Eur J Appl Physiol (2012) 112:41634172 DOI 10.

1007/s00421-012-2403-z

ORIGINAL ARTICLE

Relevance of hand dominance to the bilateral decit phenomenon

Andrew Cornwell Nazareth Khodiguian Eun Jung Yoo

Received: 15 September 2011 / Accepted: 4 April 2012 / Published online: 25 April 2012 Springer-Verlag 2012

Abstract During maximal voluntary contractions, the sum of forces exerted by homonymous muscles when activated unilaterally (UL) is, typically, larger than the sum of forces when activated bilaterally (BL). This phenomenon is known as the bilateral decit (BLD). Our purpose was to determine if the dominant limb would be inhibited to a greater degree in the BL condition, thereby reducing any disparity in force output between the limbs. Maximum voluntary handgrip strength was measured in 40 left-handed and 40 right-handed individuals under both BL and UL conditions. The righthanders displayed 10.4 % greater right hand strength in both conditions; the left-handers exhibited 5.5 % greater left hand strength in the UL and 4.3 % in the BL condition. A BLD (-1.30 %) was present in the left-handed group only but a reduction in the force disparity between the hands was not evident. It was observed, however, that seven individuals from each group exhibited greater UL force with their nondominant hand. Accordingly, we re-analyzed the data after rearranging the groups based on unilateral hand grip strength dominance. A signicant reduction in force disparity between the hands occurred for the left-handed group only, the result of a signicant inhibition of the stronger left hand. A trend towards a similar reduction occurred for the right-

handers because of a signicant force reduction of the stronger right hand. Consequently, it appears that for maximum handgrip contractions, the BLD may be related to preferential inhibition of the stronger hand, especially for individuals who are left-hand-strength-dominant in terms of unilateral force output. Keywords Bilateral decit Hand dominance Handgrip strength

Introduction The bilateral decit (BLD) describes the occurrence of a reduction in maximal voluntary force output when homonymous muscles are activated bilaterally (simultaneously) compared to unilateral (independent) action of the same muscles. Although not evident in all studies (Hakkinen et al. 1995; Jakobi and Cafarelli 1998), this phenomenon has been demonstrated by the majority of investigations (Henry and Smith 1961; Howard and Enoka 1991; Kawakami et al. n 1998; Matkowski et al. 2011; Rejc et al. 2010; Van Diee et al. 2003). Consequently, the combined maximal force output from homonymous muscles in the bilateral condition is, typically, less than the combined maximal force output in the unilateral condition. For example, a 725 % reduction in bilateral force production compared to unilateral output has been reported for isometric knee extension and combined isometric hip and knee extension (Howard and Enoka n et al. 2003). Dynamic 1991; Koh et al. 1993; Van Diee contractions also display this phenomenon (Taniguchi 1998; Vandervoort et al. 1984, 1987), and a similar effect has been found for the upper limbs, although the decit is generally smaller compared to the lower limbs (for a review, see Jakobi and Chilibeck 2001).

Communicated by Dick F. Stegeman. A. Cornwell (&) N. Khodiguian School of Kinesiology and Nutritional Science, California State University, Los Angeles, 5151 State University Drive, Los Angeles, CA 90032, USA e-mail: acornwe@calstatela.edu E. J. Yoo College of Physical Education, Kyung Hee University, Seocheon-dong, Giheung-gu, Yongin-si, Gyeonggi-do 446-701, Korea

123

4164

Eur J Appl Physiol (2012) 112:41634172

At present, the underlying mechanism(s) responsible for the BLD is (are) not fully known. Nonetheless, it is generally accepted that some kind of neural inhibition prevails although the neural pathways responsible for such inhibition remain unclear (Jakobi and Chilibeck 2001). One possibility is that the BLD is caused by mutual inhibition of the two cerebral cortices via transcallosal bers. Oda and Moratani (1995b) investigated cortical involvement in the BLD by recording cortical activity via movement-related cortical potentials in subjects that performed maximal bilateral and unilateral handgrip contractions. In addition to lower force output and electrical activity of the exor digitorum supercialis muscle in the bilateral compared to unilateral condition, cortical activity was also depressed lending support to the notion that neural inhibition is partly responsible for the BLD and that the origin of inhibition lies within the motor cortex. Corroborating these ndings, Post et al. (2007) also found a reduction in cortical activity during maximal bilateral contractions of the index ngers. Our work (Khodiguian et al. 2003), however, has shown the BLD to be expressed during reexively evoked contractions, suggesting that a spinal mechanism may also be involved. We measured reexive force output and EMG activity of the quadriceps femoris muscle group after initiation of the myotatic patellar reex under unilateral and bilateral conditions. A BLD of 9.3 and 16.8 % was found for reexive force output and EMG activity, respectively. However, because similar ndings were not evident during maximal isometric knee extensions with the same participants, we could not predict the contribution of a spinal mechanism to the BLD under the condition of maximal voluntary activation. It is possible, though, that such a mechanism may be more inuential in dynamic movements rather than isometric conditions. When more complex movement patterns such as dynamic, multi-joint tasks have been studied, a BLD is more often prevalent (Owings and Grabiner 1998; Taniguchi 1998; Vandervoort et al. 1984; Van Soest et al. 1985). Activation of cross-inhibitory circuitry, therefore, whether at the level of the spinal cord or in the motor cortex, may be part of the underlying mechanism behind the BLD phenomenon. It is also not known whether this phenomenon has evolved for a specic purpose, or if it is just a by-product of an optimization process that has occurred to provide the most suitable neural circuitry to control the majority of human motion. Most human movement, such as locomotion, is reciprocal in nature, which involves inhibiting the homonymous muscle groups and stimulating the antagonist muscle groups of the contralateral limb. Consequently, during bilateral actions, which are not performed as frequently as reciprocal, unilateral actions, mutual neural inhibition of homonymous agonist muscles is a possibility if the inhibitory pathways are not fully silenced.

Alternatively, the BLD may be the result of a specic attempt by the nervous system to equate the maximal force output of each limb, as a signicant disparity in maximal force output between the limbs is not ideal during bilateral actions. Several studies have shown that the dominant limb is typically stronger than its non-dominant counterpart, with values ranging between 0 and 40 % (Clerke and Clerke 2001; Crosby et al. 1994; Kamarul et al. 2006), although the non-dominant limb may exhibit greater strength in some individuals (Clerke and Clerke 2001; Crosby et al. 1994; Incel et al. 2002). Such a dominance effect, however, has been shown to be more prevalent in right-handed individuals than in left-handers, with some studies showing no strength disparity between the hands for left-handers (Armstrong and Oldham 1999; Crosby et al. 1994; Mitsionis et al. 2009). Furthermore, it has been observed by several authors (Henry and Smith 1961; Oda and Moritani 1994, 1995a, b; Ohtsuki 1981) investigating the BLD phenomenon that a greater reduction in force was evident in the dominant limb, although this observation has generally being limited to actions involving the upper limbs. This inconsistency might be due to the differences in movement patterns of the upper and lower limbs during activities of daily living. Over the lifespan, a difference in the levels of physical activity exists between the upper and not the lower limbs (Jakobi and Chilibeck 2001), which may give rise to more frequent dominance effects in terms of strength for the upper limbs. When studying the upper limbs in terms of the BLD, investigations have, generally, recruited individuals who were right-side dominant, and the nding that the dominant side tends to display a greater reduction in force output has been an a posteriori observation rather than the result of tested a priori hypotheses. No study thus far has specically tested the relevance of limb dominance in regard to the BLD phenomenon and has not investigated separate groups that represent right- and left-side dominance. The purpose of this study, therefore, was to test the handgrip strength of two groups of subjects under bilateral and unilateral conditions. One group was left-side dominant, and the other right-side dominant. It was hypothesized that bilateral handgrip strength would be lower than that demonstrated under unilateral conditions, and that each group would display a greater force decrement in the dominant side. We further hypothesized that the dominant side would display greater unilateral strength and, consequently, the disparity in handgrip force output between the right and left hands would be smaller in the bilateral condition for both groups. Finally, because right-handers tend to display a greater strength difference between the dominant and nondominant hands for grip strength, we predicted that these outcomes would be of greater magnitude in the right-handed group.

123

Eur J Appl Physiol (2012) 112:41634172

4165

Methods Participants Forty left-handed (12 males, 28 females) and forty righthanded (19 males, 21 females) individuals volunteered to participate in this study (for subject characteristics see Table 1). Hand dominance was ascertained by asking the participants which hand they use to perform well-learned skills such as writing and throwing. If anyone had previously switched from predominantly using their naturally preferred hand for whatever reason, either as a child or during adulthood, they were declined further participation. The participants were healthy and had not suffered any recent injuries to the upper extremity which might have inhibited their ability to produce a maximum voluntary contraction (MVC) of the nger exors. Furthermore, as bilateral resistance training has been shown to diminish a BLD (Taniguchi 1998), none of the participants were engaged in, or had recently undergone, a habitual rigorous strength-training regimen. Finally, all participants were naive to the hypotheses of the study and, after reading a consent form that outlined the procedures and any potential risks, informed consent was provided by each individual prior to the testing session. The consent form and procedures were approved by the appropriate institutional human subjects review committee. Experimental protocol Maximum voluntary handgrip strength was measured under both bilateral and unilateral conditions, together with the electrical activity of the exor digitorum supercialis muscle via associated electromyogram (EMG) recordings. EMG data allowed us to determine if any force decit was associated with a decrease in neural activation. After placement of the EMG electrodes, a warm-up set of ten sub-maximal contractions at approximately 75 % of perceived maximum effort was rst conducted. Two minutes following the completion of the warm-up, a MVC accommodation trial was performed under bilateral conditions. The participants then performed multiple maximum effort handgrip contractions, each trial lasting for a period of 3 s. Three bilateral trials (BL) and three unilateral trials (UL) for each hand were performed, making a total of six contractions for each limb. The trials were divided into
Table 1 Subject characteristics

three sets, with one set comprising of three trialseither, right hand only, left hand only, or both hands simultaneously. The order was predetermined for a particular subject and was selected in a random, counterbalanced fashion. A 1-min recovery period was allowed between trials and 3 min between sets to prevent fatigue. The peak force for each trial and associated integrated EMG was determined and the results for each hand were averaged over the three trials for each condition. Measurement of maximal force output Unilateral and bilateral maximal isometric nger exion force output (handgrip strength) was measured using electronic hand dynamometers (Model SS25LA, Biopac Systems, Inc., Santa Barbara, CA). To collect and analyze the raw signal, an analog software package [Ariel performance analysis system (APAS), Ariel Dynamics, Trabuco Canyon, CA] was used to sample the data at 1,000 Hz for a period of 5 s. The participants performed the trials whilst seated with their upper arms hanging naturally beside their torso and their elbows exed to a 90 angle so that their forearms were parallel with the ground and their hands in a semipronated/supinated position. In the unilateral condition, the non-participating arm was rested on the ipsilateral thigh and kept relaxed throughout the trial. The participants looked directly ahead focusing on an object that was strategically placed to achieve a head and neck posture as dened by the anatomical position. They were instructed to generate an MVC as rapidly as possible on the verbal command of Ready? Go. Verbal encouragement was given during each MVC, which was maintained for approximately 3 s, but no performance feedback was given. Measurement of electrical activity Electrical activity of the exor digitorum supercialis muscle in each forearm was monitored during each contraction by placing EMG electrodes over the belly of the muscle on the surface of the skin. The electrodes were positioned according to Zipp (1982). An ink mark was made distal to the medial epicondyle at a distance equivalent to one quarter of the length measured between the medial epicondyle and the skin fold at the wrist to denote

Number of right-handers Males Females 19 21

Number of left-handers 12 28

Age (years) 32.1 12.7 28.2 9.1

Height (m) 1.76 0.08 1.61 0.06

Weight (kg) 83.5 14.9 61.6 12.1

123

4166

Eur J Appl Physiol (2012) 112:41634172

the location of the electrode center. The muscle belly was then palpated whilst the ngers were exed against a resistance to ne tune the nal electrode position. The electrodes were bipolar surface electrodes (Delsys, Boston, MA), consisting of two parallel silver bars, each 10 mm long, 1.0 mm wide, with a center-to-center inter-electrode distance of 10 mm. The bony patella was used to attach the reference ground electrode. Before placing the electrodes, the skin was shaved if necessary, cleansed with acetone, and abraded to reduce impedance between the skin and electrode. The raw signal was pre-amplied (1,000 gain) at a xed bandwidth of 20450 Hz, full-wave rectied, and integrated (iEMG) to assess the activity of the exor digitorum supercialis during each MVC. Data were sampled at 1,000 Hz for a period of 5 s using the same equipment that was used to sample the force data (APAS system). Data reduction The APAS software was used to determine the maximum force generated during each trial and the associated iEMG which indicated the degree of muscle activation during the time of maximal force generation. An iEMG was obtained by integrating the area of the EMG-time curve over a 500 ms window which spanned 250 ms either side of the point in time when peak force was attained. The rate of force generation (RFG) was also calculated by measuring the slope of the force/time curve (DF/Dt) between the time points corresponding to 10 and 50 % of the peak force. This part of the force trace was chosen as it typically represented the most linear portion of the graph that denoted a rise in force. The initial part of the trace was quite curvilinear, possibly due to the initial compression of the soft tissues of the hand as force was applied to the dynamometer. The muscle activity between the time points used to calculate the RFG was also assessed by integrating the EMG signal over this period. An iEMG value per unit time was then obtained by dividing this value by the time duration of the period. This was necessary to normalize for duration of activity as the time between 10 and 50 % of peak force varied from trial to trial. Bilateral indices Bilateral indices for peak force (BIF), iEMG associated with peak force (BIEMG1), rate of force generation (BIRFG), and iEMG associated with rate of force generation (BIEMG2) were calculated to express any relative difference in force and EMG output between unilateral and bilateral conditions. The calculations were performed according to Howard and Enoka (1991) outlined as follows:

BI % 100 right bilateral left bilateral right unilateral left unilateral 100: A divergence of the BI from zero demonstrates that there is a difference between the unilateral and bilateral conditions. A BI which is \0, or negative, indicates that the bilateral performance is less than the combined unilateral performance This indicates a BLD. A BI which is [0, or positive, is indicative of a greater bilateral than combined unilateral performance. This indicates a bilateral facilitation. Analysis of results A paired sample t test was used for the bilateral-to-unilateral comparison of means and for dominant to nondominant limb comparisons within a condition. A single sample t test was used to test if the mean bilateral indices were signicantly different from 0. Statistical signicance was accepted at an alpha level of 0.05.

Results A bilateral force decit was not evident in the right-handed group; however, the left-handed group displayed a small but signicant decit (-1.30 0.46 %; P \ 0.01) (Table 2). The associated iEMG values were not concomitant. A signicant bilateral facilitation (BIEMG1 = ?3.50 1.16 %; P \ 0.01) resulted for the right-handed individuals, and no signicant difference was found for the left-handers (BIEMG1 = ?1.97 1.36 %; P = 0.08). When considering each limb individually, no signicant force decits for either right or left hand were found for the right-handed group. However, for iEMG values, both hands displayed a signicant (P \ 0.05) bilateral facilitation. For the left-handed group, only the left hand displayed a signicant (P \ 0.01) reduction in force in the bilateral compared to the unilateral condition, although this was not matched by any signicant decit in iEMG. A signicant (P \ 0.01) increase in iEMG, however, was found for the right hand. For both groups under both unilateral and bilateral conditions, the dominant hand displayed the greatest force compared to the non-dominant hand. In the right-handed group, the right hand was 10.4 2.1 % stronger (P \ 0.001) in the bilateral condition and also 10.4 2.1 % stronger (P \ 0.001) in the unilateral condition. In the lefthanded group, the left hand was 4.3 1.6 % stronger (P \ 0.05) in the bilateral condition and 5.5 1.5 % stronger (P \ 0.005) in the unilateral condition. These strength disparities between the hands were not signicantly different between the conditions for either group.

123

Eur J Appl Physiol (2012) 112:41634172

4167

Table 2 Peak forces and associated iEMG values for both right- and left-handed groups during the performance of maximal voluntary unilateral and bilateral handgrip contractions Right-handers Bilateral RH force (N) LH force (N) RH iEMG (lV s) LH iEMG (lV s) BIF (%) BIEMG1 (%) 355.9 17.5e 327.3 17.1 59.7 6.3 52.7 5.4r -0.17 0.46 ?3.50 1.16
r

Left-handers Unilateral 357.6 17.9e 327.0 16.9 58.3 6.2 50.3 4.8 Bilateral 293.0 14.9 302.8 14.7,e 51.0 5.2 56.5 6.1
r

Unilateral 295.3 15.1 307.8 14.5e 48.9 5.1 57.3 6.4

-1.30 0.46* ?1.97 1.36

Bilateral indices for force (BIF) and EMG (BIEMG) are also presented Values are mean standard error of the mean iEMG integrated electromyogram, RH right hand, LH left hand
r

Value in bilateral condition is signicantly smaller (P \ 0.05) than the value in unilateral condition Value in bilateral condition is signicantly greater (P \ 0.05) than the value in unilateral condition Value is signicantly greater than 0 (P \ 0.01)

* Value is signicantly smaller than 0 (P \ 0.01)

e

Dominant hand value is signicantly greater than the non-dominant hand value (P \ 0.05)

Table 3 Rate of force generation and associated iEMG values for both right- and left-handed groups during the performance of maximal voluntary unilateral and bilateral handgrip contractions Right-handers Bilateral RH RFG (N/s) LH RFG (N/s) RH iEMG (lV) LH iEMG (lV) BIRFG (%) BIEMG2 (%) 1,905.1 126.4 1,684.3 116.7 104.5 12.3 82.4 7.0 -3.48 1.57* -2.18 2.21
,e

Left-handers Unilateral 1,992.6 127.0 1,688.3 106.0 107.9 13.1 86.7 7.8
e

Bilateral 1,528.9 117.8 1,606.8 122.1e 87.0 9.4 92.1 8.8 -2.70 1.56* -4.64 1.98*

Unilateral 1,548.8 111.5 1,618.8 109.3e 91.0 8.8 95.5 8.4

Bilateral indices for rate of force generation (BIRFG), and associated EMG (BIEMG2) are also presented Values are mean standard error of the mean RFG rate of force generation, iEMG integrated electromyogram, RH right hand, LH left hand

Value in bilateral condition is signicantly smaller (P \ 0.05) than the value in unilateral condition Dominant hand value is signicantly greater than the non-dominant hand value (P \ 0.05)

* Value is signicantly smaller than 0 (P \ 0.05)

e

Table 3 shows similar information for the rate at which force was generated (RFG). Both groups demonstrated a signicant BLD for RFG; a BIRFG of -3.48 1.57 % and -2.70 1.56 % (P \ 0.05) was calculated for the right- and left-handers, respectively. However, only the left-handed group showed a similar signicant decit (-4.64 1.98 %; P \ 0.05) for iEMG (BIEMG2). When analyzing each limb independently, only the right hand of the right-handed group demonstrated a signicant (P \ 0.05) reduction in RFG but no concomitant reduction for iEMG was realized. For the left-handers, neither hand displayed a signicant decline in RFG or iEMG.

Under both unilateral and bilateral conditions, a significantly greater RFG value resulted for the dominant limb regardless of the group. In the right-handed group, the RFG for the right hand was 15.7 2.7 % higher (P \ 0.001) than for the left hand in the bilateral condition, and 20.6 5.0 % (P \ 0.001) higher in the unilateral condition. This disparity in RFG between the hands in the unilateral condition was signicantly (P \ 0.05) greater than the disparity in the bilateral condition. In the left-handed group, the RFG for the left hand was 6.5 2.9 % higher (P \ 0.05) than the right hand in the bilateral condition, and 7.4 2.8 % (P \ 0.05) higher in the unilateral

123

4168

Eur J Appl Physiol (2012) 112:41634172

condition. However, these disparities were not signicantly different between the unilateral and bilateral conditions. After examining the individual strength differences between each hand for each participant, it was observed that seven right-handed individuals actually exhibited a greater unilateral force in their non-dominant left hand, and an equal number of left-handed individuals displayed a greater unilateral force in their non-dominant right hand. Because we theorized that a BLD may arise because of a need to balance the force output between each limb when contracting homonymous limbs simultaneously, and that this would be achieved by preferentially inhibiting the stronger limb, we performed further analysis after rearranging the groups based on unilateral hand grip strength. Thus, the seven right-handed participants who had a stronger left hand were placed in the left-handed group, and the seven left-handed individuals who had a stronger right hand were placed in the right-handed group. Hence, we had a right-hand-strength-dominant group and a lefthand-strength-dominant group. The results are shown below in Table 4. For the right-hand-strength-dominant group, the right hand was signicantly (P \ 0.001) stronger than the left by 13.4 1.9 % in the unilateral condition compared to 12.0 2.0 % (P \ 0.001) in the bilateral condition. For the left-hand-strength-dominant group, the left hand was signicantly (P \ 0.001) stronger than the right hand by 8.2 1.0 % in the unilateral condition, whereas the difference was only 5.9 1.5 % (P \ 0.001) in the bilateral condition. Unlike the ndings using the original grouping (hand dominance determined by hand preference to perform common skills), the force disparity between the hands

was signicantly (P \ 0.001) smaller in the bilateral condition compared to the unilateral condition for the left-hand-strength-dominant group. This was due to a signicant (P \ 0.01) reduction in force output from the dominant left hand as the peak force developed by the right hand remained unchanged. A signicant (P \ 0.01) BLD of -1.0 0.4 % was also calculated for this group. For the right-hand-strength-dominant group, there was a signicant (P \ 0.05) reduction in force output of the dominant right hand but not the non-dominant left in the bilateral compared to the unilateral condition. However, the resulting smaller force disparity between the hands in the bilateral condition was not signicant (P = 0.06). Similarly, a resulting BLD of -0.5 % was not signicant. For the iEMG data associated with the peak force readings, rearranging the groups in terms of unilateral force dominance changed the results for both groups. For the righthand-strength-dominant group, only a trend towards a signicant bilateral facilitation (BIEMG1 = ?2.6 1.3 %; P = 0.05) occurred instead of a signicant nding as with the original grouping. Also, no signicant increase in iEMG was found for each limb. For the left-hand-strength-dominant group, a signicant bilateral facilitation occurred (BIEMG1 = ?3.3 1.3 %; P \ 0.01). This was due to a signicant increase in iEMG of the right hand only. Analysis of the data with respect to RFG yielded no new information.

Discussion One of the primary goals of this investigation was to determine if handgrip strength is reduced more in the

Table 4 Peak forces and associated iEMG values for both right- and left-hand-strength-dominant groups based on unilateral hand grip strength Right-hand-strength-dominant Bilateral RH force (N) LH force (N) RH iEMG (lV s) LH iEMG (lV s) BIF (%) BIEMG1 (%) 357.5 18.3,e 322.0 17.0 59.6 7.1 58.0 6.1 -0.5 0.5 ?2.6 1.3 Unilateral 361.2 18.4e 321.3 17.0 58.6 7.0 56.5 6.5 Left-hand-strength-dominant Bilateral 291.4 14.0 308.1 14.9,e 50.0 3.4 50.6 4.6 -1.0 0.4* ?3.3 1.3
r

Unilateral 291.7 14.3 313.5 14.4e 47.5 3.8 50.3 4.6

Bilateral indices for force (BIF) and EMG (BIEMG) are also presented Values are mean standard error of the mean iEMG integrated electromyogram, RH right hand, LH left hand
r

Value in bilateral condition is signicantly smaller (P \ 0.05) than the value in unilateral condition Value in bilateral condition is signicantly greater (P \ 0.05) than the value in unilateral condition Value is signicantly greater than 0 (P \ 0.01)

* Value is signicantly smaller than 0 (P \ 0.01)

e

Dominant hand value is signicantly greater than the non-dominant hand value (P \ 0.01)

123

Eur J Appl Physiol (2012) 112:41634172

4169

dominant hand than the non-dominant hand when performing MVCs under bilateral conditions as compared to performing similar contractions under unilateral conditions. It was reasoned that the dominant side would display greater unilateral strength, and should a BLD be evident, a greater reduction of force in the stronger, dominant hand would occur to reduce the disparity in handgrip force output between the two hands. It was also theorized that such an outcome would be of greater magnitude in the right-handed group because right-handers tend to display a greater grip strength difference between the dominant and non-dominant hands (Crosby et al. 1994; Mitsionis et al. 2009; Kamarul et al. 2006). Such a nding may be due to the fact that, in general, equipment such as tools and appliances are designed for right-handed individuals, thereby resulting in left-handed people performing more physical activity with their non-dominant hand than righthanders (Crosby et al. 1994). Our results indicate that a signicant BLD (-1.3 %) occurred in the left-handed group only, and that when considering each limb independently, only the left hand of the left-handed group displayed a signicant reduction in force output. Thus, the hypothesis that the dominant hand would display a greater amount of inhibition under bilateral conditions was realized for the left-handed group only. This was in spite of the left-handers demonstrating the least force disparity between the hands under unilateral conditions. As hypothesized, both groups displayed a stronger dominant hand under unilateral conditions, and this strength difference was more evident in the right-handed group (10.4 vs. 5.5 % for right- and left-handed groups, respectively). However, this did not lead to the righthanded individuals experiencing a greater bilateral force decrement of the dominant hand in relation to the nondominant hand than the left-handed participants. In fact, neither hand of the right-handers was subject to signicant bilateral inhibition. Furthermore, despite the left-handers demonstrating a small reduction in force of the dominant hand, the amount of inhibition was not sufcient to realize a signicant reduction in the force disparity between the two hands in the bilateral as compared to the unilateral condition. Consequently, our initial ndings do not fully support the notion that the BLD is the result of a neural strategy employed by the nervous system to minimize the force disparity between the hands when performing maximum voluntary bilateral contractions. One confounding factor, however, was that not all the participants were stronger in their dominant hand. Analysis of the force outputs of the hands revealed that an equal number of seven (17.5 %) individuals from each group exhibited a greater unilateral force with their non-dominant hand. Similar ndings have been reported by Incel et al. (2002) who observed that 10.9 % of right-handed

individuals (n = 128) and 33.3 % of left-handed individuals (n = 21) were stronger in their non-dominant side when performing maximum handgrip contractions. Consequently, we decided to further analyze the data after rearranging the groups based on unilateral hand grip strength dominance. Thus, the seven right-handed participants who had a stronger left hand were placed in the lefthanded group, and the seven left-handed individuals who had a stronger right hand were placed in the right-handed group. Hence, we had a right-hand-strength-dominant group and a left-hand-strength-dominant group. As expected, the unilateral strength difference between the hands increased compared to the original grouping (13.4 vs. 10.4 % for the right-handed group, and 8.2 vs. 5.5 % for the left-handed group). Moreover, a signicant reduction (from 8.2 to 5.9 %) in this force disparity between the hands became evident for the left-handstrength-dominant group under the bilateral compared to the unilateral condition, and a trend towards a signicant reduction (from 13.4 to 12.0 %; P = 0.06) also materialized for the right-hand-strength-dominant group. This smaller force disparity in the left-hand-strength- dominant group was due to a signicant reduction in force output of the stronger left hand as no signicant inhibition was noted for the right hand. This resulted in a signicant BLD of -1.0 %. For the right-handers, a signicant drop in force occurred for the stronger right hand, but not the left. However, as mentioned above, this was not sufcient to create a signicant reduction in the force disparity between the hands or a signicant BLD. Thus, it appears that when hand dominance is based on unilateral force capability, more compelling evidence in support of our hypotheses is forthcoming. Consequently, future research might involve a similar study but with a greater number of participants and group selection based on unilateral strength dominance rather than the typical methods of assessing hand dominance such as hand preference for performing ne motor skills (e.g., writing and throwing). A greater number of participants may also elucidate if left-hand-strengthdominant individuals are truly more susceptible to bilateral inhibition than right-hand-strength-dominant individuals as indicated by the current ndings. It has been suggested that the underlying mechanism for such bilateral inhibition is neural in origin as several studies (Koh et al. 1993; Oda and Moritani 1994, 1995b; n et al. 2003) have found a concomitant reduction Van Diee in electrical activity during bilateral contractions. Our iEMG data, however, do not corroborate our ndings for force output regardless of how the groups were arranged. Thus, no evidence was obtained to suggest that any of the decits found were the result of neural inhibition. In fact, for the right-handed group, unexpectedly, a bilateral facilitation for iEMG was evident (BIEMG1 = ?3.5%).

123

4170

Eur J Appl Physiol (2012) 112:41634172

Analyzing each limb independently, both hands displayed an increase in iEMG. These results are surprising as no concomitant bilateral facilitation for force was found, and as mentioned, several studies (Koh et al. 1993; Oda and n et al. 2003) have shown Moritani 1994, 1995b; Van Diee a BLD with regard to EMG activity. It is possible that in the present study, co-contraction or increased co-contraction of the digit extensor muscles in the bilateral as compared to the unilateral condition occurred, thereby countering the force generated by the exor muscles. Several studies have shown that during maximal voluntary contractions, simultaneous activation of agonist and antagonist muscles can occur (Bazzuchi et al. 2006, 2008; Carolan and Cafarelli 1992). For example, Bazzuchi et al. (2006) demonstrated that both the elbow exors and extensors display antagonistic activity during maximal isokinetic elbow extension and exion contractions, respectively. The advantage of such co-contraction activity is increased joint stability (Bazzuchi et al. 2006; Kornecki 1992). Although plausible, the existence of co-contraction in the present study must remain speculative as we did not collect EMG data from the extensor muscles of the ngers. If, however, such co-contraction was present, it is unclear why it would occur, or occur to a greater extent, in the bilateral compared to the unilateral condition. Some studies kkinen et al. 1998, 2001) (Carolan and Cafarelli 1992; Ha have shown that one of the early neural adaptations to a resistance training program is that the antagonist muscles reduce their level of activation; that is, co-contraction is minimized. Perhaps the level of initial co-contraction in untrained subjects is related to the total effort involved which would increase when more muscles are required to be activated to perform the task. As the bilateral condition required maximal activation of both left and right nger exor muscle groups simultaneously, more overall effort was demanded from our untrained participants, which may have led to greater co-contraction of the extensor muscles because of the need for greater stabilization. We acknowledge, however, that cross-talk from neighboring muscles may have contaminated the recordings from the exor digitorum supercialis muscle. We did not have the resources to develop a spatial lter via double differentiation of the signal and, therefore, could not attenuate such effects. Because the forearm contains many slender muscles that often overlap, the potential for crosstalk is high. For example, the wrist exors, comprising of the exor carpi radialis, exor carpi unlaris and the palmaris longus, all lie in close proximity to the exor digitorum supercilalis. However, as the same type of contraction was performed under both the bilateral and unilateral conditions, it could be argued that the relative proportion of cross-talk would remain fairly consistent across trials and conditions. Since we were comparing the

relative difference in electrical activity between unilateral and bilateral conditions, such cross-talk may have been mathematically cancelled out after applying the BLD calculation. Still, we recognize that by not attempting to control for cross-talk, a major limitation with respect to the EMG data exists. Our ndings in terms of the magnitude of bilateral force decit are also in contrast to several studies that have investigated maximal grip strength differences between bilateral and unilateral conditions. Post et al. (2007), Oda n et al. and Moritani (1995b), Ohtsuki (1981), and Van Diee (2003) all demonstrated a greater magnitude of bilateral force decit with means ranging from -5 to -22 %. Ohtsuki (1981) also measured the contribution of each individual nger to the overall grip strength. Interestingly, for the males, the ranked order from the strongest to the weakest nger (middle nger, followed by the ring, index, and little nger) corresponded exactly to the order of each ngers contribution (from largest to smallest) to the decrease in strength in the bilateral condition. In other words, the strongest nger suffered the greatest inhibition. This nding is in line with the notion that the hand displaying the greatest grip strength as a whole might be inhibited more when a BLD in force is evident. As mentioned, our study demonstrated this to be true for the lefthanded group only but the amount of decit was quite smallonly -1.3 % whereas Ohtsuki (1981) in the abovementioned study found decits up to -13.6 %. Even n et al. higher decits have been observed by Van Diee (2003) who found a mean decit of -22 % in the right hand and -20 % in the left hand of 10 right-handed individuals. Our results for the left-handers are more in line with those of Henry and Smith (1961) and Morehouse et al. (2000) who observed relatively small BLDs of -3 and -1.9 to -5.7 %, respectively, using relatively large numbers of participants. Differences in the type of device used to measure force output, instructions given to the participants on how to exert the contractions, the number of trials, the length of the rest intervals, the posture maintained by the participants, and the number of participants are some of the potential reasons that might account for the wide range of bilateral force decits found. In the present study, however, further supporting evidence was forthcoming when we investigated how quickly the force was developed, or the RFG. One of the key instructions given to our participants was to contract and reach peak force as rapidly as possible on hearing the command to start. This produced a relatively steep rise in force as displayed by the force/time trace. From the rise portion of the force trace, the RFG was calculated as outlined in the methods. We found that both groups displayed a signicant BLD for RFG (-3.48 and -2.70 % for the

123

Eur J Appl Physiol (2012) 112:41634172

4171

right- and left-handers, respectively), and that the dominant hand was inhibited more than the non-dominant hand in the right-handed group only. In fact, only the dominant right hand displayed a signicant force reduction in the righthanded group. In the left-handed group, while the combined force outputs of both hands led to a signicant BLD, the individual force decits realized for each hand were not signicant when analyzed independently. Furthermore, under both unilateral and bilateral conditions, a signicantly greater RFG value resulted for the dominant limb regardless of the group. However, this disparity in RFG between the hands in the unilateral condition was signicantly greater than the disparity in the bilateral condition for the right-handed group only. Consequently, it appears that the outcome for RFG is in contrast with the results for peak force: that is, the right-handed group displays results that are more consistent with our hypotheses when considering RFG, but when considering peak force, the results for the left-handed group tend to support our hypotheses more. The reasons for this are unclear. Other studies have also shown a BLD to occur during the rise in force period prior to reaching peak force (Koh n et al. 2003). et al. 1993; Oda and Moritani 1994; Van Diee n et al. Interestingly, both Koh et al. (1993) and Van Diee (2003) found larger BLDs for RFG than for peak force when participants performed maximal isometric knee extensions. In the present study, a BLD for RFG was found for both right- and left-handed groups, whereas only the left-handed group demonstrated a BLD for peak force. It is possible, therefore, that the inhibition of force output under bilateral conditions may be more prevalent during the rising phase of the force development if the contraction is elicited as quickly as possible. In line with this premise, Vandervoort et al. (1984, 1987) found that greater BLDs were evident with increasing isokinetic bench press (Vandervoort et al. 1987) and leg press (Vandervoort et al. 1984) velocities and speculated that this was due to the preferential inhibition of fast twitch motor units. Since then, several researchers, including ourselves, have found supporting evidence for this notion (Koh et al. 1993; Oda and Moritani 1994; Kawakami et al. 1998; Khodiguian et al. 2003) but further work needs to be conducted to strengthen such speculations. In conclusion, we have shown that the dominant hand is typically stronger than the non-dominant hand and that this unilateral strength difference is greater in right-handed individuals. The greater unilateral force disparity in righthanders, however, did not translate to a greater inhibition of the dominant hand compared to left-handers when performing maximum bilateral contractions. In fact, only the left-handers displayed a small but signicant BLD for maximum force output, the result of a force reduction in the dominant left hand only. This decit was not sufcient

to cause a signicant reduction in the strength inequality between the hands in the bilateral compared to the unilateral condition. It was observed, however, that several of the participants from each group were actually stronger in their non-dominant hand. After rearranging the groups based on unilateral strength dominance, a signicant reduction in the strength difference between the hands in the bilateral condition did result for the left-hand-strength-dominantgroup, and this was due to a decrease in force output from the left hand as opposed to the right. Furthermore, a trend towards a similar reduction was found for the right-handstrength-dominant group, which was the result of a decrease in force output from the right hand rather than the left. Consequently, it appears that for maximum handgrip contractions, the BLD may be related to preferential inhibition of the stronger hand, especially for individuals who are left-hand-strength-dominant in terms of unilateral force output.
Conict of interest The authors declare no conict of interest.

References
Armstrong CA, Oldham JA (1999) A comparison of dominant and non-dominant hand strengths. J Hand Surg 24(4):421425 Bazzuchi I, Riccio ME, Felici F (2008) Tennis players show a lower coactivation of the elbow antagonist muscles during isokinetic exercises. J Electromyogr Kinesiol 18:752759 Bazzuchi I, Sbriccoli P, Marzattinocci G, Felici F (2006) Coactivation of the elbow antagonist muscles is not affected by the speed of movement in isokinetic exercise. Muscle Nerve 33:191199 Carolan B, Cafarelli E (1992) Adaptations in coactivation after isometric resistance training. J Appl Physiol 73:911917 Clerke A, Clerke J (2001) A literature review of the effect of handedness on isometric grip strength differences of the left and right hands. Am J Occup Ther 55:206211 MA, Mawr B (1994) Hand strength: normative Crosby CA, Wehbe values. J Hand Surg 19(4):665670 kkinen K, Pastinen UM, Karsikas R, Linnamo V (1995) NeuroHa muscular performance in voluntary bilateral and unilateral contraction and during electrical stimulation in men at different ages. Eur J Appl Physiol 70:518527 kkinen K, Kallinen M, Izquierdo M et al (1998) Changes in Ha agonist-antagonist EMG, muscle CSA and force during strength training in middle-aged and older people. J Appl Physiol 84:13411349 kkinen K, Kraemer WJ, Newton RU, Alen M (2001) Changes in Ha electromyographic activity, muscle ber and force production characteristics during heavy resistance/power strength training in middle-aged and older men and women. Acta Physiol Scand 171(1):5162 Henry FM, Smith LE (1961) Simultaneous vs. separate bilateral muscular contractions in relation to neural overow theory and neuromotor specicity. Res Q Exerc Sport 32:4246 Howard JD, Enoka RM (1991) Maximum bilateral contractions are modied by neurally mediated interlimb effects. J Appl Physiol 70:306316 Incel NA, Ceceli E, Durukan PB, Erdem HR, Yorgancioglu ZR (2002) Grip strength: effect of hand dominance. Singap Med J 43(5):234237

123

4172 Jakobi JM, Cafarelli E (1998) Neuromuscular drive and force production are not altered during bilateral contractions. J Appl Physiol 84:200206 Jakobi JM, Chilibeck PD (2001) Bilateral and unilateral contractions: possible differences in maximal voluntary force. Can J Appl Physiol 26:1233 Kamarul T, Loh WYC, Ahmad TS (2006) Normal hand grip strength in the Malaysian population. J Orthop Surg 14:172177 Kawakami Y, Sale DG, MacDougall JD, Moroz JS (1998) Bilateral decit in plantar exion: relation to knee joint position, muscle activation, and reex excitability. Eur J Appl Physiol 77:212216 Khodiguian N, Cornwell A, Lares E, DiCaprio PA, Hawkins SA (2003) Expression of the bilateral decit during reexively evoked contractions. J Appl Physiol 94:171178 Koh TJ, Grabiner MD, Clough CA (1993) Bilateral decit is larger for step than for ramp isometric contractions. J Appl Physiol 74:12001205 Kornecki S (1992) Mechanism of muscular stabilization process in joints. J Biomech 25(3):235245 Matkowski B, Martin A, Lepers R (2011) Comparison of maximal unilateral versus bilateral voluntary contraction force. Eur J Appl Physiol 111(8):15711578 Mitsionis G, Pakos EE, Stalas KS, Paschos N, Papakostas T, Beris AE (2009) Normative data on handgrip strength in a Greek adult population. Int Orthop 33(3):713717 Morehouse SA, Szeligo F, DiTommaso E (2000) Characteristics of the bimanual decit using grip strength. Laterality 5(2):167185 Oda S, Moritani T (1994) Maximal isometric force and neural activity during bilateral and unilateral elbow exion in humans. Eur J Appl Physiol 69:240243 Oda S, Moritani T (1995a) Cross-correlation of bilateral differences in fatigue during sustained maximal voluntary contraction. Eur J Appl Physiol 70:305310

Eur J Appl Physiol (2012) 112:41634172 Oda S, Moritani T (1995b) Movement-related cortical potentials during handgrip contractions with special reference to force and electromyogram bilateral decit. Eur J Appl Physiol 72:15 Ohtsuki T (1981) Decrease in grip strength induced by simultaneous bilateral exertion with reference to nger strength. Ergonomics 24:3748 Owings TM, Grabiner MD (1998) Fatigue effects on the bilateral decit are speed dependent. Med Sci Sports Exerc 30:12571262 Post M, van Duinen H, Steens A, Renken R, Kuipers B, Maurits N, Zijdewind I (2007) Reduced cortical activity during maximal bilateral contractions of the index nger. Neuroimage 35(1):1627 Rejc E, Lazzer S, Antonutto G, Isola M, di Prampero PE (2010) Bilateral decit and EMG activity during explosive lower limb contractions against different overloads. Eur J Appl Physiol 108(1):157165 Taniguchi Y (1998) Relationship between the modications of bilateral decit in upper and lower limbs by resistance training in humans. Eur J Appl Physiol 78:226230 Vandervoort AA, Sale DG, Moroz JR (1984) Comparison of motor unit activation during unilateral and bilateral leg extension. J Appl Physiol 56:4651 Vandervoort AA, Sale DG, Moroz JR (1987) Strength-velocity relationship and fatigability of unilateral versus bilateral arm extension. Eur J Appl Physiol 56:201205 n JH, Ogita F, De Haan A (2003) Reduced Neural drive in Van Diee bilateral exertions: a performance limiting factor? Med Sci Sports Exerc 35(1):111118 Van Soest AJ, Roebroeck ME, Robbert MJ, Huling PA, Van Ingen Schenau GJ (1985) A comparison of one-legged and two-legged countermovement jumps. Med Sci Sports Exerc 17:635639 Zipp P (1982) Recommendations for the standardization of lead positions in surface electromyography. Eur J Appl Physiol 50:4144

123