- Email: [email protected]
Effect of mental imagery of a motor task on the Hoffmann reflex
Behavioural Brain Research 142 (2003) 81–87
Effect of mental imagery of a motor task on the Hoffmann reflex B.S. Hale∗ , J.S. Raglin, D.M. Koceja Department of Kinesiology, Indiana University, HPER 112, Bloomington, IN 47404, USA Received 16 February 2002; received in revised form 8 November 2002; accepted 8 November 2002
Abstract Research has found that mental imagery of a motor task may influence the Hoffmann reflex (H-reflex). However, this work has not examined the potential influence of background EMG (BEMG) on the H-reflex. In this study 23 adult participants (M = 23.3 years, S.D. = 3.2) were instructed to mentally image plantar flexion of the right foot at four intensities: 40, 60, 80 and 100% of maximum voluntary contraction (MVC) after completing practice trials of actual contractions at these intensities. Dependent measures were the BEMG activity and the peak-to-peak amplitude of the H-reflex. On each trial the peak-to-peak amplitude of the soleus H-reflex was measured in the right leg and averaged, BEMG (40 ms bin) was measured in the soleus and tibialis anterior of both legs. Following trials of plantar flexion at the target intensities participants completed 5 imagery trials at each intensity and 15 trials while performing this motor task. Five resting control trials were administered prior to and following the completion of all test trials. Administration of test trials was randomized within conditions. A main effect (P < 0.05) for trial blocks was observed for H-reflex amplitude but not BEMG. The H-reflex increased linearly throughout testing, suggesting that the H-reflex was modified by the practice of imagery rather than the intensity of the imagined task. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Hoffmann reflex; Motor imagery; Background EMG
1. Introduction Some research indicates physiological responses to imagined movement are similar to the actual performance of motor activity [4,9,11,12], with the degree of physiological activity corresponding with the effort of the imagined task [10]. Recently, researchers have examined Hoffmann reflex (H-reflex) responses during imagery of motor tasks. The results of these studies have been equivocal [6,8,9]. Hashimoto and Rothwell [6] reported no change in the H-reflex amplitude while subjects were imaging repetitive wrist flexion and extension. However, they also reported greater EMG responses in the flexor muscle while the subject was imaging flexion, with larger EMG responses in the extensor muscle while the subject was imaging extension. They concluded “mental rehearsal could dynamically modulate the excitability of the motor cortex without evoking descending volleys to the spinal cord” (p. 79). In contrast, Jeannerod [8] reported a small elevation in H-reflex amplitude and a greater increase in tendon reflexes in subjects performing imagery which was attributed to increased motor neuron excitability. The increase in motor neuron excitability suggests the presence ∗
Corresponding author. Tel.: +1-812-856-4590; fax: +1-812-855-6778. E-mail address: [email protected] (B.S. Hale).
of a descending volley of activity. A depression in H-reflex amplitude during imagery of speed skating was reported by Oishi et al. [9]. Oishi et al. suggested that the depression in H-reflex during mental imagery is a result of a descending neural mechanism which reduces motor neuron excitability. Inconsistencies in the previous findings may be influenced in part by BEMG activity, as it is known that with an elevation in BEMG, H-reflex amplitude also increases [1]. Unfortunately BEMG was not assessed in two [8,9] of these studies, and there remains a need to determine if H-reflex responses to mental imagery are mediated by BEMG activity. The current study was conducted to compare the Hreflex amplitude and BEMG activity between an actual motor movement and mental imagery of the same motor movement.
2. Materials and methods 2.1. Participants The study was initiated following approval by the local institutional review board. Prior to the initiation of testing all participants completed general information questionnaires and human subjects consent forms. Participants were
0166-4328/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0166-4328(02)00397-2
82
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
23 student volunteers (10 males and 13 females; mean age = 23.3 years, S.D. = 3.2 years), enrolled at Indiana University. Prior to testing participants were screened for previous right leg injuries or surgeries and for any prior history of neurological disorder. 2.2. Methodology Participants were tested on actual performance and mental imagery involving plantar flexion of the right foot. Surface electrodes (Therapeutics Unlimited, active electrodes), with a 2-cm interelectrode distance, were placed longitudinally over the right and left soleus muscles, approximately 2 cm below the point where the two heads of the gastrocnemius join the Achilles tendon. Additionally, surface electrodes were placed over the right and left tibialis anterior muscles. Ground electrodes were positioned over the right and left ankles of the participants. Recording and stimulating electrode placement was unchanged throughout testing. A transcutaneous electrical stimulus (1-ms pulse duration; Grass Instruments, Model S88), the Grass Stimulator was connected in series to a constant current unit (Grass Instruments, Model CCU 1 A) and a stimulus isolation unit (Grass Instruments, Model SIU 5), was delivered to the posterior tibial nerve in the popliteal fossa to elicit the H-reflex.
Soleus H-wave recruitment curves were determined for each participant utilizing procedures outlined by Hugon [7]. The current was increased by 2 mA increments from zero until a maximum motor response was obtained. Testing was done at 20% of maximum M-wave amplitude. On each trial, the current was monitored with a current probe, to ensure stability of the current delivery. The peak-to-peak amplitude of the H-reflex was used as the dependent variable. Data were sampled and digitized on-line with a microcomputer (sample rate = 2 kHz), which allowed for the measurement of the BEMG activity (for entire trial approximately 7 s, the BEMG used for baseline analysis was sampled for the 40 ms immediately prior to the administration of the electrical stimulation) and the peak-to-peak amplitudes of the corresponding EMG waves. Independent variables were mental imaging, motor task performance, and percentage of contraction. Sample traces from each condition (rest, actual contraction, and imagined contraction) are presented in Fig. 1. Maximum voluntary contraction was determined as participants plantar flexed their right foot on a footplate connected to a Cybex (Type II) dynamometer and was determined using feedback from a Tektronix (TDS 310) oscilloscope. Participants performed three maximum contractions and the trial with the greatest voluntary force was used to calculate the percentages for the sub-maximal trials.
Fig. 1. Sample 150 ms traces from participants in the current study at each of the experimental conditions. The samples are shown from 50 ms immediately prior to the presentation of the experimental stimulation to 50 ms following the H-reflex response. The peak-to-peak amplitude of the H-reflex was measured.
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
Each participant was tested on 1 day under all conditions. The testing conditions were rest, imagery trials at 40, 60, 80, 100% maximum voluntary contraction (MVC), actual motor performance trials at 40, 60, and 80% MVC, and rest. The participants were tested on each condition and the trials were randomized within condition. Participants were in a seated position with their eyes closed and their knees and hips flexed at 90◦ angles with their ankles in a neutral (non-flexed) position for the duration of the testing. The right foot of each participant was secured to a footplate during testing and their left foot was resting on an attached footplate. Individual H-reflex and BEMG data from trials were converted into means for the five trials at each condition. Instructions were recorded and played to the participants for all imagery trials at the onset of each block of the specific percentage of MVC. See sample imagery script. Soon I would like you to imagine that you are making a contraction of your right soleus at (40, 60, 80% or maximal) of your maximal possible effort. In other words, the imagined effort should be exactly (40, 60, 80% or maximal) of the hardest contraction you could make. While forming the image recall what it felt like to contract your soleus at (40, 60, 80% or maximal) of maximal effort. You should imagine what it felt like to make the contraction. Close your eyes and form the image of (40, 60, 80% or maximal) of maximal contraction. While forming the image keep your leg still. Now begin imagining that you are contracting your soleus at (40, 60, 80% or maximal) of your maximal possible effort. Nod your head when you have formed this image. Hold this image until you feel the stimulation. Previous literature indicates that responses to internal imagery are more closely associated with actual exercise than are the responses to external imagery [10]. Therefore, the imagery script was written to encourage participants to perform mental imagery using the internal perspective. Upon completion of the imagery trials participants rated the vividness of the images formed on a 5-point Likert scale, and this rating was an overall assessment of all imagery trials. 2.3. Data analysis Initially, a 2 × 3 (condition by percentage of MVC) ANOVA was performed to determine H-reflex differences between mental imaging and actual motor performance and the various levels of maximum voluntary contraction. To address the specificity of imaging an additional series of 2 × 4 (involved versus non-involved and percentage of MVC) ANOVAs were performed to analyze BEMG responses to testing. To examine the influence of trial block order one-way ANOVAs were performed analyzing trial block order results. A Dunnett’s post hoc test was performed on the data for cases in which significant effects were observed. Further analysis for effect size was performed to measure the magnitude of changes using standardized mean
83
differences effect size equation [3]. All statistical tests were performed using the 0.05 level of significance.
3. Results The participants rated the vividness of the images formed upon completion of imagery trials. The range of possible responses on the vividness scale were from 1 (very vivid) to 5 (not at all vivid). The mean vividness score was 2.4 (S.D. = 0.6) corresponding to a verbal rating of “vivid”. No participants reported vividness ratings greater than three (“somewhat vivid”). 3.1. H-reflex and soleus BEMG responses for the actual and imagery conditions An ANOVA was performed to examine if H-reflex amplitude differed from rest during the experimental trials (40, 60, 80, and maximal) involving either actual or imagined contractions. A significant (F(3,132) = 3.39, P < 0.05) main effect was found for the actual condition. Post hoc analysis indicated that H-reflex amplitude was elevated above rest in all of the actual contraction conditions, this result was expected. However, there was no main effect for imagery conditions (Fig. 2). The effect size for H-reflex for the actual and imagery trials by intensity from 40, 60, 80% and maximal were 0.37, 0.26, 0.39 and 0.17, 0.10, 0.11, 0.13, respectively. The ANOVA for involved soleus BEMG responses indicated a significant (F(3,132) = 94.97, P < 0.05) main effect for the actual condition. Post hoc analysis indicated that BEMG responses were elevated above the rest condition in all of the actual contraction conditions, this elevation is consistent with previous research and was anticipated. A significant main effect was not observed in the imagery condition (F(3,132) = 0.01, P = 0.99) (Fig. 3). The effect size for soleus BEMG for the actual and imagery trials by intensity from 40, 60, 80% and maximal were 2.21, 2.47, 2.35 and 0.22, 0.22, 0.44, 0.53, respectively. The ANOVA for non-involved soleus BEMG responses also revealed differences from rest within each condition (i.e. actual, imagery). A significant (F(3,132) = 17.23, P < 0.05) main effect was found for the actual condition. Post hoc analysis indicated that BEMG responses were elevated above rest in all of the actual contraction conditions. A significant main effect was not observed in the imagery condition (F(3,132) = 0.32, P = 0.81). BEMG activity was also assessed in the tibialis anterior of the involved and non-involved legs. An ANOVA indicated that there were no changes in BEMG activity of this muscle during imagery. However, a significant (F(3,132) = 47.98, P < 0.05) main effect was found for the actual condition. Post hoc analysis indicated that BEMG responses were elevated above rest in all of the actual contraction conditions (Fig. 4).
84
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
Fig. 2. Soleus H-reflex activity during actual and imagined contractions across percentages of MVC. The mean H-reflex responses are presented as they relate to the mean maximum motor response, as the H/M ratio, as well as standard errors for both conditions.
Fig. 3. Soleus BEMG activity in the involved and non-involved (contralateral) legs during imagery trials. The mean BEMG data are presented as well as the measure of standard error.
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
85
Fig. 4. Tibialis anterior BEMG activity in the involved and non-involved (contralateral) legs during imagery trials. The mean BEMG data are presented as well as the measure of standard error.
3.2. H-reflex and BEMG responses by trial order The H-reflex amplitude values in the imagery conditions were analyzed for a trials effect and the ANOVA revealed a significant (F(4,88) = 3.36, P < 0.05) main effect. H-reflex amplitude generally increased from the rest condition as the imagery trials progressed. Post hoc analysis indicated this increase reached significance in the final two blocks of imagery trials (Fig. 5). The effect size for H-reflex by trials
by trial block from block 1 to block 4 were 0.04, 0.07, 0.21 and 0.25, respectively. A significant main effect (F(4,88) = 2.67, P < 0.05) was observed for the soleus in the involved leg. Post hoc analysis indicated significant increases from rest in soleus BEMG of the involved leg during the first and third blocks of trials (Fig. 6). The effect size for soleus BEMG by trials by trial block from block 1 to block 4 were 0.44, 0.22, 0.44 and 0.22, respectively.
Fig. 5. H-reflex activity in the soleus across imagery trial blocks. The mean H-reflex responses are presented as they relate to the maximum motor response, as the H/M ratio, as well as the standard error.
86
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
Fig. 6. Soleus BEMG activity in the involved and non-involved (contralateral) legs across imagery blocks. The mean BEMG and standard error data are presented.
Changes in tibialis anterior BEMG activity for the involved leg were not observed (F(4,88) = 1.96, P > 0.05). A significant main effect (F(4,88) = 2.98, P < 0.05) was observed for the tibialis anterior in the non-involved leg. Post hoc analysis indicated all conditions were significantly elevated over rest for trial blocks.
4. Discussion The present study was conducted to examine the consequences of mental imagery of a simple motor task on the H-reflex and BEMG. One finding was the lack of a significant H-reflex response or BEMG changes during imagery of plantar flexion at differing intensities. According to the psychoneuromuscular theory of mental imagery [11], increased physiological activity should occur with imagined movement. Further, the physiological changes should correspond to the intensity of the imagined effort, although at a lower magnitude than would be found with actual movement [4,10]. However, as noted by Williams et al. [11] “fully convincing support for this theory remains elusive” (p. 284). Specifically, it was found that H-reflex amplitude and soleus BEMG activity were each unchanged from baseline during imagery trials. Additionally, there was no evident dose–response relationship with H-reflex responses or BEMG with increased intensities of imagined effort. As
expected during the actual contraction conditions, significant increases in both H-reflex amplitude and BEMG activity were observed in both the tibialis anterior and the soleus. The changes in soleus BEMG of the involved leg increased linearly with elevations in the intensity of contraction. Although H-reflex did not change significantly during any of the imagined intensity conditions, a significant main effect was found when the data were analyzed by trial order, independent of intensity. The primary finding was an observed increase in H-reflex amplitude that reached significance in the third and final (i.e. fourth) block of trials. The pattern of soleus BEMG activity for trials was different, with elevations above rest occurring in the first and third trial blocks. Thus, H-reflex responses to imagery were independent of BEMG, a factor known to influence H-reflex [1]. Thus the present H-reflex changes appear to be independent of BEMG activity. Although changes in BEMG were noted future research might want to only sample imagery trials that show no change in BEMG. Previous research involving H-reflex and imagery has yielded differing results. Oishi et al. [9] observed that elite speed skaters, who were experienced mental imagers, decreased the amplitude of the H-reflex during imagery of a speed skating event. The elevation in H-reflex amplitude across imagery trial blocks observed in this study is consistent with prior findings by Jeannerod [8] in which mental practice was associated with an increase in motor neuron excitability and motor unit recruitment. In contrast, Bon-
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
net et al. (unpublished, [8]) found only a small increase in H-reflex in subjects who imaged weak or strong presses on a foot pedal. They reported the increase in T-reflexes, an indicator of gamma motor neuron activity, in imagery of a strong or weak press on a foot pedal. The elevation in T-reflex was found to be only slightly less than what they observed during actual movement. Based on this finding the authors contend that imagery had a greater effect on the gamma-afferent activity and research will be needed to test this hypothesis. While the present investigation did not test for differences in H-reflex response to the motor imagery task based on sex, future research investigating these responses should include an analysis based on sex. Barton [2] reported no differences between females and males on a performance-based assessment of imagery effectiveness. Whereas, Epstein [5] reported differences in dart throwing performance based on sex. Males were reported to perform better than females after imagery training. The equivocal nature of these findings suggests that future research should continue to assess the difference of imagery based on sex. The participants in this study were given some practice in the activity prior to the imagery conditions. It is possible that larger changes in H-reflex amplitude could have occurred with additional physical practice prior to imagery, but this is unlikely given the simple nature of the task used in this study. Increases in H-reflex amplitude observed for trial order during imagery trials may also have resulted from operant conditioning, or from the stress of repeated electrical stimulation elevating the H-reflex. However, H-reflex amplitude did not differ from the pre-test baseline value, indicating that the changes in H-reflex amplitude were not a simple habituation or practice effect. Thus it appears that H-reflex amplitude during imagery is modulated by practice independent of the percent of contraction imagined. Based on current findings future research investigating the effects of motor imagery on the H-reflex should control for trial effects. The current data indicates that the modulations observed in H-reflex are a function of practice.
87
In summary, the present study found that H-reflex and BEMG activity during imagery was unaffected by the effort of the imagined task. However, H-reflex activity increased across trials. Significant changes in BEMG also occurred, however, based upon the difference in magnitude of these responses it is suggested that these changes are unrelated to the H-reflex changes during imagery trials.
References [1] Angulo-Kinzler R, Mynark R, Koceja D. Soleus H-reflex gain in elderly and young adults: modulation due to body position. J Gerontol 1998;53:M120–5. [2] Barton K. The effect of mental imagery on sport climbing performance of college students. Microform Publications, International Institute for Sport and Human Performance, University of Oregon, 1997. [3] Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum; 1988. [4] Decety J, Jeannerod M, Germain M, Pastene J. Vegetative response during imagined movement is proportional to mental effort. Behav Brain Res 1991;42:1–5. [5] Epstein ML. Relationship of mental imagery and mental rehearsal to performance of a motor task. J Sport Psychol 1980;2:211–20. [6] Hashimoto R, Rothwell JC. Dynamic changes in cortico spinal excitability during motor imagery. Exp Brain Res 1999;125:75–81. [7] Hugon M. Methodology of the Hoffmann reflex in man. New developments in electromyography and clinical neurophysiology 1973;3:277–93. [8] Jeannerod M. Mental imagery in the motor context. Neuropsychologia 1995;33:1419–32. [9] Oishi K, Kimura M, Yasukawa M, Yoneda T, Maeshima T. Amplitude reduction of H-reflex during mental movement simulation in elite athletes. Behav Brain Res 1994;62:55–61. [10] Wang Y, Morgan W. The effect of imagery perspectives on the psychophysiological responses to imagined exercise. Behav Brain Res 1992;52:167–74. [11] Williams J, Rippon G, Stone B, Annett J. Psychophysiological correlates of dynamic imagery. Br J Psychol 1995;86:283–300. [12] Yue G, Cole K. Strength increases from the motor program: comparison of training with maximal voluntary and imagined muscle contractions. J Neurophysiol 1992;67:1114–23.
Effect of mental imagery of a motor task on the Hoffmann reflex B.S. Hale∗ , J.S. Raglin, D.M. Koceja Department of Kinesiology, Indiana University, HPER 112, Bloomington, IN 47404, USA Received 16 February 2002; received in revised form 8 November 2002; accepted 8 November 2002
Abstract Research has found that mental imagery of a motor task may influence the Hoffmann reflex (H-reflex). However, this work has not examined the potential influence of background EMG (BEMG) on the H-reflex. In this study 23 adult participants (M = 23.3 years, S.D. = 3.2) were instructed to mentally image plantar flexion of the right foot at four intensities: 40, 60, 80 and 100% of maximum voluntary contraction (MVC) after completing practice trials of actual contractions at these intensities. Dependent measures were the BEMG activity and the peak-to-peak amplitude of the H-reflex. On each trial the peak-to-peak amplitude of the soleus H-reflex was measured in the right leg and averaged, BEMG (40 ms bin) was measured in the soleus and tibialis anterior of both legs. Following trials of plantar flexion at the target intensities participants completed 5 imagery trials at each intensity and 15 trials while performing this motor task. Five resting control trials were administered prior to and following the completion of all test trials. Administration of test trials was randomized within conditions. A main effect (P < 0.05) for trial blocks was observed for H-reflex amplitude but not BEMG. The H-reflex increased linearly throughout testing, suggesting that the H-reflex was modified by the practice of imagery rather than the intensity of the imagined task. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Hoffmann reflex; Motor imagery; Background EMG
1. Introduction Some research indicates physiological responses to imagined movement are similar to the actual performance of motor activity [4,9,11,12], with the degree of physiological activity corresponding with the effort of the imagined task [10]. Recently, researchers have examined Hoffmann reflex (H-reflex) responses during imagery of motor tasks. The results of these studies have been equivocal [6,8,9]. Hashimoto and Rothwell [6] reported no change in the H-reflex amplitude while subjects were imaging repetitive wrist flexion and extension. However, they also reported greater EMG responses in the flexor muscle while the subject was imaging flexion, with larger EMG responses in the extensor muscle while the subject was imaging extension. They concluded “mental rehearsal could dynamically modulate the excitability of the motor cortex without evoking descending volleys to the spinal cord” (p. 79). In contrast, Jeannerod [8] reported a small elevation in H-reflex amplitude and a greater increase in tendon reflexes in subjects performing imagery which was attributed to increased motor neuron excitability. The increase in motor neuron excitability suggests the presence ∗
Corresponding author. Tel.: +1-812-856-4590; fax: +1-812-855-6778. E-mail address: [email protected] (B.S. Hale).
of a descending volley of activity. A depression in H-reflex amplitude during imagery of speed skating was reported by Oishi et al. [9]. Oishi et al. suggested that the depression in H-reflex during mental imagery is a result of a descending neural mechanism which reduces motor neuron excitability. Inconsistencies in the previous findings may be influenced in part by BEMG activity, as it is known that with an elevation in BEMG, H-reflex amplitude also increases [1]. Unfortunately BEMG was not assessed in two [8,9] of these studies, and there remains a need to determine if H-reflex responses to mental imagery are mediated by BEMG activity. The current study was conducted to compare the Hreflex amplitude and BEMG activity between an actual motor movement and mental imagery of the same motor movement.
2. Materials and methods 2.1. Participants The study was initiated following approval by the local institutional review board. Prior to the initiation of testing all participants completed general information questionnaires and human subjects consent forms. Participants were
0166-4328/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S0166-4328(02)00397-2
82
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
23 student volunteers (10 males and 13 females; mean age = 23.3 years, S.D. = 3.2 years), enrolled at Indiana University. Prior to testing participants were screened for previous right leg injuries or surgeries and for any prior history of neurological disorder. 2.2. Methodology Participants were tested on actual performance and mental imagery involving plantar flexion of the right foot. Surface electrodes (Therapeutics Unlimited, active electrodes), with a 2-cm interelectrode distance, were placed longitudinally over the right and left soleus muscles, approximately 2 cm below the point where the two heads of the gastrocnemius join the Achilles tendon. Additionally, surface electrodes were placed over the right and left tibialis anterior muscles. Ground electrodes were positioned over the right and left ankles of the participants. Recording and stimulating electrode placement was unchanged throughout testing. A transcutaneous electrical stimulus (1-ms pulse duration; Grass Instruments, Model S88), the Grass Stimulator was connected in series to a constant current unit (Grass Instruments, Model CCU 1 A) and a stimulus isolation unit (Grass Instruments, Model SIU 5), was delivered to the posterior tibial nerve in the popliteal fossa to elicit the H-reflex.
Soleus H-wave recruitment curves were determined for each participant utilizing procedures outlined by Hugon [7]. The current was increased by 2 mA increments from zero until a maximum motor response was obtained. Testing was done at 20% of maximum M-wave amplitude. On each trial, the current was monitored with a current probe, to ensure stability of the current delivery. The peak-to-peak amplitude of the H-reflex was used as the dependent variable. Data were sampled and digitized on-line with a microcomputer (sample rate = 2 kHz), which allowed for the measurement of the BEMG activity (for entire trial approximately 7 s, the BEMG used for baseline analysis was sampled for the 40 ms immediately prior to the administration of the electrical stimulation) and the peak-to-peak amplitudes of the corresponding EMG waves. Independent variables were mental imaging, motor task performance, and percentage of contraction. Sample traces from each condition (rest, actual contraction, and imagined contraction) are presented in Fig. 1. Maximum voluntary contraction was determined as participants plantar flexed their right foot on a footplate connected to a Cybex (Type II) dynamometer and was determined using feedback from a Tektronix (TDS 310) oscilloscope. Participants performed three maximum contractions and the trial with the greatest voluntary force was used to calculate the percentages for the sub-maximal trials.
Fig. 1. Sample 150 ms traces from participants in the current study at each of the experimental conditions. The samples are shown from 50 ms immediately prior to the presentation of the experimental stimulation to 50 ms following the H-reflex response. The peak-to-peak amplitude of the H-reflex was measured.
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
Each participant was tested on 1 day under all conditions. The testing conditions were rest, imagery trials at 40, 60, 80, 100% maximum voluntary contraction (MVC), actual motor performance trials at 40, 60, and 80% MVC, and rest. The participants were tested on each condition and the trials were randomized within condition. Participants were in a seated position with their eyes closed and their knees and hips flexed at 90◦ angles with their ankles in a neutral (non-flexed) position for the duration of the testing. The right foot of each participant was secured to a footplate during testing and their left foot was resting on an attached footplate. Individual H-reflex and BEMG data from trials were converted into means for the five trials at each condition. Instructions were recorded and played to the participants for all imagery trials at the onset of each block of the specific percentage of MVC. See sample imagery script. Soon I would like you to imagine that you are making a contraction of your right soleus at (40, 60, 80% or maximal) of your maximal possible effort. In other words, the imagined effort should be exactly (40, 60, 80% or maximal) of the hardest contraction you could make. While forming the image recall what it felt like to contract your soleus at (40, 60, 80% or maximal) of maximal effort. You should imagine what it felt like to make the contraction. Close your eyes and form the image of (40, 60, 80% or maximal) of maximal contraction. While forming the image keep your leg still. Now begin imagining that you are contracting your soleus at (40, 60, 80% or maximal) of your maximal possible effort. Nod your head when you have formed this image. Hold this image until you feel the stimulation. Previous literature indicates that responses to internal imagery are more closely associated with actual exercise than are the responses to external imagery [10]. Therefore, the imagery script was written to encourage participants to perform mental imagery using the internal perspective. Upon completion of the imagery trials participants rated the vividness of the images formed on a 5-point Likert scale, and this rating was an overall assessment of all imagery trials. 2.3. Data analysis Initially, a 2 × 3 (condition by percentage of MVC) ANOVA was performed to determine H-reflex differences between mental imaging and actual motor performance and the various levels of maximum voluntary contraction. To address the specificity of imaging an additional series of 2 × 4 (involved versus non-involved and percentage of MVC) ANOVAs were performed to analyze BEMG responses to testing. To examine the influence of trial block order one-way ANOVAs were performed analyzing trial block order results. A Dunnett’s post hoc test was performed on the data for cases in which significant effects were observed. Further analysis for effect size was performed to measure the magnitude of changes using standardized mean
83
differences effect size equation [3]. All statistical tests were performed using the 0.05 level of significance.
3. Results The participants rated the vividness of the images formed upon completion of imagery trials. The range of possible responses on the vividness scale were from 1 (very vivid) to 5 (not at all vivid). The mean vividness score was 2.4 (S.D. = 0.6) corresponding to a verbal rating of “vivid”. No participants reported vividness ratings greater than three (“somewhat vivid”). 3.1. H-reflex and soleus BEMG responses for the actual and imagery conditions An ANOVA was performed to examine if H-reflex amplitude differed from rest during the experimental trials (40, 60, 80, and maximal) involving either actual or imagined contractions. A significant (F(3,132) = 3.39, P < 0.05) main effect was found for the actual condition. Post hoc analysis indicated that H-reflex amplitude was elevated above rest in all of the actual contraction conditions, this result was expected. However, there was no main effect for imagery conditions (Fig. 2). The effect size for H-reflex for the actual and imagery trials by intensity from 40, 60, 80% and maximal were 0.37, 0.26, 0.39 and 0.17, 0.10, 0.11, 0.13, respectively. The ANOVA for involved soleus BEMG responses indicated a significant (F(3,132) = 94.97, P < 0.05) main effect for the actual condition. Post hoc analysis indicated that BEMG responses were elevated above the rest condition in all of the actual contraction conditions, this elevation is consistent with previous research and was anticipated. A significant main effect was not observed in the imagery condition (F(3,132) = 0.01, P = 0.99) (Fig. 3). The effect size for soleus BEMG for the actual and imagery trials by intensity from 40, 60, 80% and maximal were 2.21, 2.47, 2.35 and 0.22, 0.22, 0.44, 0.53, respectively. The ANOVA for non-involved soleus BEMG responses also revealed differences from rest within each condition (i.e. actual, imagery). A significant (F(3,132) = 17.23, P < 0.05) main effect was found for the actual condition. Post hoc analysis indicated that BEMG responses were elevated above rest in all of the actual contraction conditions. A significant main effect was not observed in the imagery condition (F(3,132) = 0.32, P = 0.81). BEMG activity was also assessed in the tibialis anterior of the involved and non-involved legs. An ANOVA indicated that there were no changes in BEMG activity of this muscle during imagery. However, a significant (F(3,132) = 47.98, P < 0.05) main effect was found for the actual condition. Post hoc analysis indicated that BEMG responses were elevated above rest in all of the actual contraction conditions (Fig. 4).
84
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
Fig. 2. Soleus H-reflex activity during actual and imagined contractions across percentages of MVC. The mean H-reflex responses are presented as they relate to the mean maximum motor response, as the H/M ratio, as well as standard errors for both conditions.
Fig. 3. Soleus BEMG activity in the involved and non-involved (contralateral) legs during imagery trials. The mean BEMG data are presented as well as the measure of standard error.
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
85
Fig. 4. Tibialis anterior BEMG activity in the involved and non-involved (contralateral) legs during imagery trials. The mean BEMG data are presented as well as the measure of standard error.
3.2. H-reflex and BEMG responses by trial order The H-reflex amplitude values in the imagery conditions were analyzed for a trials effect and the ANOVA revealed a significant (F(4,88) = 3.36, P < 0.05) main effect. H-reflex amplitude generally increased from the rest condition as the imagery trials progressed. Post hoc analysis indicated this increase reached significance in the final two blocks of imagery trials (Fig. 5). The effect size for H-reflex by trials
by trial block from block 1 to block 4 were 0.04, 0.07, 0.21 and 0.25, respectively. A significant main effect (F(4,88) = 2.67, P < 0.05) was observed for the soleus in the involved leg. Post hoc analysis indicated significant increases from rest in soleus BEMG of the involved leg during the first and third blocks of trials (Fig. 6). The effect size for soleus BEMG by trials by trial block from block 1 to block 4 were 0.44, 0.22, 0.44 and 0.22, respectively.
Fig. 5. H-reflex activity in the soleus across imagery trial blocks. The mean H-reflex responses are presented as they relate to the maximum motor response, as the H/M ratio, as well as the standard error.
86
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
Fig. 6. Soleus BEMG activity in the involved and non-involved (contralateral) legs across imagery blocks. The mean BEMG and standard error data are presented.
Changes in tibialis anterior BEMG activity for the involved leg were not observed (F(4,88) = 1.96, P > 0.05). A significant main effect (F(4,88) = 2.98, P < 0.05) was observed for the tibialis anterior in the non-involved leg. Post hoc analysis indicated all conditions were significantly elevated over rest for trial blocks.
4. Discussion The present study was conducted to examine the consequences of mental imagery of a simple motor task on the H-reflex and BEMG. One finding was the lack of a significant H-reflex response or BEMG changes during imagery of plantar flexion at differing intensities. According to the psychoneuromuscular theory of mental imagery [11], increased physiological activity should occur with imagined movement. Further, the physiological changes should correspond to the intensity of the imagined effort, although at a lower magnitude than would be found with actual movement [4,10]. However, as noted by Williams et al. [11] “fully convincing support for this theory remains elusive” (p. 284). Specifically, it was found that H-reflex amplitude and soleus BEMG activity were each unchanged from baseline during imagery trials. Additionally, there was no evident dose–response relationship with H-reflex responses or BEMG with increased intensities of imagined effort. As
expected during the actual contraction conditions, significant increases in both H-reflex amplitude and BEMG activity were observed in both the tibialis anterior and the soleus. The changes in soleus BEMG of the involved leg increased linearly with elevations in the intensity of contraction. Although H-reflex did not change significantly during any of the imagined intensity conditions, a significant main effect was found when the data were analyzed by trial order, independent of intensity. The primary finding was an observed increase in H-reflex amplitude that reached significance in the third and final (i.e. fourth) block of trials. The pattern of soleus BEMG activity for trials was different, with elevations above rest occurring in the first and third trial blocks. Thus, H-reflex responses to imagery were independent of BEMG, a factor known to influence H-reflex [1]. Thus the present H-reflex changes appear to be independent of BEMG activity. Although changes in BEMG were noted future research might want to only sample imagery trials that show no change in BEMG. Previous research involving H-reflex and imagery has yielded differing results. Oishi et al. [9] observed that elite speed skaters, who were experienced mental imagers, decreased the amplitude of the H-reflex during imagery of a speed skating event. The elevation in H-reflex amplitude across imagery trial blocks observed in this study is consistent with prior findings by Jeannerod [8] in which mental practice was associated with an increase in motor neuron excitability and motor unit recruitment. In contrast, Bon-
B.S. Hale et al. / Behavioural Brain Research 142 (2003) 81–87
net et al. (unpublished, [8]) found only a small increase in H-reflex in subjects who imaged weak or strong presses on a foot pedal. They reported the increase in T-reflexes, an indicator of gamma motor neuron activity, in imagery of a strong or weak press on a foot pedal. The elevation in T-reflex was found to be only slightly less than what they observed during actual movement. Based on this finding the authors contend that imagery had a greater effect on the gamma-afferent activity and research will be needed to test this hypothesis. While the present investigation did not test for differences in H-reflex response to the motor imagery task based on sex, future research investigating these responses should include an analysis based on sex. Barton [2] reported no differences between females and males on a performance-based assessment of imagery effectiveness. Whereas, Epstein [5] reported differences in dart throwing performance based on sex. Males were reported to perform better than females after imagery training. The equivocal nature of these findings suggests that future research should continue to assess the difference of imagery based on sex. The participants in this study were given some practice in the activity prior to the imagery conditions. It is possible that larger changes in H-reflex amplitude could have occurred with additional physical practice prior to imagery, but this is unlikely given the simple nature of the task used in this study. Increases in H-reflex amplitude observed for trial order during imagery trials may also have resulted from operant conditioning, or from the stress of repeated electrical stimulation elevating the H-reflex. However, H-reflex amplitude did not differ from the pre-test baseline value, indicating that the changes in H-reflex amplitude were not a simple habituation or practice effect. Thus it appears that H-reflex amplitude during imagery is modulated by practice independent of the percent of contraction imagined. Based on current findings future research investigating the effects of motor imagery on the H-reflex should control for trial effects. The current data indicates that the modulations observed in H-reflex are a function of practice.
87
In summary, the present study found that H-reflex and BEMG activity during imagery was unaffected by the effort of the imagined task. However, H-reflex activity increased across trials. Significant changes in BEMG also occurred, however, based upon the difference in magnitude of these responses it is suggested that these changes are unrelated to the H-reflex changes during imagery trials.
References [1] Angulo-Kinzler R, Mynark R, Koceja D. Soleus H-reflex gain in elderly and young adults: modulation due to body position. J Gerontol 1998;53:M120–5. [2] Barton K. The effect of mental imagery on sport climbing performance of college students. Microform Publications, International Institute for Sport and Human Performance, University of Oregon, 1997. [3] Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum; 1988. [4] Decety J, Jeannerod M, Germain M, Pastene J. Vegetative response during imagined movement is proportional to mental effort. Behav Brain Res 1991;42:1–5. [5] Epstein ML. Relationship of mental imagery and mental rehearsal to performance of a motor task. J Sport Psychol 1980;2:211–20. [6] Hashimoto R, Rothwell JC. Dynamic changes in cortico spinal excitability during motor imagery. Exp Brain Res 1999;125:75–81. [7] Hugon M. Methodology of the Hoffmann reflex in man. New developments in electromyography and clinical neurophysiology 1973;3:277–93. [8] Jeannerod M. Mental imagery in the motor context. Neuropsychologia 1995;33:1419–32. [9] Oishi K, Kimura M, Yasukawa M, Yoneda T, Maeshima T. Amplitude reduction of H-reflex during mental movement simulation in elite athletes. Behav Brain Res 1994;62:55–61. [10] Wang Y, Morgan W. The effect of imagery perspectives on the psychophysiological responses to imagined exercise. Behav Brain Res 1992;52:167–74. [11] Williams J, Rippon G, Stone B, Annett J. Psychophysiological correlates of dynamic imagery. Br J Psychol 1995;86:283–300. [12] Yue G, Cole K. Strength increases from the motor program: comparison of training with maximal voluntary and imagined muscle contractions. J Neurophysiol 1992;67:1114–23.