Directory UMM :Data Elmu:jurnal:A:Advances in Physiology Education:Vol277.Issue6.Dec1999:

(1)

ELECTROPHY SIOLOGY OF THE UNDERGRADUATE

NEUROSCIENCE STUDENT: A LABORATORY

EX ERCISE IN HUMAN ELECTROMY OGRAPHY

Robert C. Lennartz

Depa rtm ent of Psychology a nd Progra m in Neuroscience, Willia m s College, Willia m stown, Ma ssa chusetts 01267

A

laboratory exercise is described in which students in a neuroscience, psychobi-ology, or similar laboratory course record the electromyogram (EMG) from themselves, using surface electrodes (placed on the skin). This exercise is intended to give students a firsthand demonstration that electrical activity is produced within them and to allow the students to use this activity to study biological and psychological concepts. The students study the nature of the EMG (changes with tension and the temporal relationship with limb movement) and the concepts of flexion and extension, reaction time, and the patellar (‘‘knee jerk’’) reflex. In postlaboratory evaluations, undergraduate introductory neuroscience students indicated that they appreciated the opportunity to record electrical activity from their own bodies. The students found the exercise enjoyable, believed that they had learned from it, and indicated that it should be a regular part of the course. If electrophysiology in animal preparations is already part of the course, this exercise requires minimal additional equipment, some of which is easily constructed and the remainder of which is available inexpensively.

AM. J. PHYSIOL. 277 (ADV. PHYSIOL. EDUC. 22): S42–S50, 1999.

Key words:electromyogram; muscle; reflex; reaction time

The laboratory component of undergraduate neurosci-ence, psychobiology, and similar courses often in-cludes at least one exercise in electrophysiology that allows students to record electrical activity from living tissue. This type of exercise gives students exposure to electrophysiological techniques and the use of these techniques in answering experimental ques-tions. For example, the students in the Introduction to Neuroscience course at Williams College record ac-tion potentials from a cockroach leg to study sensory coding by the nervous system.

One problem with these preparations is that the students do not always fully appreciate that such activity occurs within themselves as well. Thus a valuable addition to such laboratory courses is an

exercise that allows students to detect, and use in experiments, electrical signals produced by their own bodies. A limitation is that any such exercise must use noninvasive methods, and thus any electrical activity must be detectable from the skin surface.

Neural activity in the form of the electroencephalo-gram (EEG) and evoked potentials can be recorded from the human scalp; however, sophisticated equip-ment is required, as is special treatequip-ment of the data (such as the averaging of responses over many trials). Muscles produce another type of electrical activity detectable from the human skin surface, the electro-myogram (EMG). The EMG is easily recorded, and if electrophysiology is already part of the course, the basic equipment is available. Only inexpensive

(2)

elec-trodes and some readily available and easily con-structed additional items are needed.

Among the concepts investigated by the students in this exercise are 1) the nature of the EMG—how it changes with muscle tension, and the temporal rela-tionship between muscle action potentials (which comprise the EMG) and limb movement; 2) flexion and extension;3) reaction time; and4) the patellar, or ‘‘knee-jerk,’’ reflex. In addition to providing the stu-dents with an opportunity to record electrical activity from their own bodies, these experiments involve students in active learning of ideas that they have been exposed to in lecture courses. For example, the idea that a reflex is considerably faster than a similar voluntary response to a stimulus is a fundamental concept; in this laboratory exercise, this is studied empirically. Furthermore, to encourage critical think-ing, the students can be asked to predict and explain results. Thus, for example, they can predict, on the basis of their knowledge of the EMG source, whether a change in this signal will precede a movement or follow it, and then test their prediction.

EQUIPMENT

Several companies manufacture surface EMG elec-trodes (typically silver-silver chloride) with attached leads. These require electrode gel, for skin-electrode continuity, and adhesive disks, for securing the elec-trodes. General-purpose EMG electrodes work well for this experiment; the miniature electrodes (which are more selective and can be more closely spaced) should be avoided. It is more difficult to maintain good skin contact with these smaller electrodes, and this can lead to increased noise problems; furthermore, the increased recording selectivity provided by the miniature electrodes is not necessary for this exercise. Self-adhesive, pregelled, disposable electrodes are avail-able inexpensively (Biopac Systems, Santa Barbara, CA), and were used here; they attach to reusable leads.

From the electrodes, the signal goes to a preamplifier (Grass P511 was used) set to record activity within a range of,30–1,000 Hz. The gain should initially be

5,000, with changes made as necessary. The signal is then sent to a storage oscilloscope. A digital scope is useful, but not necessary.

Three other instruments are required:1) an inexpen-sive ‘‘handgrip’’ exerciser; 2) a simple ‘‘telegraph’’-style switch that produces a voltage change when a finger is lifted; and 3) a percussion hammer that triggers an oscilloscope when struck. The switch is easily constructed, as illustrated in Fig. 1A. The percussion hammer, although commercially available, can be constructed less expensively. A simple percus-sion hammer is available (Carolina Biological Supply, Burlington, NC) and is prepared as shown in Fig. 1B. A Faraday cage to shield out electrical noise should not be necessary. If there is considerable noise, the ground connection should first be checked. An elec-trode attached to the wrist and connected to the proper amplifier input serves as the ground. An electrode such as those used for the EMG recordings may be sufficient for this purpose (as it was with the disposable electrodes used here); however, a ground electrode with a very large surface area may further reduce the noise level. Another potential cause of excessive 60-Hz noise is poor electrode-skin contact. This can occur if insufficient electrode gel has been used (or, when using the disposable electrodes, if the gel has dried) or if the electrode has pulled away from the adhesive. If the noise remains after all of the electrodes have been checked, the 60-Hz filter on the amplifier can be activated (if the amplifier is so equipped). The use of such filters is not recom-mended in EMG experiments because they filter out part of the EMG signal (6); however, no fine analysis of the signal is done in this exercise, and so this loss is fairly inconsequential to the experimental results. Electr ode Placements

Three recording sites are used, each requiring three electrode placements: two placements for differential EMG recording, and one for ground. The two forearm sites are based on descriptions by Davis as reported in Lippold (13); the quadriceps site is from Hale et al. (8). These surface recordings detect activity from more than one muscle. Rather than identifying all the possible muscles that may contribute to the record-ings, the nomenclature of ‘‘forearm flexors’’ and ‘‘forearm extensors’’ is used for the electrode place-ments on the arms. Likewise, the leg muscles relevant to this exercise are identified as the quadriceps.

(3)

For ear m flex ors. The electrodes for this site are attached to the ventral (bottom) surface of the fore-arm. With the forearm held with the palm of the hand up, the two EMG electrodes are placed on an imagi-nary line running diagonally across the entire length of the forearm, from the outer edge of the wrist (just below the thumb side of the hand) to the inner, most posterior part of the forearm. One electrode is posi-tioned at a point on this line one-third of the way from the posterior part of the forearm, and the second electrode is placed on this line ,5 cm (distances are

based on the electrode centers) from the first, toward the wrist. The ground electrode is placed on the wrist. For ear m ex tensors.The electrodes for this site are attached to the dorsal (top) surface of the forearm. With the arm held with the palm of the hand down, the two EMG electrodes are placed on an imaginary line running straight back along the surface of the forearm, from the outer edge of the wrist (just below the little finger side of the hand) to the outer, most posterior part of the dorsal surface of the forearm (near the elbow). One electrode is placed at a point on this line one-third of the way from the posterior part of the forearm, and the second electrode is placed on the line,5 cm from the first, toward the wrist. The

ground electrode can be left in place from the forearm flexor placement.

Quadriceps. On the dorsal (top) surface of the leg, the point approximately one-half the distance be-tween the hip and the knee, along the midline, is located, and one electrode is placed anterior (toward the knee) and one posterior (toward the body) to this point,,3 cm apart. The ground electrode is placed on

the wrist. EXPERIMENTS

Basic concepts such as motor units, flexion, exten-sion, and simple reflexes should be discussed before the laboratory exercise is done. The following points regarding the EMG should also be stressed. 1) The muscle action potentials, which comprise the EMG, occur before the actual contraction of the muscle; it is the processes initiated by these potentials that lead to the contraction. 2) The surface EMG represents the FIG. 1.

Two easily constructed pieces of equipment r equir ed for the laboratory ex er cise. A: a simple ‘‘telegraph’’-style switch. Size specifics ar e not critical; the wooden base used her e was 1.3 cm thick, 15.2 cm long, and 6.4 cm wide. An aluminum strip (D0.4 mm thick, 8.9 cm long, and 1.3 cm wide) is fastened at one end to the wood and bent upwar ds. It can be pushed down (to complete cir cuit) with a finger, but r etur ns to its original position as the finger is raised. A scr ew is inserted into a hole in the fr ee end of the strip, thr eaded thr ough a bolt on the bottom of the strip, and tightened; electrician’s tape or shrink tubing is placed ar ound the end, with the scr ew pr otruding. The scr ew in the wood beneath the fr ee end of the strip should be offset slightly to pr event the upper scr ew fr om getting lodged in the slot. B: a per cussion hammer rigged to trigger an oscilloscope. One end of a r ectangular piece of aluminum (D0.4 mm thick, 3.5 cm long, and 1.1 cm wide) is attached to the hammerhead by a scr ew; once a starting hole is made, the scr ew is driven easily into the rubber head. Another scr ew is placed in the hammerhead under the fr ee end of the aluminum strip. The strip is bent such that it is almost touching this second scr ew. The scr ews secur e the leads of a thin, two-conductor cable. The other end of the cable (D2 m) is the input to the oscilloscope trigger. A small 9-Vbattery and a 100-kVr esistor ar e placed in series with this cir cuit. When the hammerhead is struck (on the ‘‘pointed’’ end, opposite the switch), the aluminum strip makes contact with the second scr ew, com-pleting the cir cuit and triggering the oscilloscope. Strain r elief on the cable is pr ovided by attaching it to the hammer handle with shrink tubing.

(4)

summed population response (from muscle action potentials) reaching the skin from muscle fibers within a motor unit and across many motor units (6), and the contribution of these fibers to a movement can be assessed by measuring the EMG amplitude. 3) The electrodes usually record from more than one muscle (thus the recordings done in these experiments are from muscle groups). 4) As a muscle contracts with increasing force, both an increase in the firing rate of motor units and the recruitment of additional motor units are involved, with the relative contribution of these two factors partially dependent on the size of the muscle (3).

In addition to this basic knowledge about the nature of the EMG, its use in studying reflex and reaction times should also be covered, because this is relevant to two of the experiments in this exercise. The term ‘‘reaction’’ is used here as it is in describing reaction

time experiments, to indicate a voluntary response to a stimulus. Figure 2 shows how the time from a stimulus to an overt skeletal response (whether a reflex or a reaction) can be partitioned into two components.

The first component represents the time from the stimulus to a change in the EMG from the muscles required for the response. For reflex time, this compo-nent is the spinal reflex latency. The EMG change in this case occurs at the end of a reflex arc that has a single synapse in the spinal cord. For reaction time, this first component is the supraspinal response initia-tion time. The involvement of the brain (hence, ‘‘supraspinal’’) is required for the response as well as for detection of the stimulus.

The second component is the time from the EMG change to an overt response; this is the electromechani-cal delay (for both reflex and reaction times). It includes the time required for the processes leading up to muscle contraction and the time required for sufficient contraction to result in a detectable move-ment.

Temporal partitions similar to these have been used in studies on reflex (8, 9, 10, 12, 14) and reaction (1, 4, 5, 8, 11, 15) times.

If there are two people per station, one can be the subject for the forearm sites experiments and the other for the quadriceps site experiments.

FOREARM FLEXORS Flex ion

Concepts addr essed.This exercise serves to familiar-ize the students with the EMG signal and (along with a later exercise) to demonstrate that different muscle groups are responsible for flexion and extension of limbs.

Temporal partitioning of a skeletal r esponse to a stimulus. The first component is the time fr om stimulus to a change in the electr omyogram (EMG) fr om the muscles mediating the r esponse. In the case of a spinal r eflex , this first component is the spinal r eflex latency, and for a voluntary r eaction, it is the supraspinal r esponse initiation time (‘‘supraspinal’’ r e-flects the involvement of the brain). The second component is the electr omechanical delay and r epr esents the time r equir ed for the pr ocesses occurring between the muscle action potentials and the movement.

(5)

the back of the wrist (extension) to determine for which movement more EMG activity is observed. Pr edicted r esults. There should be more activity during flexion compared with extension. Muscles (and muscle groups) can only move a limb in one direction. Because these electrodes are on the ventral surface of the forearm, activity (leading to contrac-tion) in the muscles under these electrodes would be expected when the hand moves toward the front of the wrist (flexion). This is evident in the sample data shown in Fig. 3A.

For ce and EMG

Concepts addr essed. This exercise demonstrates what happens to the EMG as the force exerted by muscles is increased.

Pr ocedur e. The handgrip ex erc iser is slowly squeezed, with the oscilloscope at a slow speed (500 ms/div), and any change in the EMG is noted.

Pr edicted r esults.The prediction is that there will be an increase in the peak-to-peak amplitude (from the peak of a negative wave to the peak of the next positive wave) of the surface EMG as the exerciser is squeezed. There is an increase in the frequency of muscle action potentials within motor units and an increased number of active motor units as muscle tension is increased, and this electrical activity sum-mates. The increase in the EMG amplitude is illus-trated in Fig. 3B.

FOREARM EXTENSORS Ex tension

Concepts addr essed.As in the flexion exercise, this exercise demonstrates that different muscle groups perform flexion and extension.

Pr ocedur e. The arm is held with the elbow by the side of the body and the forearm extended with the fingers curled into a loose fist. The hand is slowly moved toward the front of the wrist (flexion) and then the back of the wrist (extension) to determine for which movement more EMG activity is observed. Pr edicted r esults. There should be more activity during extension compared with flexion. Because these electrodes are on the dorsal surface of the forearm, activity (leading to contraction) in the muscles under these electrodes would be expected when the hand moves toward the back of the wrist (extension). Note: The focus during the wrist extension (and the earlier flexion) should be on what happens as the hand moves from the midline position because, for example, during extreme flexion there may be EMG activity in the extensors due to these muscles in the FIG. 3.

Data fr om for ear m flex or ex periments.A: oscilloscope trace showing activity in for ear m flex or muscles during move-ment of the hand. Ther e is gr eater activity during flex ion (as the hand is moved towar d the fr ont of the wrist, or flex ed) than during ex tension (as the hand is moved towar d the back of the wrist, or ex tended).B: change in activity of for ear m flex ors as for ce ex erted by the muscles is incr eased. EMG amplitude incr eases as a handgrip ex er ciser is squeezed.

(6)

antagonist position becoming active to stabilize the wrist (2). If desired, the students can demonstrate this for themselves.

Temporal Relationship Between EMG and Movement

Concepts addr essed.This experiment addresses the issue of the timing of the EMG change relative to movement of a finger.

Pr ocedur e.The hand is placed palm down with the forefinger on the switch. The output of the switch goes to one channel on the oscilloscope and the EMG to the other channel. With the sweep speed at 20 or 50 ms/div, one person starts a sweep of the oscillo-scope and the subject quickly lifts the forefinger. This step may need to be repeated to get both the EMG and finger lift on the oscilloscope. (If the oscilloscope allows you to see the ‘‘pretrigger’’ period, then the channel with the output of the switch can be used to trigger the scope.) The students determine whether the EMG change occurs first, the finger movement occurs first, or both occur simultaneously.

Pr edicted r esults. The prediction is that the EMG change precedes the finger lift. It is the muscle action potentials, which comprise the EMG, that lead to the process of contraction. Figure 4 shows sample data for this experiment, with the change in the EMG clearly proceeding the lifting of the finger. This illustrates the electromechanical delay described in Fig. 2.

Note: When doing this and the next experiment (reaction time), the EMG change preceding the finger

lift is easiest to discern if the hand and forearm are relaxed, with minimal baseline EMG activity, before the finger is lifted.

Reaction Time

Concepts addr essed.This experiment is an investiga-tion of reacinvestiga-tion time, which is the time required to make a voluntary response after a stimulus. Specifi-cally, reaction time is determined here as the amount of time required for a subject to lift a finger after the hammer strikes the table. Weiss (15) divided reaction time into two components corresponding to the supraspinal response initiation time and the electrome-chanical delay, illustrated in Fig. 2 (Weiss used the terms ‘‘premotor time’’ and ‘‘motor time,’’ respec-tively.) He was trying to determine whether variables, including motivation, that influence reaction time have their major effects on the events occurring between the stimulus and the change in the EMG, or in the events occurring between this change and the overt response. Others have since done similar analy-ses to examine the effects of such variables as fatigue (11), temperature (1), and ethanol (4) on the two components. Here, the students observe the reaction time variability that occurs over trials and then deter-mine which of the two components, the supraspinal response initiation time or the electromechanical delay, accounts for most of this variability.

Pr ocedur e.The EMG and switch outputs are left on separate oscilloscope channels, and the hammer is connected to the oscilloscope trigger. The subject sits with eyes closed and forefinger on the switch. An-other student strikes the hammer on the table or An-other surface (using the ‘‘pointed’’ part of the hammerhead, and not the switch), and the subject lifts the finger rapidly on hearing the strike. (If more than one oscilloscope sweep occurs per hammer strike due to ‘‘bounce’’ in the switch, the oscilloscope can be set to ‘‘single sweep’’). On the oscilloscope, three times are determined: reaction time (hammer strike—when the oscilloscope is triggered—to finger lift), the supraspi-nal response initiation time (hammer strike to start of change in EMG activity), and the electromechanical delay (start of change in EMG activity to finger lift). Students do 10 trials and plot the means 6 SE of reaction time and its components. They then plot a scatter graph of reaction time for the 10 trials versus FIG. 4.

Oscilloscope trace showing temporal r elationship between EMG and finger movement. The switch battery was con-nected such that a voltage incr ease was indicated on the oscilloscope scr een when the finger was lifted.

(7)

each of the two components and calculate correlation coefficients to determine which component is better correlated with reaction time; Botwinick and Thomp-son (5) did a similar analysis to address the issue of which reaction time component accounts for most of the reaction time variability. The present experiment takes advantage of the reaction time variability that tends to occur across trials; no variables are manipu-lated. The basic question that is addressed is whether this variability is mainly due to processes occurring between the occurrence of the stimulus and the muscle action potentials or between the muscle action potentials and the lifting of the finger. This will be reflected as a greater correlation between that particular component and the total reaction time. Alternatively, both processes could be making equal contributions to the variability in the reaction time, in which case the two correlations will be about the same.

Pr edicted r esults.The prediction is that the supraspi-nal response initiation time will be more highly correlated with the reaction time. This first compo-nent includes the time required to detect and respond to the stimulus. This component will be influenced by any psychological variables—attention, for example— that may vary from trial to trial. The second compo-nent—the electromechanical delay—is largely a bio-chemical and physical process, and less trial-to-trial variability would be expected. The sample data (Fig. 5) show that the supraspinal response initiation time is more highly correlated with the reaction time. QUADRICEPS

Reflex Versus Reaction Times

Concepts addr essed. This experiment serves to investigate whether the time required for a response to a stimulus—jerk of the leg (or, in this case, the EMG change preceding this) in response to a tap—varies depending on whether it is a reflex or a reaction. Pr ocedur e.Students should bring short pants to wear for this experiment. The subject sits with the leg with the electrodes crossed over the other leg, and the experimenter strikes the hammer on the patellar tendon just below the kneecap, triggering the oscillo-scope and eliciting a knee jerk. Ten trials are given this

way and ten more trials are given in which the hammer is struck at a point near the tendon, which does not elicit a reflex; the subject jerks the leg when detecting the tap. For each trial, the time between the stimulus and the EMG change is measured. For the first 10 trials, this represents the spinal reflex latency; for the second 10 trials, it is the supraspinal response initiation time (Fig. 2). The means 6SE are plotted. The purpose of this exercise is to determine whether one of these time periods is greater than the other, and if so, which one is greater.

Pr edicted r esults. The reaction, unlike the reflex, requires brain involvement and thus should take considerably longer. The sample data in Fig. 6,A(raw data) andB(plotted data), show this. This experiment is essentially an empirical test of a similar idea dis-cussed in a textbook by Carlson (7 ).

EVALUATION OF EXERCISE

This exercise was performed in the 1996 fall semester in the Introduction to Neuroscience course at Wil-liams College. All students had previously performed laboratory exercises on the dissection of the sheep brain, Golgi staining of the rat cerebellum, and sen-sory coding in the cockroach. Twenty of the ninety-three students in the course chose the EMG exercise for the last laboratory session (they had 2 other options: an exercise on hippocampal long-term

poten-FIG. 5.

Reaction time data fr om one subject. Supraspinal r esponse initiation time is better corr elated with r eaction time than is electr omechanical delay.r, Corr elation coefficient.

(8)

tiation or an exercise on hemispheric and gender differences in frontal cortex EEG). The 20 students were distributed over 3 laboratory sections. There were usually two (and never more than 3) students per setup. The exercise took approximately two hours, including a brief lecture to explain the EMG. The students were required to write a laboratory report, the grade of which comprised 10% of their total grade for the course.

Of the 20 students who performed the EMG exercise, 17 opted to evaluate it. These evaluations (on a scale of 1 to 7) were done immediately at the end of the session, before the students wrote up their results. The lab was evaluated along three dimensions. 1) Enjoyment, with 1 5not at all enjoyable and 7 5 very enjoyable (4 5 indifferent): mean 5 6.2,

range 5 5–7. Although this should not be the main criterion by which a lab exercise is judged, students obviously will be more involved in experiments that they enjoy doing.

2) Learning from the exercise, with 1 5 I learned nothing from doing this laboratory exercise and 75I learned a lot from doing this laboratory exercise (45I learned a moderate amount from doing this laboratory exercise): mean 5 5.2, range 5 4–6. This is a self-report measure of learning and thus is not as valid as test scores, etc. However, it does indicate that the students seemed to believe that the exercise was worthwhile.

3) Whether the exercise should be continued as part of the course, with 1 5 I think that this laboratory exercise should definitely be dropped from the course and 7 5 I think that this laboratory exercise should definitely be a part of this course (45I really have no opinion either way): mean5 5.8, range5 5–7. This was the first time that this exercise was done, and this item was included as a way to gauge the students’ overall opinion of it and to give some indication of whether it should remain as part of the laboratory course.

There was also space on the evaluation form for comments regarding the exercise. Of the 16 students answering the question, ‘‘What did you like most about today’s lab?,’’ eight included as at least part of the answer that they enjoyed the use of human subjects. These responses included ‘‘it was interesting to use our own bodies,’’ ‘‘being hooked up to a computer,’’ and ‘‘[as] student subjects we saw how these motor systems functioned in ourselves (as op-posed to cockroaches, rats, etc.).’’

DISCUSSION

One goal of this exercise was for the students to gain an appreciation for the idea that the electrical pro-cesses they have heard and read about in courses, and perhaps observed in animal preparations, are occur-ring in their own bodies. The student comments cited previously demonstrated that they did enjoy the opportunity to observe this activity. Another goal was for the students to engage in active learning. One student did indicate in the laboratory evaluation that he or she ‘‘...enjoyed seeing a somewhat ‘practical’ FIG. 6.

Knee jerk as a r eflex and as a r eaction.A: oscilloscope traces fr om quadriceps.Top tr ace: EMG r esponse of a subject to a tap on the patellar tendon (r eflex ). Bottom tr ace: EMG r esponse when subject voluntarily moved the leg after feeling a tap near the tendon (r eaction).B: means6SE of supraspinal r esponse initiation time and spinal r eflex la-tency. Ten trials wer e given for both conditions.

(9)

application of the material learned in class,’’ and another found it ‘‘...interesting to see the correlation between the APs and movement.’’ A third goal was to encourage critical thinking. As a method for accom-plishing this, the students were required to write a laboratory report on this exercise to describe and offer explanations for their findings, which did require them to think about the data. However, what could have aided this would have been to provide a work-sheet the week previous to the exercise (along with the laboratory description) in which the students would be required to predict the results and offer explanations of the predictions. (They could be al-lowed to modify these after the laboratory lecture, and before doing the experiments, in the event that the lecture clarified any poorly understood concepts). The predictions could then be incorporated into the lab reports.

The experiments are described here in a procedurally logical order. However, for a postlaboratory assign-ment (whether a list of questions to answer or a laboratory report), it would be best if the results were organized in a conceptual manner. Thus the experi-ments on muscle tension and the temporal relation-ship of the EMG to movement could be grouped together because both investigate aspects of the nature of the EMG. The flexion and extension experi-ments could similarly be grouped in a postlaboratory assignment.

Although this does work extremely well as a labora-tory exercise, the instructor and one or two students could do all or part of it as a demonstration, perhaps to supplement another laboratory exercise.

This exercise is relatively inexpensive, requiring mini-mal additional equipment over that found in a typical neuroscience laboratory. The use of the percussion hammer, designed for studying reflexes, to also pro-duce the stimulus and trigger the oscilloscope for the reaction time experiment eliminates the need for an additional apparatus for this purpose. Furthermore, there are no expenses for animal care because the subjects take care of their own housing and feeding. I thank Dr. Steven J. Zottoli for advice, Emile Ouellette for technical assistance, Felicity Adams for serving as a subject during develop-ment of this exercise, and the Introduction to Neuroscience students for participation and evaluations.

This work was supported by a grant from the Essel Foundation to Williams College.

This paper is based on a presentation given at the 26th Annual Meeting of the Society for Neuroscience, November 16–21, 1996, in Washington, DC.

Address for reprint requests and other correspondence: R. C. Lennartz, Dept. of Psychology, College of William and Mary, PO Box 8795, Williamsburg, VA 23187-8795 (E-mail: rclenn@wm.edu). Received 2 April 1999; accepted in final form 20 July 1999.

Refer ences

1. Air d, J. W., R. D. G. Webb, and J. Hoar e. Heat exposure-induced changes in motor outflow component of reaction time.

Percept. Mot. Sk ills56: 699–706, 1983.

2. Ba¨ckdahl, M., and S. Carlso¨o¨. Distribution of activity in muscles acting on the wrist (an electromyographic study).Acta Morph. Neerl. Sca nd. 4: 136–144, 1961.

3. Basmajian, J. V., and C. J. De Luca. Muscles Alive: Their Functions Revea led by Electrom yogra phy(5th ed.). Baltimore, MD: Williams and Wilkins, 1985.

4. Baylor, A. M., C. S. Layne, R. D. Mayfield, L. Osbor ne, and W. W. Spir duso. Effects of ethanol on human fractionated response times.Drug Alcohol Depend. 23: 31–40, 1989. 5. Botwinick, J., and L. W. Thompson. Premotor and motor

components of reaction time.J. Exp. Psychol. 71: 9–15, 1966. 6. Cacioppo, J. T., L. G. Tassinary, and A. J. Fridlund. The

skeletomotor system. In:Principles of Psychophysiology: Physi-ca l, Socia l, a nd Inferentia l Elem ents, edited by J. T. Cacioppo and L. G. Tassinary. Cambridge, UK: Cambridge Univ. Press, 1990, p. 325–384.

7. Carlson, N. R.Physiology of Beha vior(6th ed.). Boston, MA: Allyn and Bacon, 1998.

8. Hale, B. D., D. M. Landers, R. Snyder Bauer, and N. L. Goggin.Iris pigmentation and fractionated reaction and reflex time.Biol. Psychol. 10: 57–67, 1980.

9. Hart, B. A. Fractionated myotatic reflex times in women by activity level and age.J. Gerontol. 41: 361–367, 1986.

10. Hayes, K. C.Jendrassik maneuver facilitation and fractionated patellar reflex times.J. Appl. Physiol. 32: 290–295, 1972. 11. Klimovitch, G.Startle response and muscular fatigue effects

upon fractionated hand grip reaction time.J. Mot. Beha v. 9: 285–292, 1977.

12. Kr oll, W. Patellar reflex time and reflex latency under Jen-drassik and crossed extensor facilitation.Am . J. Phys. Med. 47: 292–301, 1968.

13. Lippold, O. C. J.Electromyography. In:A Ma nua l of Psycho-physiologica l Methods, edited by P. H. Venables and I. Martin. New York: Elsevier, 1967, p. 245–297.

14. Morris, A. F. Effects of fatiguing isometric and isotonic exercise on fractionated patellar tendon reflex components.

Res. Q. 48: 121–128, 1977.

15. Weiss, A. D. The locus of reaction time change with set, motivation, and age.J. Gerontol. 20: 60–64, 1965.

(1)

summed population response (from muscle action

potentials) reaching the skin from muscle fibers within

a motor unit and across many motor units (6), and the

contribution of these fibers to a movement can be

assessed by measuring the EMG amplitude.

3 ) The

electrodes usually record from more than one muscle

(thus the recordings done in these experiments are

from muscle groups).

4 ) As a muscle contracts with

increasing force, both an increase in the firing rate of

motor units and the recruitment of additional motor

units are involved, with the relative contribution of

these two factors partially dependent on the size of

the muscle (3).

In addition to this basic knowledge about the nature

of the EMG, its use in studying reflex and reaction

times should also be covered, because this is relevant

to two of the experiments in this exercise. The term

‘‘reaction’’ is used here as it is in describing reaction

time experiments, to indicate a voluntary response to

a stimulus. Figure 2 shows how the time from a

stimulus to an overt skeletal response (whether a

reflex or a reaction) can be partitioned into two

components.

The first component represents the time from the

stimulus to a change in the EMG from the muscles

required for the response. For reflex time, this

compo-nent is the spinal reflex latency. The EMG change in

this case occurs at the end of a reflex arc that has a

single synapse in the spinal cord. For reaction time,

this first component is the supraspinal response

initia-tion time. The involvement of the brain (hence,

‘‘supraspinal’’) is required for the response as well as

for detection of the stimulus.

The second component is the time from the EMG

change to an overt response; this is the

electromechani-cal delay (for both reflex and reaction times). It

includes the time required for the processes leading

up to muscle contraction and the time required for

sufficient contraction to result in a detectable

move-ment.

Temporal partitions similar to these have been used in

studies on reflex (8, 9, 10, 12, 14) and reaction (1, 4, 5,

8, 11, 15) times.

If there are two people per station, one can be the

subject for the forearm sites experiments and the

other for the quadriceps site experiments.

FOREARM FLEXORS

Flex ion

Concepts addr essed.

This exercise serves to

familiar-ize the students with the EMG signal and (along with a

later exercise) to demonstrate that different muscle

groups are responsible for flexion and extension of

limbs.

Pr ocedur e.

The arm is held with the elbow by the

side of the body and the forearm extended with the

fingers curled into a loose fist. The hand is slowly

moved toward the front of the wrist (flexion) and then

FIG. 2.

(2)

the back of the wrist (extension) to determine for

which movement more EMG activity is observed.

Pr edicted r esults.

There should be more activity

during flexion compared with extension. Muscles

(and muscle groups) can only move a limb in one

direction. Because these electrodes are on the ventral

surface of the forearm, activity (leading to

contrac-tion) in the muscles under these electrodes would be

expected when the hand moves toward the front of

the wrist (flexion). This is evident in the sample data

shown in Fig. 3

A

.

For ce and EMG

Concepts addr essed.

This exercise demonstrates

what happens to the EMG as the force exerted by

muscles is increased.

Pr ocedur e.

The handgrip ex erc iser is slowly

squeezed, with the oscilloscope at a slow speed (500

ms/div), and any change in the EMG is noted.

Pr edicted r esults.

The prediction is that there will be

an increase in the peak-to-peak amplitude (from the

peak of a negative wave to the peak of the next

positive wave) of the surface EMG as the exerciser is

squeezed. There is an increase in the frequency of

muscle action potentials within motor units and an

increased number of active motor units as muscle

tension is increased, and this electrical activity

sum-mates. The increase in the EMG amplitude is

illus-trated in Fig. 3

B

.

FOREARM EXTENSORS

Ex tension

Concepts addr essed.

As in the flexion exercise, this

exercise demonstrates that different muscle groups

perform flexion and extension.

Pr ocedur e.

The arm is held with the elbow by the

side of the body and the forearm extended with the

fingers curled into a loose fist. The hand is slowly

moved toward the front of the wrist (flexion) and then

the back of the wrist (extension) to determine for

which movement more EMG activity is observed.

Pr edicted r esults.

There should be more activity

during extension compared with flexion. Because

these electrodes are on the dorsal surface of the

forearm, activity (leading to contraction) in the muscles

under these electrodes would be expected when the

hand moves toward the back of the wrist (extension).

Note: The focus during the wrist extension (and the

earlier flexion) should be on what happens as the

hand moves from the midline position because, for

example, during extreme flexion there may be EMG

activity in the extensors due to these muscles in the

FIG. 3.

(3)

antagonist position becoming active to stabilize the

wrist (2). If desired, the students can demonstrate this

for themselves.

Temporal Relationship Between

EMG and Movement

Concepts addr essed.

This experiment addresses the

issue of the timing of the EMG change relative to

movement of a finger.

Pr ocedur e.

The hand is placed palm down with the

forefinger on the switch. The output of the switch

goes to one channel on the oscilloscope and the EMG

to the other channel. With the sweep speed at 20 or

50 ms/div, one person starts a sweep of the

oscillo-scope and the subject quickly lifts the forefinger. This

step may need to be repeated to get both the EMG and

finger lift on the oscilloscope. (If the oscilloscope

allows you to see the ‘‘pretrigger’’ period, then the

channel with the output of the switch can be used to

trigger the scope.) The students determine whether

the EMG change occurs first, the finger movement

occurs first, or both occur simultaneously.

Pr edicted r esults.

The prediction is that the EMG

change precedes the finger lift. It is the muscle action

potentials, which comprise the EMG, that lead to the

process of contraction. Figure 4 shows sample data for

this experiment, with the change in the EMG clearly

proceeding the lifting of the finger. This illustrates the

electromechanical delay described in Fig. 2.

Note: When doing this and the next experiment

(reaction time), the EMG change preceding the finger

lift is easiest to discern if the hand and forearm are

relaxed, with minimal baseline EMG activity, before

the finger is lifted.

Reaction Time

Concepts addr essed.

This experiment is an

investiga-tion of reacinvestiga-tion time, which is the time required to

make a voluntary response after a stimulus.

Specifi-cally, reaction time is determined here as the amount

of time required for a subject to lift a finger after the

hammer strikes the table. Weiss (15) divided reaction

time into two components corresponding to the

supraspinal response initiation time and the

electrome-chanical delay, illustrated in Fig. 2 (Weiss used the

terms ‘‘premotor time’’ and ‘‘motor time,’’

respec-tively.) He was trying to determine whether variables,

including motivation, that influence reaction time

have their major effects on the events occurring

between the stimulus and the change in the EMG, or

in the events occurring between this change and the

overt response. Others have since done similar

analy-ses to examine the effects of such variables as fatigue

(11), temperature (1), and ethanol (4) on the two

components. Here, the students observe the reaction

time variability that occurs over trials and then

deter-mine which of the two components, the supraspinal

response initiation time or the electromechanical

delay, accounts for most of this variability.

Pr ocedur e.

The EMG and switch outputs are left on

separate oscilloscope channels, and the hammer is

connected to the oscilloscope trigger. The subject sits

with eyes closed and forefinger on the switch.

An-other student strikes the hammer on the table or An-other

surface (using the ‘‘pointed’’ part of the hammerhead,

and not the switch), and the subject lifts the finger

rapidly on hearing the strike. (If more than one

oscilloscope sweep occurs per hammer strike due to

‘‘bounce’’ in the switch, the oscilloscope can be set to

‘‘single sweep’’). On the oscilloscope, three times are

determined: reaction time (hammer strike—when the

oscilloscope is triggered—to finger lift), the

supraspi-nal response initiation time (hammer strike to start of

change in EMG activity), and the electromechanical

delay (start of change in EMG activity to finger lift).

Students do 10 trials and plot the means

6 SE of

reaction time and its components. They then plot a

scatter graph of reaction time for the 10 trials versus

FIG. 4.

(4)

each of the two components and calculate correlation

coefficients to determine which component is better

correlated with reaction time; Botwinick and

Thomp-son (5) did a similar analysis to address the issue of

which reaction time component accounts for most of

the reaction time variability. The present experiment

takes advantage of the reaction time variability that

tends to occur across trials; no variables are

manipu-lated. The basic question that is addressed is whether

this variability is mainly due to processes occurring

between the occurrence of the stimulus and the

muscle action potentials or between the muscle

action potentials and the lifting of the finger. This will

be reflected as a greater correlation between that

particular component and the total reaction time.

Alternatively, both processes could be making equal

contributions to the variability in the reaction time, in

which case the two correlations will be about the

same.

Pr edicted r esults.

The prediction is that the

supraspi-nal response initiation time will be more highly

correlated with the reaction time. This first

compo-nent includes the time required to detect and respond

to the stimulus. This component will be influenced by

any psychological variables—attention, for example—

that may vary from trial to trial. The second

compo-nent—the electromechanical delay—is largely a

bio-chemical and physical process, and less trial-to-trial

variability would be expected. The sample data (Fig.

5) show that the supraspinal response initiation time

is more highly correlated with the reaction time.

QUADRICEPS

Reflex Versus Reaction Times

Concepts addr essed.

This experiment serves to

investigate whether the time required for a response

to a stimulus—jerk of the leg (or, in this case, the EMG

change preceding this) in response to a tap—varies

depending on whether it is a reflex or a reaction.

Pr ocedur e.

Students should bring short pants to wear

for this experiment. The subject sits with the leg with

the electrodes crossed over the other leg, and the

experimenter strikes the hammer on the patellar

tendon just below the kneecap, triggering the

oscillo-scope and eliciting a knee jerk. Ten trials are given this

way and ten more trials are given in which the

hammer is struck at a point near the tendon, which

does not elicit a reflex; the subject jerks the leg when

detecting the tap. For each trial, the time between the

stimulus and the EMG change is measured. For the

first 10 trials, this represents the spinal reflex latency;

for the second 10 trials, it is the supraspinal response

initiation time (Fig. 2). The means

6 SE are plotted.

The purpose of this exercise is to determine whether

one of these time periods is greater than the other,

and if so, which one is greater.

Pr edicted r esults.

The reaction, unlike the reflex,

requires brain involvement and thus should take

considerably longer. The sample data in Fig. 6,

A

(raw

data) and

B

(plotted data), show this. This experiment

is essentially an empirical test of a similar idea

dis-cussed in a textbook by Carlson (7 ).

EVALUATION OF EXERCISE

This exercise was performed in the 1996 fall semester

in the Introduction to Neuroscience course at

Wil-liams College. All students had previously performed

laboratory exercises on the dissection of the sheep

brain, Golgi staining of the rat cerebellum, and

sen-sory coding in the cockroach. Twenty of the

ninety-three students in the course chose the EMG exercise

for the last laboratory session (they had 2 other

options: an exercise on hippocampal long-term

poten-FIG. 5.

Reaction time data fr om one subject. Supraspinal r esponse initiation time is better corr elated with r eaction time than is electr omechanical delay.r, Corr elation coefficient.

(5)

tiation or an exercise on hemispheric and gender

differences in frontal cortex EEG). The 20 students

were distributed over 3 laboratory sections. There

were usually two (and never more than 3) students

per setup. The exercise took approximately two

hours, including a brief lecture to explain the EMG.

The students were required to write a laboratory

report, the grade of which comprised 10% of their

total grade for the course.

Of the 20 students who performed the EMG exercise,

17 opted to evaluate it. These evaluations (on a scale

of 1 to 7) were done immediately at the end of the

session, before the students wrote up their results.

The lab was evaluated along three dimensions.

1 ) Enjoyment, with 1

5 not at all enjoyable and 7

5 very enjoyable (4

5 indifferent): mean

5 6.2,

range

5 5–7. Although this should not be the main

criterion by which a lab exercise is judged, students

obviously will be more involved in experiments that

they enjoy doing.

2 ) Learning from the exercise, with 1

5 I learned

nothing from doing this laboratory exercise and 7

5 I

learned a lot from doing this laboratory exercise (4

5 I

learned a moderate amount from doing this laboratory

exercise): mean

5 5.2, range

5 4–6. This is a

self-report measure of learning and thus is not as valid

as test scores, etc. However, it does indicate that the

students seemed to believe that the exercise was

worthwhile.

3 ) Whether the exercise should be continued as part

of the course, with 1

5 I think that this laboratory

exercise should definitely be dropped from the course

and 7

5 I think that this laboratory exercise should

definitely be a part of this course (4

5 I really have no

opinion either way): mean

5 5.8, range

5 5–7. This

was the first time that this exercise was done, and this

item was included as a way to gauge the students’

overall opinion of it and to give some indication of

whether it should remain as part of the laboratory

course.

There was also space on the evaluation form for

comments regarding the exercise. Of the 16 students

answering the question, ‘‘What did you like most

about today’s lab?,’’ eight included as at least part of

the answer that they enjoyed the use of human

subjects. These responses included ‘‘it was interesting

to use our own bodies,’’ ‘‘being hooked up to a

computer,’’ and ‘‘[as] student subjects we saw how

these motor systems functioned in ourselves (as

op-posed to cockroaches, rats, etc.).’’

DISCUSSION

One goal of this exercise was for the students to gain

an appreciation for the idea that the electrical

pro-cesses they have heard and read about in courses, and

perhaps observed in animal preparations, are

occur-ring in their own bodies. The student comments cited

previously demonstrated that they did enjoy the

opportunity to observe this activity. Another goal was

for the students to engage in active learning. One

student did indicate in the laboratory evaluation that

he or she ‘‘...enjoyed seeing a somewhat ‘practical’

FIG. 6.

(6)

application of the material learned in class,’’ and

another found it ‘‘...interesting to see the correlation

between the APs and movement.’’ A third goal was to

encourage critical thinking. As a method for

accom-plishing this, the students were required to write a

laboratory report on this exercise to describe and offer

explanations for their findings, which did require

them to think about the data. However, what could

have aided this would have been to provide a

work-sheet the week previous to the exercise (along with

the laboratory description) in which the students

would be required to predict the results and offer

explanations of the predictions. (They could be

al-lowed to modify these after the laboratory lecture, and

before doing the experiments, in the event that the

lecture clarified any poorly understood concepts).

The predictions could then be incorporated into the

lab reports.

The experiments are described here in a procedurally

logical order. However, for a postlaboratory

assign-ment (whether a list of questions to answer or a

laboratory report), it would be best if the results were

organized in a conceptual manner. Thus the

experi-ments on muscle tension and the temporal

relation-ship of the EMG to movement could be grouped

together because both investigate aspects of the

nature of the EMG. The flexion and extension

experi-ments could similarly be grouped in a postlaboratory

assignment.

Although this does work extremely well as a

labora-tory exercise, the instructor and one or two students

could do all or part of it as a demonstration, perhaps to

supplement another laboratory exercise.

This exercise is relatively inexpensive, requiring

mini-mal additional equipment over that found in a typical

neuroscience laboratory. The use of the percussion

hammer, designed for studying reflexes, to also

pro-duce the stimulus and trigger the oscilloscope for the

reaction time experiment eliminates the need for an

additional apparatus for this purpose. Furthermore,

there are no expenses for animal care because the

subjects take care of their own housing and feeding.

I thank Dr. Steven J. Zottoli for advice, Emile Ouellette for technical assistance, Felicity Adams for serving as a subject during develop-ment of this exercise, and the Introduction to Neuroscience students for participation and evaluations.

This work was supported by a grant from the Essel Foundation to Williams College.

This paper is based on a presentation given at the 26th Annual Meeting of the Society for Neuroscience, November 16–21, 1996, in Washington, DC.

Refer ences