterior platform displacements, where early neck flexor muscle responses were observed during the ascending sequence on the flexor surface of the body. Platform rotations caused fewer responses in the neck and upper trunk muscles than translations, and all muscle responses occurred simultaneously rather than sequentially. Probable differences in the stimulation of vestibular and neck proprioceptive inputs and the mechanical demands of the rotation and translation paradigms are discussed.     

Correlation between intra-abdominal pressure and pulmonary volumes after superior and inferior abdominal surgery

OBJECTIVE:

Patients undergoing abdominal surgery are at risk for pulmonary complications. The principal cause of postoperative pulmonary complications is a significant reduction in pulmonary volumes (FEV₁ and FVC) to approximately 65-70% of the predicted value. Another frequent occurrence after abdominal surgery is increased intra-abdominal pressure. The aim of this study was to correlate changes in pulmonary volumes with the values of intra-abdominal pressure after abdominal surgery, according to the surgical incision in the abdomen (superior or inferior).

METHODS:

We prospectively evaluated 60 patients who underwent elective open abdominal surgery with a surgical time greater than 240 minutes. Patients were evaluated before surgery and on the 3^rd postoperative day. Spirometry was assessed by maximal respiratory maneuvers and flow-volume curves. Intra-abdominal pressure was measured in the postoperative period using the bladder technique.

RESULTS:

The mean age of the patients was 56±13 years, and 41.6% 25 were female; 50 patients (83.3%) had malignant disease. The patients were divided into two groups according to the surgical incision (superior or inferior). The lung volumes in the preoperative period showed no abnormalities. After surgery, there was a significant reduction in both FEV₁ (1.6±0.6 L) and FVC (2.0±0.7 L) with maintenance of FEV₁/FVC of 0.8±0.2 in both groups. The maximum intra-abdominal pressure values were similar (p = 0.59) for the two groups. There was no association between pulmonary volumes and intra-abdominal pressure measured in any of the groups analyzed.

CONCLUSIONS:

Our results show that superior and inferior abdominal surgery determines hypoventilation, unrelated to increased intra-abdominal pressure. Patients at high risk of pulmonary complications should receive respiratory care even if undergoing inferior abdominal surgery.     

Postural adjustments in sitting humans following external perturbations: muscle activity and kinematics

There are several controversies concerning the organization and induction of postural adjustments in standing humans. Some investigators suggest the responses are triggered by somatosensory inputs (especially from the ankle in standing subjects), while others emphasize the vestibular input induced by head acceleration. We examined postural responses in sitting subjects in order to describe the muscle activation pattern during various perturbations and to test whether somatosensory or vestibular stimulation elicited the responses. The kinematics and EMG patterns in respons to perturbations caused by movements of the support surface were studied in adults. The postural muscle activation following a backward sway was mainly the same, whether it was elicited by a forward translation or a legs-up rotation. This is remarkable, since, except for pelvis rotation, the movements of all body segments including the head differed in the two conditions. Furthermore, a second experiment showed that the direction of the initial head movement could be reversed with retainment of the same postural muscle activation pattern. The results suggest that somatosensory signals derived from the backward rotation of the pelvis, and not vestibular information from the head, trigger postural responses during sitting. There was a slight but consistent difference in the muscle activation pattern, whether the backward sway was elicited by a forward translation or legs-up rotation. The difference seemed to reflect the sensory information from head and other body parts (except the pelvis). This finding allowed us to speculate in a central pattern generator for postural adjustments containing two levels. At the first level, a simple format of the muscle activation would be generated; at the second level, the centrally generated pattern could be shaped and timed by interaction from the entire somatosensory, vestibular, and visual input.     

Trunk muscle reactions to sudden unexpected and expected perturbations in the absence of upright postural demand

The aim was to increase the understanding of the multifunctional role of the trunk muscles in spine control, particularly transversus abdominis (TrA). In 11 healthy males, intramuscular fine-wire electromyography (EMG) was obtained bilaterally from TrA, obliquus externus (OE), rectus abdominis (RA) and erector spinae (ES). The subjects lay on their right side on a horizontal swivel-table with immobilized pelvis and lower limbs and the trunk strapped to a movable platform. Unexpected or expected release of loads attached to the table by steel cables produced a perturbation inducing either trunk flexion or extension. The timing and the amplitude of activation of TrA were independent of direction of induced trunk movement. Furthermore, timing of TrA activation was simultaneous to or later than that of the more superficial abdominal muscles. Expectation of the perturbation caused a general shortening of onset latencies. The results indicate a direction independent function of TrA in lumbar spine control. Balancing the trunk vertically appears to add specific demands, since the recruitment of TrA in relation to the other abdominal muscles differed from earlier experiments in standing.     

The regulation of vestibular afferent information during monocular vision while standing

The purpose of this study was to investigate the contribution of the vestibular system to postural control during monocular vision using binaural-bipolar galvanic vestibular stimulation (GVS). Four visual (both eyes, dominant eye, non-dominant eye, and no vision) conditions were tested during GVS in five healthy subjects while focusing on a target placed in front of them. GVS evoked similar upper body postural sway during both monocular and no vision conditions that were significantly greater to those during binocular vision. Changes in ground reaction forces to the anode side followed that same trend, although data for vision with the dominant eye were not significantly different from that for binocular vision. These data suggest an increase in the weighting of vestibular afferent information during monocular vision for standing postural control.     

Anticipatory postural muscle activity associated with bilateral arm flexion while standing in individuals with spastic diplegic cerebral palsy: A pilot study

Compared to automatic postural responses to external perturbation, little is known about anticipatory postural adjustments in individuals with spastic diplegic cerebral palsy. In this study, we examined whether anticipatory activation of postural muscles would be observed before voluntary arm movement while standing in individuals with spastic diplegia. Seven individuals with spastic diplegia (SDCP_group, 12–22 years) and 7 age- and gender-matched individuals without disability (Control_group) participated in this study. Participants performed bilateral arm flexion at maximum speed at their own timing while standing, during which electromyographic (EMG) activities of focal and postural muscles were recorded. In both groups, the erector spinae (ES) and medial hamstring (MH) muscles were activated in advance of the anterior deltoid muscle (AD), which is a focal muscle of arm flexion. Although start times of ES and MH with respect to AD were similar in the 2 groups, increases in EMG amplitudes of ES and MH in the anticipatory range from −150 ms to +50 ms, with respect to burst onset of AD, were significantly smaller in the SDCP_group than in the Control_group. These findings suggest that individuals with spastic diplegia have the ability to anticipate the effects of disturbance of posture and equilibrium caused by arm movement and to activate postural muscles in advance of focal muscles. However, it is likely that the anticipatory increase in postural muscle activity is insufficient in individuals with spastic diplegia.     

Surface muscle pressure as a measure of active and passive behavior of muscles during gait

While surface electromyography (SEMG) can accurately register electrical activity of muscles during gait, there are no methods to estimate muscular force non-invasively. To better understand the mechanical behavior of muscle, we evaluated surface muscle pressure (SMP) in conjunction with SEMG. Changes in anterior thigh radial pressure during isometric contractions and gait were registered by pressure sensors on the limb. During isometric knee extensions by a single subject, SMP waveforms correlated well with SEMG (r=0.97), and SEMG onsets preceded those of SMP by 35-40 ms. SMP and SEMG signals were simultaneously recorded from the quadriceps of 10 healthy subjects during gait at speeds of 0.4, 0.8, 1.1, 1.4 and 2.2m/s. Muscle activity onset and cessation times were objectively determined for both modalities, and results showed high intra-class correlations. SMP waveforms were highly consistent from stride to stride, while SEMG waveforms varied widely. SEMG waveforms were typically brief, while SMP waveforms tended to be biphasic and outlasted the SEMG by approximately 40% of gait cycle at all speeds. These results are consistent with mechanical models of muscle, and demonstrate the use of SMP to estimate the timing of knee extensor muscle stiffness during gait.     

Changes in multi-segmented body movements and EMG activity while standing on firm and foam support surfaces

Postural control ensures stability during both static posture and locomotion by initiating corrective adjustments in body movement. This is particularly important when the conditions of the support surface change. We investigated the effects of standing on a compliant foam surface using 12 normal subjects (mean age 26 years) in terms of: linear movements at the head, shoulder, hip and knee; EMG activity of the tibialis anterior and gastrocnemius muscles and torques towards the support surface. As subjects repeated the trials with eyes open or closed, we were also able to determine the effects of vision on multi-segmented body movements during standing upon different support surface conditions. As expected, EMG activity, torque variance values and body movements at all measured positions increased significantly when standing on foam compared with the firm surface. Linear knee and hip movements increased more, relative to shoulder and head movements while standing on foam. Vision stabilized the head and shoulder movements more than hip and knee movements while standing on foam support surface. Moreover, vision significantly reduced the tibialis anterior EMG activity and torque variance during the trials involving foam. In conclusion, the foam support surface increased corrective muscle and torque activity, and changed the firm-surface multi-segmented body movement pattern. Vision improved the ability of postural control to handle compliant surface conditions. Several essential features of postural control have been found from recording movements from multiple points on the body, synchronized with recording torque and EMG.     

The effects of muscle fatigue on rapid finger oscillations

Two types of model for the control of rapid finger oscillations can be contrasted. According to the first type, any change of muscle characteristics induced by sustained isometric contractions should result in kinematic changes. According to the second type, kinematic characteristics should remain invariant while the timing of bursts of muscle activity exhibits compensatory adjustments. Following sustained isometric contractions of finger flexors and extensors, we observed essentially unchanged durations of the flexion and extension components of rapid finger oscillations, while the pattern of inter-burst intervals was modified so as to compensate the changed phasing of electromyographic bursts. These findings strongly suggest that even in extremely simple rapid finger oscillations motor control is configured to result in invariant kinematics. Electronic Publication     

Influence of lumbar spine rhythms and intra-abdominal pressure on spinal loads and trunk muscle forces during upper body inclination

Improved knowledge on spinal loads and trunk muscle forces may clarify the mechanical causes of various spinal diseases and has the potential to improve the current treatment options. Using an inverse dynamic musculoskeletal model, this sensitivity analysis was aimed to investigate the influence of lumbar spine rhythms and intra-abdominal pressure on the compressive and shear forces in L4-L5 disc and the trunk muscle forces during upper body inclination.Based on in vivo data, three different spine rhythms (SRs) were used along with alternative settings (with/without) of intra-abdominal pressure (IAP). Compressive and shear forces in L4-L5 disc as well as trunk muscle forces were predicted by inverse static simulations from standing upright to 55° of intermediate trunk inclination.Alternate model settings of intra-abdominal pressure and different spine rhythms resulted in significant variation of compression (763 N) and shear forces (195 N) in the L4-L5 disc and in global (454 N) and local (156 N) trunk muscle forces at maximum flexed position. During upper body inclination, the compression forces at L4-L5 disc were mostly released by IAP and increased for larger intervertebral rotation in a lumbar spine rhythm.This study demonstrated that with various possible assumptions of lumbar spine rhythm and intra-abdominal pressure, variation in predicted loads and muscles forces increase with larger flexion. It is therefore, essential to adapt these model parameters for accurate prediction of spinal loads and trunk muscle forces.     

背向行走足底压力分布特征对平衡能力的影响

目的研究背向行走过程中足底压力的变化,从而探究其对人体步态平衡能力的影响。方法使用足底压力测量系统(Pedar-X压力鞋垫)采集10名受试者在正向和背向两种行走模式下的足底压力数据。试验在跑步机上进行,以正向行走作为对照试验组。依据试验条件及预试验时受试者的直观感受,选取4个速度(2.0、2.5、3.0、3.5 km/h)分别进行两种行走模式下的试验,分析不同速度下足底压力中心轨迹、足底压力、足-地接触时间等参数的变化情况。结果不同速度的两种行走模式下,人在行走过程中的足底压力有所不同。背向行走过程中,足底压力中心从足前部向足跟过渡,从内侧向外侧过渡。相比正向行走,背向行走时的足底压力显著减小,行走过程中的足-地接触时间有所增加。结论研究背向行走的压力分布有助于更全面地理解人的动态平衡机制,并为行走稳定性的相关研究提供新视角。     

The effect of muscle length on motor-unit recruitment during isometric plantar flexion in humans

The triceps surae muscle group, consisting of the mono-articular soleus (SOL) and bi-articular gastrocnemius (GAS) muscles, primarily generates plantar flexor torque. Since the GAS muscle crosses the knee joint, flexion of the knee reduces the length of this muscle, thus limiting its contribution to torque output. However, it is not clearly understood how the central nervous system activates muscles that are at inefficient or non-optimal force-producing lengths. Therefore, the present study was designed to determine the effect of muscle length on motor-unit recruitment in the medial GAS muscle. Single motor-unit activity was recorded from the medial GAS muscle while electromyographic (EMG) activity was recorded from the SOL muscle in nine male subjects. With the ankle angle held constant at 90 degrees, the knee angle was changed from 180 degrees to 90 degrees, corresponding to a long and short GAS muscle length, respectively. Levels of voluntary plantar flexor torque were produced at a rate of 2 Nm.s-1 until motor-unit activity was detected. A total of 229 motor units were recorded, of which 121 and 108 were obtained at the long and short muscle lengths, respectively. At the short length, onset of motor-unit activity occurred at significantly higher levels of plantar flexor torque and SOL EMG. Onset of motor-unit activity occurred at 2.97 +/- 7.78 Nm and 32.14 +/- 10.25 Nm, corresponding to 0.045 +/- 0.075 mV and 0.231 +/- 0.129 mV of SOL EMG in the long and short positions, respectively. No individual GAS motor unit could be recorded at both muscle lengths. Motor units in the shortened GAS muscle may be influenced by peripheral afferents capable of reducing the excitability of the motoneurone pool. This may also reflect a specific inhibition of motor units having shortened, non-optimal fascicle lengths, and they are thereby incapable of contributing to plantar flexor torque.     

Shoulder and elbow muscle activity in goal-directed arm movements

 This research examined the electromyographic (EMG) activity of shoulder and elbow muscles during reaching movements of the upper limb. Subjects performed goal-directed arm movements in the horizontal plane. Movements which varied in amplitude, speed, and direction were performed in different sections of the workspace. EMG activity was recorded from the pectoralis major, posterior deltoid, biceps brachii short head, brachioradialis, triceps brachii long head, and triceps brachii lateral head; motion recordings were obtained with an optoelectric system. The analysis focused on the magnitude and timing of opposing muscle groups at the shoulder and elbow joints. For hand movements within any given direction of the workspace direction, kinematic manipulations changed agonist and antagonist EMG magnitude and intermuscle timing in a manner consistent with previous single-joint findings. To produce reaching movements in different directions and areas of the workspace, shoulder and elbow agonist EMG magnitude increased for those hand motions which required higher angular velocities, while the timing between opposing muscle groups at each joint was invariant. Received: 11 January 1996 / Accepted: 24 February 1997     

The influence of artificially increased hip and trunk stiffness on balance control in man

Lightweight corsets were used to produce mid-body stiffening, rendering the hip and trunk joints practically inflexible. To examine the effect of this artificially increased stiffness on balance control, we perturbed the upright stance of young subjects (20–34 years of age) while they wore one of two types of corset or no corset at all. One type, the half-corset, only increased hip stiffness, and the other, the full-corset, increased stiffness of the hips and trunk. The perturbations consisted of combined roll and pitch rotations of the support surface (7.5 deg, 60 deg/s) in one of six different directions. Outcome measures were biomechanical responses of the legs, trunk, arms and head, and electromyographic (EMG) responses from leg, trunk, and upper arm muscles. With the full-corset, a decrease in forward stabilising trunk pitch rotation compared to the no-corset condition occurred for backward pitch tilts of the support surface. In contrast, the half-corset condition yielded increased forward trunk motion. Trunk backward pitch motion after forwards support-surface perturbations was the same for all corset conditions. Ankle torques and lower leg angle changes in the pitch direction were decreased for both corset conditions for forward pitch tilts of the support-surface but unaltered for backward tilts. Changes in trunk roll motion with increased stiffness were profound. After onset of a roll support-surface perturbation, the trunk rolled in the opposite direction to the support-surface tilt for the no-corset and half-corset conditions, but in the same direction as the tilt for the full-corset condition. Initial head roll angular accelerations (at 100 ms) were larger for the full-corset condition but in the same direction (opposite platform tilt) for all conditions. Arm roll movements were initially in the same direction as trunk movements, and were followed by large compensatory arm movements only for the full-corset condition. Leg muscle (soleus, peroneus longus, but not tibialis anterior) balance-correcting responses were reduced for roll and pitch tilts under both corset conditions. Responses in paraspinals were also reduced. These results indicate that young healthy normals cannot rapidly modify movement strategies sufficiently to account for changes in link flexibility following increases in hip and trunk stiffness. The changes in leg and trunk muscle responses failed to achieve a normal roll or pitch trunk end position at 700 ms (except for forward tilt rotations), even though head accelerations and trunk joint proprioception seemed to provide information on changed trunk movement profiles over the first 300 ms following the perturbation. The major adaptation to stiffness involved increased use of arm movements to regain stability. The major differences in trunk motion for the no-corset, half-corset and full-corset conditions support the concept of a multi-link pendulum with different control dynamics in the pitch and roll planes as a model of human stance. Stiffening of the hip and trunk increases the likelihood of a loss of balance laterally and/or backwards. Thus, these results may have implications for the elderly and others, with and without disease states, who stiffen for a variety of reasons.     

Motor-unit activity in the trapezius muscle during rest, while inputting data, and during fast finger tapping

The Cinderella hypothesis postulates the continuous activity of specific motor units during low-level muscle contraction and contradicts the concept of motor-unit substitution. Constant trapezius muscle activity has been reported in typical visual-display-unit-related tasks. If it can be shown that constant muscle activity can be caused by the continuous firing of single motor units, this could explain the frequent complaints of muscular neck pain reported by computer users. The present study was undertaken to investigate motor-unit activity in the trapezius muscle during resting with closed eyes, while inputting three-digit numbers with auditory presentation at a rate of 0.5 Hz, and while tapping on a key with the right index finger at a rate of 5 Hz. Electrodes with four fine wires were inserted into the right upper trapezius muscle of six healthy subjects, and three-channel intramuscular electromyography was recorded. The decomposition programme MAPQuest, developed to analyse short-term one-channel signals, was complemented with MAPView, a programme that merges the short-term results of 10 s to a 3-min analysis. The results showed that activity in the trapezius muscle was induced in one subject while resting, in two subjects while inputting data, and in five subjects while finger tapping. Long-lasting single motor-unit firing was observed in two subjects while inputting data and in one subject while finger tapping. Whilst our findings may support the Cinderella hypothesis, the measurement periods are too short to confirm it fully, and for further discussion it is necessary to record and analyse for longer periods. Accepted: 19 June 2000