Confin. neurol. 26: 57-75 (1965)
From the Neurosurgical Service of the Clinica de la Concepei6n, Madrid
(Dir.: Prof. S. Obrador)
By E. MARG and G. DIERSSEN
Introduction
What inquisitive mind has not asked itself how it works ? How do, for example, visual perceptions arise from the firing of neurons within the brain ? What sort of activity and organization must occur in the visual pathways of the brain to see a straight line, a color, a curve, a luminous field ? How many neurons must be active and in what temporal and spatial combinations to see a threshold point of light ? What are the basic or elementary units of visual perception and how arc they related to receptive fields ? Where, indeed, are the perceptual images developed ?
We have started on a path of investigation which we believe is taking us in the direction of the answers to these questions and others. Only a beginning has been made because the difficulties are formidable but we are confident that they can be overcome in time.
History
Although there is a legend that Fritsch stimulated the brain in dressing a war wound in 1864 [27], Bartholow [3], a Cincinnati surgeon, in 1874, was probably the first to stimulate electrically the brain of a conscious patient. Electrodes insulated to their tips were inserted through an open ulcer which extended through the parietal aura mater. This was followed by Ransom [39] in 1892 who stimulated the brain of an epileptic patient with a pair of electrodes pushed through the scalp and a previous trephine of the skull. Motor and sensory responses were obtained without anesthesia. Cashing [5] in 1909
5 Confinia Neurologica, Vol. 26, No. 2 (1965)
58 Marg,Dierssen, Reported Visual Percepts from Stimulation
showed in two patients that stimulation of the postcentral gyrus elicited sensations where the precentral gyrus gave motor responses as found in apes. This confirmed in man that the postcentral gyrus was a sensory area.
Foerster [13-15] and later Penfield [35] and his colleagues systematically mapped the sensory and motor responses to direct cortical stimulation, mainly during operations for epilepsy. Penfield and his co-workers, reported by Penfield and Rasmussen [36], stimulated the occipital cortex and obtained numerous visual responses. They will be discussed later.
The advent of human stereotaxic surgery by Spiegel and Wycis [40] and now widely used for a number of therapeutic and diagnostic procedures has made possible the incidental gathering of information from deep within the brain of conscious patients relatively quickly, easily, and under excellent physiological conditions. A number of investigators as well as one of us (G. D.) [1, 7-10, 17-19, 32, 34, 40, 41] have studied the effect of stimulation of brain structures as part of the diagnostic and therapeutic procedures used to aid patients requiring brain surgery for various disorders such as dystonia including Parkinson's disease and intention tremor, brain tumor, and epilepsy. All these surgeon-investigators shocked the brain with macroelectrodes, stimulating several cubic millimeters of tissue containing as many as a million neurons.
The finding of excitatory and inhibitory interplay between elements in receptive fields in the brain [26] made it even clearer than ever that gross stimulation of many thousands of neurons might be eliciting percepts as unnormal as they were complex, beclouding basic mechanisms. It became evident that a potentially fruitful approach was to stimulate as few neurons as possible for a percept in order to try to analyze elementary units of perception.
The development of microelectrodes makes limited and even single neuron stimulation possible. However, a new microelectrode had to be devised.for the human brain since existing forms did not have the requisite ruggedness, reliability, and could not be sterilized for neurosurgical procedures. One of us (E. M.) devised a microelectrode with these qualities for the purpose of stimulating the human brain [29] and we adapted it to the simplified, goniometric stereotaxic method of the other (G. D.) [11]. The initial results made possible by this union are reported here.
of the Human Brain with Mieroeleetrodes during Therapeutic Surgery 59
Limitations
We have already mentioned that there are formidable difficulties which Emit the quantity and to some extent the quality of our results. It is self-evident that absolutely nothing in the investigation be done that increases the usual hazards of the diagnostic and therapeutic surgery. Any deviation from this surgical program can be made only with the patient's knowledge and consent. The surgery cannot be unduly prolonged and the patient cannot be pressed too strongly for some of the details of a percept he may find difficult to describe. These limitations along with a limited number of patients and the difficulty in eliciting percepts account at this time for the relatively small sample of visual percepts. Hopefully, time will increase our sample and experience will make our techniques more effective. Despite these limitations, we have been able to obtain some relatively simple visual percepts from stimulating the visual pathways of the brain with microelectrodes. Analysis of these relatively simple percepts into disks, annul), and lines, all of various colors, have led us to consider them tentatively as elementary units of visual perception.
Methods
The mieroeleetrodes were usually made of 0.25 mm diameter tungsten wire, etched to a short taper and eeated with seven separately baked layers of an insulating varnish, Isonel 31 [29]. A few were 0.5 mm in diameter. The tip was generally opened mechanically by striking it a half dozen times against a rubber bottle stopper. At first all mieroeleetrodes were tested for their ability to pick up isolated unit potentials in a eat brain. After we learned the kind of responses elicited by stimulating the human brain, we generally tested subsequent electrodes by measuring their resistance in saline solution with an electrometer eireuit. The electrodes were about 20 em long and were soldered to a wire which led to a stimulator. White epoxy cement was used to mark off each centimeter (Fig. 1). Rectangular waves for stimulation were generated by a solid-state device designed and built for this investigation*. In addition, a Grass S4 Stimulator was used. The indifferent electrode was a 20 em square lead plate wrapped in gauze which had been wetted with saline and fitted around one leg. This electrode was used earner as the remote one for high frequency Coagulation of blood vessels at the site of entry. The electrode was continuously monitored with a Tektronix cathode ray oscilloscope and all stimulation potentials observed. The stimulus pulse duration was 1 msee and the frequency 400 eps, unless otherwise specified. Usually the electrode was made positive since negative current tended to erode the mierotip at the regmred voltages.
* The square wave generator was designed by Frederic W. Jenkinson of Biotronies, Ine., and built by Dr. Burton Chidden.
60 Marg, Dierssen. Renorted Visual Percepts from Stimulation
It was not possible to measure electrode current during the surgery. Laboratory measurements of these electrodes show a resistance of about 1 teraohm or more at the tip before it is opened, and an impedance of about 100 megohms to the 1 msee pulses at 400 eps. After the tip is opened, the resistance may drop to 100 megohms and the impedance to 10 megohms. At 10 volts the current is about 1 HA flowing durmg most of the pulse. When the tip is broken oflf or otherwise destroyed the resistance and impedance are the same, usuaUy between l/2 and 1 megehm. In the earlier cases, we often recorded on a small tape dictation machme the patient's description of visual responses. In the later eases, a two-channel tape recorder was used to record the complete verbal exchange between the patient and the surgeon on one channel and the stimulus on the other.
In some instances a dual electrode was used which consisted of two 0.25 mm diameter mieroeleetrodes cemented together with parallel shafts but in one of them the shaft was bared of insulation within 2 mm of the tip. This provided mieroeleetrode and maeroeleetrode tips within a 1/2 mm of each other.
The stereotaxie procedure is based on a goniometrie method which along with the simplified instrument (Fig. 1) has been described [11]. Some minor improvements have since been made in the instrument itself, including a self-threading screw which fits into the 1.5 cm diameter trephine or burr hole.
The electrode was inserted manually through the guide of the stereotaxie instrument in the direction of therapeutic interest (Fig. 1). Radiographs were taken in the
Fig. 1. A microelectrode is poised for
entry into the brain through the stereotaxic guide tube. The
electrode, which is marked off by white dots a cm apart, is 1/2
mm in diameter, although a 1/4 mm electrode was usually used. The
stereotaxic device is screwed into a burr hole in the skull
behind the surgical drapes.
of the Human Brain with Microelectrodes during Therapeutic Surgery 61
frontal and sagittal planes. The electrode could be clearly seen in the plates (Fig. 2) and allowed calculation of the correction for track orientation and depth with regard for X-ray distortion by a simple geometric method [37, 38].
While the X-ray plates were being developed, the electrode was used to stimulate the brain. The electrode tip locus was determined from the X-ray plates with an estimated average error of ~ 3 mm. The position was determined with the aid of a human stereotaxic atlas [42] and a special series of three-dimensional maps derived from it [12]. Because of the inherent error of the method, we could never be certain if the electrode tip lay, for example, in the lateral geniculate body or in the adjacent, incoming optic tract or outgoing radiation fibers, even when visual responses were elicited. In some cases the optic radiations were approached through an occipital burr hole and trocar track which was required for ventrienlography. Here the electrode tip was in the optic radiations, well removed from the geniculate and no such ambiguity existed. Location was most exactly determined in the thalamie area since it has been most carefully studied for current therapeutic stereotaxie procedures and the atlases are most detailed. In addition, the anatomic references most eommonly used are the anterior and posterior eommissures which grossly separate the thalamie structures from the subthalamie ones. These were visible after cerebrospinal
Fig. 1. X-ray photographs showing the
electrode in position in the brain. On the left is the frontal
view. On the right the sagittal view is seen with Talairach's
coordinates for human stereotaxic surger [42] superimposed. The
reference mark (black dot on the electrode) is in the region of
the lateral geniculate body.
62 Marg,Dierssen, Reported Visual Percepts from Stimulation
fluid was withdrawn and air injected in its place. In some eases, an iodine radioopaque oil * emulsified with eerebrospinal fluid was injected into the ventricles to make localization of the anatomic landmarks more certain [31].
Generally the patient was given no internal medication except for hdocaine 2% as a local anesthetic injected into the scalp in the area of entry. It was necessary for him to be alert for the therapeutic procedures and the hdocaine did not generally affect his attentiveness. He was asked to keep his eyes shut to provide a dark, uniform field.
The entire procedure except for the use of a fine mieroeleetrode in place of a gross maeroeleetrode is the same procedure used previously by one of us (G. D.) [7-10, 34] and is basically the same one widely used today [1, 17-19, 32, 40, 41]. The fibers of the internal capsule are found by stimulation so that they can be avoided when a therapeutic lesion is made in the adjacent lateral thalamus. The visual fibers are along the same electrode track.
The visual responses obtained were screened from the responses of 20 patients for this report; those which were considered likely to be spurious, suggested, or complex were rejected on the basis of five criteria. The rejected responses were: (1) not associated with the electrical stimulus; (2) not supported by the radiographic location of the electrode tip in anatomical primary visual pathways; (3) associated with nonvisual responses (paresthesias, seizures); (4) in the ipsilateral visual field; and (5) apparently suggested in the interrogation as determined from an examination of the transcript of the surgeon-patient dialogue.
At times we would withdraw the electrode lead plug from the stimulator and appear to continue the stimulation and interrogation as before. This was one of the ways we could determine how closely the response was associated with the electrical stimulus.
Anyone who has tried to demonstrate entoptie phenomena, especially "stopped images" such as Purkinje images or Hadinger brushes, even to intelligent and interested people knows the difficulty there is in seeing them for the first time. Once they are perceived, the observer knows what to look for henceforth. The same appears true of our entoptie "stopped images". It was necessary to instruct the patient just before stimulation and to make specific suggestions of simple forms which might appear such as Ones, lights, rectangles, ete. If the patient were asked if he saw anything without this orientation, he would be inclined to say no because he would expect a familiar and concrete picture such as a donkey grazing in a field.
No regular pattern of trains of stimulation was used. In the earner eases, we simply left the stimulus on steadily for many seconds until either a response was elicited or we moved on to the next stimulus parameter or new electrode position. In the later eases we tended to give the stimulus im irregular bursts, much as a doorbell is rung insistently.
Results
Patient A. M. R. With the mieroeleetrode in the region of the posterior and later the superior border of the lateral genienlate body, the patient reported seeing numerous yellow and green disks which moved slowly in the left visual field. The
* Manufactured as Lipiodol "F", Laboratories Andre Guerbet, Paris.
of the Human Bram with Mieroeleetrodes during Therapeutic Surgery 63
percept resulted from a series of stimulations from 10 to 1000 eps at -20 v. (This negative potential eroded the mierotip, as was later confirmed under the mieroseope, and undoubtedly injured some nerve fibers.) The exchange between the surgeon and the patient follows, except for the initial report of small yellow objects which was lost while the recorder was being turned on.
S. What size were they?
P. Very small.
S. And were there many?
P. Very many. They went passing by.
S. Good. And where did they pass to?
P. Well, one behind the other.
S. Were they lights or were they yellow spots?
P. Little yellow things.
S. But luminous?
P. Well no...
S. Were they circles?
P. I don't know.
S. Tell me if you see them now.
P. Now, no.
S. And now?
P. No.
S. How long did they last?
P. Well, they passed one behind the other.
S. All in a straight One ?
P. NO, helter-skelter.
S. And were they only yellow ?
P. Yellow, yes sir.
S. And where were they, there or on the ceiling ? (Gesturing.)
P. Yes there. They passed to here. (Gesturing.)
S. From below to above ?
P. No, like this. (Gesturing.)
S. AB if they were going from below upward around your head ?
P. Yes sir.
S. And you say that there were many ?
P. Very many.
S. Do you see the spots agam ?
P. No sir.
S. And now ?
P. Now I saw a little, yes.
S. Where?
P. There (Pointing.)
S. The same as before?
P. Fewer.
S. Fewer spots?
P. Fewer spots, yes.
Stimulating in the same region, 1/2 cm further out:
S. Do you see something ?
64 Marg, Dierssen, Reported Visual Percepts from Stimulation
P. No. . . yes. . . I saw the tattle spots again.
S. The same tattle spots as before ? What were they like ?
P. The same, of the same color, yellow and green.
S. Were they green before ?
P. They must have been green also.
S. And exactly where did you see these spots ?
P. Here. (Pointing.)
S. On the left side ?
P. Always in the same direction.
S. How big were the tattle spots ?
P. They were like a fingernail.
S. Where were they ? At what distance were they ?
P. Well I don't know. . . they passed by there. (Indicating a distance of 10 to 20 cm where the drab green surgical drape forms a hood over her head.)
S. How many were there ?
P. Very, very many.
S. And they passed very rapidly ?
P. No, no, slowly, slowly.
S. Were the tattle yellow spots mixed or arranged im order ?
P. They were all mixed.
S. And they were all the same size ?
P. Yes, I think so.
S. Do they oscillate ?
P. Yes. . . no. . . as they all go together.
S. Is the green of your spots the same as the green of the (surgical drape) cloth ?
P. Well yes, more or less.
S. It is not darker ?
P. No.
S. And the yellow, what was it like ?
P. Yellw, yellow.
S. Light or dark ?
P. Light, light.
Further stimulation in this locus gives no response. The electrode is shifted 2 em out.
S. Do you see something ?
P. I saw some tattle lights pass, little ones.
S. More here or there ? (Pointing.)
P. More toward the center.
S. And what color were they ?
P. All the same.
S. Of what color ?
P. Green and yellow.
S. They were exactly the same as before ?
P. The same, yes.
S. More toward the center ?
P. Yes sir.
S. From below upward ?
P. Yes, to there.
of the Human Brain with Microelectrodes during Therapeutic Surgery 65
After the next 1/2 cm shift of the electrode the patient started reporting somatosensations.
Patient J. M. L. A mieroeleetrode in the region of the lateral genieulate body gave a series of about 10 bright, more or less parallel, oscillating Ones. They were very thin and very long, at 60 em distance, and of a fire color (reddish-yellow or yellowish-red). In the center was a white, uniformly bright, circular disk, about 5 em diameter at lm (which subtends about 3°). This response came after a series of stimulations from 0.2 v to 30 v at 400 eps.
The patient said that he saw rays and the tape recorder was turned on to pick up the following dialogue:
S. Do you notice any movement ?
P. Yes, yes sir.
S. And what are these movements Eke ? Do you see rays ?
P. Yes, rays.
S. And these rays, are they horizontal or vertical or oblique ?
P. They are vertical.
S. And do these rays move ?
P. Yes, yes, not in the center, there is a light.
S. Close your eyes tightly. Do you still see them ?
P. Yes.
S. Open your eyes. Do you see them now ?
P. No.
S. Were the lines thin or thick ?
P. Thin.
S. How thin ? Did they vary ? How many ?
P. About 10.
S. How fine were they ? Like a finger ?
P. Yes, less, very fine.
S. At what distance did you see them ?
P. At about 60 em or SO. (Indicates about 100 em.)
The patient now indicates that the Ones were oblique, from upper left to lower right.
S. How did they move ? Did the lines oscillate, or what ?
P. Yes, they oscillated.
S. They were very thin. And how long were they ?
P. They were very thin and very long.
S. And the light which was in the center, what was it Eke ?
P. Like a light bulb.
S. But was it uniform or different ?
P. It was uniform.
S. Was it white ?
P. White, yes.
S. And the rays were also white ?
P. No, no, as I said they tended to look like fire.
S. Was the light round or square ?
P. It was round.
66 Marg,Dierssen, Reported Visual Percepts from Stimulation
S. And all uniform ?
P. Yes, yes.
S. There were no differences at all ?
P. No.
S. How large was the light ?
P. Like this. (Indicating 5 cm.)
S. At what distance ? At the same distance ?
P. At the same distance.
S. Are you sure ?
P. I think so.
S. Let's see. Close your eyes. Do you see them again ?
P. No.
The following day the patient was asked to draw the percept which is reproduced in Fig. 3. As is apparent from the illustration, he now insisted that the slope was at 45° rather than at 90° or 135° as he previously described it. He reiterated that the lines were reddish-yellow, the disk was uniformly white and 3 to 4 cm in diameter at 70 cm distance (subtending an angle of 3°). The lines were as thin as pencil fumes and the ends could not be seen. He could neither describe the oscillation nor say if the lines were equispaced.
Later inspection of the mieroeleetrode showed that the tip had broken off and the resistance was down to 3 megohms from its original value of more than 100 megohms.
Patient E. G. A stimulus of +40 v through the mierotip in the optic radiations was the basis of the following dialogue between the patient and the surgeon:
P. I see some rays of light.
S. Where ?
Fig. 3. A drawing of his percept by patient
J.M.L. The lines were reported as very thin and long, and of a
reddish-yellow color. They oscillated normal to their axes. The
central disk was uniformly a bright white.
of the Human Brain with Mieroeleetrodes during Therapeutic Surgery 67
P. In front.
S. What are these rays like ? Like lines ?
P. No.
S. Of what color ?
P. Well, blue.
S. And how were they ? Like this or this ? (S. indicating horizontal and vertical with his arm.)
P. Well, hke eireumferenees.
S. Like tattle eireles, round ?
P. Yes, yes.
S. And what size are the circles ?
P. Well, Eke the pupils of an eye. (Sie.)
S. And at what distance do you see them ?
P. At about 20 to 30 em.
S. Were they Eke hoops or like blue disks ?
P. More Eke hoops.
S. Like blue hoops ?
P. Yes, sir.
A few minutes later the question was asked:
S. How many were there, many ?
P. Many.
S. Twenty ?
P. Fewer.
S. Nine ?
P. Something like that.
The annuli were reported as movmg "very rapidly", were all at the same distance, and appeared for a "very short time".
Further stimulation in this and other loci yielded no response.
Summary of Results. In three patients, out of a total of 20, the following visual percepts were reported from mieroeleetrode stimulation: (1) yellow and green disks; (2) bright yellowish-red oblique lines with a white disk; and (3) blue annul) (about nine).
The form elements elicited by stimulation were Ones, disks, and annul). The colors reported were white, yellow, green, yellowish-red, and blue.
Comment
Penfield arid Rasmussen, as mentioned earlier, stimulated the occipital cortex, using an electrode 3 mm between poles and with sawtooth waves of up to 5 v at 60 cycles. Descriptions of the responses include ". . . a brilliant ball, a star, a streak, a wheel, a spot or a flash, a shadow, a light. Visual movement was more frequently present than absent. The image might remain still but more often it moved slowly in a certain direction or it danced, flickered, or whirled." They
68 Marg, Dierssen, Reported Visual Percepts from Stimulation
comment, "These are the basic elements of vision, no doubt, but they bear little resemblance to the images of things seen that ordinarily reach consciousness."
Even when one of their stimuli did reach consciousness there was a likelihood from the grossness of their electrode that very many neurons were being stimulated. The percepts were relatively simple when compared with percepts of common experience. But they may not have been generally elementary in the sense of being indivisible into simpler elements. Still it is surprising how often their responses appear to resemble those elicited by our microelectrodes. These include a small red mark (red disk), long white mark (white line), big white spots (white disks), blue, green, and red-colored disks, red and blue wheels (annul)), quivering and dancing (oscillating) lights, bright lights, fiery or red (yellowish-red). As yet we have not encountered their reported diamonds, stars, fawn lights, flashing, moving shadow s, brown squares, radiating spots or spokes, black and white things (after cortical ablation), or glittering silvery and gold things. The difference can be attributed to the electrode and stimuli, the specific areas of the visual structures stimulated, and perhaps also to limited sampling.
There seems to be a difference between the response of visual percepts in our microelectrode stimulation and the macroelectrode stimulation reported by others in the visual pathways. Our microstimulation elicited few percepts, often the same but thus far of no more than two of a limited number of types at a time. These are disks, annul), and lines of various colors. Macrostimulation elicits countless images of sparks, lightning, and similar widespread and complex images. This is apparently the difference between stimulating many thousands of neurons with the macroelectrode [34] and stimulating relatively few with the microelectrode.
The patient's eyes were usually closed during stimulation, providing a dark and uniform visual field, free from the interference of other visual objects. Thus it is hard to account for the difficulty in eliciting a visual sensation by stimulation of the pathways of the brain. This may be concerned with the complex coding and interaction between different channels generated by the retina producing overlap redundancy for the development of a visual percept. Retinal phosphenes are common and easy to produce, possibly because the retina encodes the signals for the percept, but optic nerve phosphenes are not so readily generated. The apparently simpler somatosensory system in
of the Human Brain with Mieroeleetrodes during Therapeutic Surgery 69
which sensations are more readily elicitable and are generally repeatable over and over again, accepts more readily the result of macroor microstimulation for percepts [30], but much more often than not the visual perceptual mechanism rejects the result of our stimulations and the patient sees nothing. It is so seldom that a response is obtained that after numerous stimulations with the microelectrode, the impression is gained that only in rare instances the visual system can be fooled into a percept with this kind of artificial stimulation. On many occasions we were almost certain that our electrode tip was in the visual pathways without obtaining any response. The lack of repeatable responses (even after the microtip is shifted a few mm) might indicate either that the probability of triggering a perceptual assembly of neurons is very small or possibly that the system learns to suppress these unnatural or unreasonable stimuli quickly.
Since our stimulations are apparently excitatory, we do not have any direct information on what is the perceptual response to inhibition of the spontaneous firing of neurons in the visual pathways. One view is that excitation is potentially seen and inhibition is not. Accordingly, an excitatory center could be seen but not its inhibitory surround. Similarly, an excitatory surround could be seen but not its inhibitory center. Another view is that only the center can be potentially seen, and the surround merely modifies it. The report of annul) by one of our subjects inclines us to the first view.
The appearance of white and the lack of black in our responses point to black being the inhibitory aspect of excitatory white. Similarly, from what has been learned in single unit studies in monkeys, rabbits, and fish [2, 6, 16, 20-25, 28, 33, 43, 44], inhibition of a neuron eliciting a specific excitatory color might be expected to elicit its opponent hue.
It has been demonstrated that some neurons of the visual pathway signal only when the retinal stimulus is a moving light contour [23, 24]. Thus the stimulation of such a neuron alone may indicate movement, which not being supported by other signals of further movement could be interpreted as a number of rapid, short movements from a given point, or an oscillation. When there is no serious conflict, stimulation of these neurons which signal movement may give abstract movement, without any part of the image actually moving, as in the waterfall illusion. This may account for the reported movement of the yellow and green disks. Large movements may be signaled by movement of the eyes, much as an after-image fixed on the retina
70 Marg, Dierssen, Reported Visual Percepts from Stimulation
appears to move when the eyes are turned. There is usually a strong desire to turn the eyes toward percepts in the peripheral field which our stimuli effectively elicit.
It is possible that each neuron represents an elementary unit of perception. However, the response of the patient who saw the same yellow and green disks before and after a change of electrode position, although the disks were shifted in position in the visual field, argues against this view. It is curious that several of each form (the lines, disks, and annah) have not been reported as overlapping at times. There are no doubt complex perceptual interactions.
A reduction in the rate of spontaneous activity can effectively occur by virtue of the sudden injury or destruction of neurons. We have little evidence that this signal, which may be a result of the injury or destruction of some neurons as a result of the penetration of the electrode or possibly strong stimulation, gives rise to a percept. The lack of parallel and redundant information over other neurons may prevent "injury inhibition" from reaching consciousness. Recording of the visual fields by an ophthalmologist in a number of patients before and after surgery showed, as expected, no change. Nor indeed have any functional deficits or changes occurred after repeated somastosensory stimulation. But this does not exclude the injury of a few fibers which would not be evident in any functional test. Repeatable responses from a microelectrode in the somatosensory system of the brain, as previously mentioned, also indicate any injury to neurons which might result from our stimulation is not extensive. In fact if the brain of a rabbit is stimulated at the usual parameters, 400 cps and 1 msec, and maintained for 5 minutes at +40 v, Nissl sections show no lesions (nor are bubbles produced) with an intact electrode. The current is about 1 PA which is below but near that reported used for making minimal lesions [25]. A broken tip may pass a current several times higher and may be presumed to do some damage although the current density may actually be lower.
It is well known to those who record single units from the mammalian brain that on occasion an electrode tip may precipitate mechanically a high-frequency burst from a cell which rapidly decreases its frequency, then disappears and presumably dies. It might be expected that such a burst would elicit a visual percept; yet we have had only one report which might be attributed to such an origin. However, this response was rejected for our series reported here because it did not
of the Human Brain with Mieroeleetrodes during Therapeutic Surgery 71
meet the first criterion, it was not associated with the electrical stimulation.
It is difficult to convey the high degree of reliability we believe the reported responses have. Through our first four criteria we have tried to eliminate complex and spurious responses. We have had data rejected by one or more of them where the patient reported a bridge over the Rio Manzanares, three balconies, his sister's bedroom, fingersized lines, a flower, a butterfly or a mushroom, a large luminous area, etc. Subsequent to this screening, the fifth criterion was used to reject otherwise apparently valid responses after a study of the tape transcript of two patients revealed that some responses might have been induced by the interrogation. We have also rejected otherwise apparently and probably valid data of disks with surrounds and also thin horizontal lines from two of our earlier cases because we had no tape recordings to analyze for possible suggestion. In order to give some indication of the spontaneity of the responses, they were included verbatim in the results. A further indication of the general validity of responses was in the patient's responses to somatosensory stimulation, of which we have hundreds for comparison [30].
It is especially interesting to compare our perceptual responses in man with physiological receptive field responses in the monkey which have been studied by Hubel arid Diesel [22-25, 44]. They have shown that receptive fields of neurons in the lateral geniculate body of the rhesus monkey fall into three main classes:
(1) Center and surround, one excitatory and the other inhibitory, have the identical response to wavelength, as in the cat. (Perceptually these could represent white disks and white annuli.)
(2) Center and surround have different wavelength sensitivity with two of three possible maxima, 580, 540 and 450 my. (According to Burnham et al. [4], typical hue names associated with these wavelengths are reddish-yellow, yellowish-green, and reddish-blue, respectively. Perceptually these could represent disks and annuli of these colors.)
(3) The opposing wavelengths have an identical spatial distribution instead of a center-surround configuration. (Perceptually these could represent disks of the same colors.)
In addition, linear receptive fields have been found in the striate cortex.
Despite the species difference, our tentative elementary visual
72 Marg, Dierssen, Reported Visual Pereepts from Stimulation
perceptual units in man show striking similarity to the visual receptive fields in monkey of Hubel and Wiesel. It raises the question if, indeed, they are not fundamentally identical. If they are, it points to perception taking place for these simple elements at the highest level at which these elements generally occur, that is, in the first neurons of the striate cortex where both circular and linear receptive fields are found. If, however, the lines were elicited by a stimulation of a fortuitous combination of what individually would be disks and annul), then the perception of the hnes may be cortical but the circular elements subcortical. In fact, the perception of multiple parallel lines resembles the organization of a cortical column of receptive fields.
We plan to collect additional reports of visual percepts elicited by microstimulation within the visual pathway. In time we should have enough data to modify our tentative schema to a more exact description of the relation between the neurons and their percepts.
Summary
Stimulation of points in the primary visual pathways with a microelectrode was done incidental to therapeutic and diagnostic brain surgery for Parkinson's disease, intention tremor and other dystonias, as well as brain tumor and epilepsy. The primary consideration of the welfare of the patient limits the quantity and the quality of the data. Responses were reported which included disks, annul), and lines, all in various colors including white. These may be elementary units of visual perception.
Zusammenfassung
Punkte der primaren optischen Bahnen wurden mit einer MikroElektrode gereizt wahrend therapeutischer und diagnostischer Hirnchirurgie fur Parkinson Erkrankung, Intentionstremor und andere Dystonien, wie auch Hirntumor und Epilepsie. Die primare Berucksichtigung des Wohlergehens des Patienten begrenzt die Quantitat und Qualitat der Beobachtungen. Die Reaktionen waren Scheiben, Ringe und Linien in verschiedenen Farben, einschliel~lich weil3. Dies durften elementare Einheiten vi-queller Wahrnehmung darstellen.
Resume
Les voies optiques primaires ont ete stimulees a l'aide d'une microelectrode, lors d'une intervention chirurgicale, pour des raisons thera-
of the Human Brain with Mieroeleetrodes during Therapeutie Surgery 73
peutiques ou diagnostiques dans des cas de maladie de Parkinson, de tremblement intentionnel, de dystonies, de tumeur cerebrale et d'epilepsie. Prenant en consideration la sante du malade, on a limite le nombre et l'importance des stimulations. On a neanmoins pu constater que les patients apercevaient des disques, des lignes et des anneaux colores ou blancs. Ceux-ci pourraient done constituer des unites elementaires de perception visuelle.
Resumen
La estimulacion de las vies visuales primerias fue producida con microelectrodos en el curve de intervenciones estereotaxicas. Las respuestas obtenidas incluyen discos, anillos y lineas de varios colores incluyendo el blanco. Pudiera tratarse de unidades elementales del sistema visual.
This investigation was supported by a grant from the U.S. National Seienee Foundation and a fellowship from the John Simon Guggenheim Foundation (to E. M.) to the College de Franee, Laboratoire de Neurophysiologie Generale, Station de l'Institut Marey, Paris 16e. It is a pleasure to thank the following colleagues for their helpful discussion: A. Fessard, H. Barlow, W. A. M. Rwiton, D. Hubel, G. Horn and W. Levick.
References
1. Alberts, W. W.; Wright, E. W.; Levin, G.; Feinstein, B. and Mueller, A. B.: Threshold stimulation of the lateral thalamus and globus pallidus in the waking human. Eleetroeneeph. elin. Neurophysiol. 13: 68 (1961).
2. Andersen, V. O.; Buchmann, B. and Lennox-Buchthal, M. A.: Single cortical units with narrow spectral sensitivity in monkey (Cerocebus Torquatus Atys). Vision Res. 2:295 (1962).
3. Bartholow, R.: Experimental investigations into the functions of the human brain. Amer. J. med. Sci. 67: 305 (1874).
4. Burnham, R. W.; Hanes, R. M. and Bartheson, C. J.: Color: A guide to basie facts and eoneepts (Wiley, New York 1963).
5. Cushing, H.: A note upon the faradic stimulation of the postcentral gyrus in conscious patients. Brain 32: 44 (1909).
6. De Valois, R. L.: Color vision mechanisms in the monkey. J. gen. Physiol. 43: 115 (1960).
7. Dierssen, G.; Sevillano, M.; Carrascosa, R. and Obrador, S.: Int. symp. physiology and neurosurgery of the extrapyramidal system, 1961 (Pa~ Montalvo, Madrid, to be published).
8. Dierssen, G.: Neurocirurgia Luso-Espanola 11: 121 (1963).
9. Dierssen, G. and Sanches Moreno, Y.: Algunas observaciones sobre defeeto censorial producido por lesiones en la radiaci6n censorial talamocortical. Rev. chary. esp. 91: 712 (1963).
6 Confinia Neurologich, Vol. 26, No. 2 (1965)
74 Marg, Dierssen, Reported Visual Percepts from Stimulation
10. Dierssen, G. and Sanchez Moreno,Y.: Int. Symp. Pain, 1962 (Paz Montalvo, Madrid, to be published).
11. Dierssen, G.: Un sistema simplificado pare el abordaje estereotaxico de estructuras subcorticales. Acta neural. Iatino-amer. 4: 366 (1959).
12. Dierssen, G.: In preparation.
13. Foerster, O.: The motor cortex in man in the light of Hughlings Jackson's doctrines. Brain 59: 135 (1936).
14. Foerster, 0.: Symptomatologie der Erkrankungen des GroBhirns. I. Motorische Felder und Bahnen; in: Bundle und Foersters Handb. d. Neurol. 6: 1 (1936).
15. Foerster, 0.: Symptomatologie der Erkrankungen des GroBhirns. II. Sensibelcorticale Felder; in Bumie und Foersters Handb. d. Neurol. 6: 358 (1936).
16. Granit, R.: Receptors and sensory perception (Yale University Press, New Haven 1955).
17. Gulot, G.; Sachs, M.; Hertzag, E.; Brian, S.; Rougerie, J.; Dallas, J. C. et Napoleone, F.: Stimulation electrique et lesions chirurgicales de la capsule interne. Neuro-chir., Paris 5: 17 (1959).
18. Guiot, G.; Hertzag, E.; Rondot, P. and Molina, P.: Arrest or acceleration of speech evoked by thalamic stimulation in the course of stereotaxic procedures for parkinsonism. Brain 84: 363 (1961).
19. Hassler, R. und Riechert, T.: Klinische Effekte bei Reizung einzelner Thalamuskerne am Menschen. Nervenarst 26: 35 (1955).
20. Hill, R. M.: Unit responses of the rabbit lateral geniculate nucleus to monochromatic light on the retina. Science 135: 98 (1962).
21. HiU, R. M. and Marg, E.: Single cell responses of the nucleus of the transpedunenlar tract in rabbit to monochromatic light on the retina. J. Neurophysiol. 26: 249 (1963).
22. Hubel, D. H. and Diesel, T. N.: Responses of monkey genieulate cells to monoehromatie and white spots of light. Physiologist 7: 162 (1964).
23. Hubel, D. H. and Diesel, T. N.: Receptive Gelds of optic nerve fibers in the spider monkey. J. Physiol., Lond. 154: 572 (1960).
24. Hubel, D. H. and Diesel, T. N.: Receptive fields, binocular interaction and functional architecture in the cat's visual cortex. J. Physiol., Lond. 160: 106 (1962).
25. Hubel, D. H. and Diesel, T. N.: Shape and arrangement of columns in cat's striate cortex. J. Physiol., Lond. 165: 559 (1963).
26. Ruffler, S. W.: Organization of mammalian retina. J. Neurophysiol. 16: 37 (1953).
27. Aunts, A.: Foundations of neurology, p. 140 (Thomas, Springfield 1953).
28. MacNichol, E. F., Jr. and Svaetichin, G.: Electric responses from the isolated retinas of fishes. Amer. J. OpEthal. 46: 26 (1958).
29. Marg, E.: A rugged rehable and sterilizable microelectrode for recording single units from the brain. Nature, Land. 202: 601 (1964).
30. Marg, E. and Dierssen, G.: In preparation.
31. Matsubara, T. and Nomura, T.: Emulsified iodized oil ventriculography. Amer. J. Roentgenol. 84: 48 (1960).
32. Monnier, M.: Contributions experimentales a la physiologic du bone cerebral chez l'homme. I. Technique de reperage, stimulation et coagulation des structures sous-corticales. Helv. physiol. pharmacol. Acta 8: 54 (1950).
of the Human Brain with Microelectrodes during Therapeutic Surgery 75
33. Motokawa, K.; Taira, N. and Okuda, J.: Spectral responses of single units in the visual cortex. Tohoku J. exp. Med. 78: 320 (1962).
34. Obrador, S. and Dierssen, G.: Communicaci6n al Congr. Latineam. de Neurocirugia, 1963 (Buenos Aires, in press).
35. Penfield, W.: The cerebral cortex in man: I. The cerebral cortex and consciousness. Arch. Neurol., Chicago 40: 417 (1938).
36. Penfield, W. and Rasmussen, T.: The cerebral cortex of man (MacmiUan, New York 1950).
37. Queimadelos, V. C.: Communicaci6n a la Soc. Espanola de Neurol., Barcelona (1963).
38. Queimadelos, V. C.: Radiologic distortion in stereotactic neurosurgery; in:Handbook of neuroradiology (foray, Barcelona, in press).
39. Ransom, W.: A case illustrating kinaesthesis. Brain 15: 437 (1892).
40. Spiegel, E. A. and Wycis, H. T.: Stereaencephalotomy, Part I (Grune and Stratton, New York 1952).
41. Talairach, J.; Hecasn, H.; David, M.; Monnier, M. et De Ajuriaguerra, J.: Recherches sur la coagulation therapeutique des structures sous-corticales chez l'homme. Rev. neural. 81: 4 (1949).
42. Talairach, J.; David, M.; Tournoux, P.; Corredor, H. et Kuasina, T.: Atlas d'anatomie stereotaxique (Masson, Paris 1958).
43. Wagner, H. G.; MacNichol, E. F.; and Wolbarsht, M. L.: The response properties of single ganglion cells in the goldfish retina. J. gen. Physiol. 43: 6, 45 (1960).
44. Wiesel, T. N. and Hubel, D. H.: Receptive fields of monkey geniculate ceUs in the dark adapted state. Physiologist 7: 287 (1964).
Authors' addresses: Dr. E. Marg, Brain Section, Neurosensory Laboratory, School of Optometry,
University of California, Berkeley, Calif. (USA)
Dr. G. Dieresen, Neurological Service, Clinica de la Concepci6n, Ciudad Universitaria,
Madrid (Spain)