toward a theory of painelevation in the electroneurogram." if all cutaneous stimuli activate...

PSYCHOLOGICAL BULLETINVOL. 56, No. 6, 1959

TOWARD A THEORY OF PAIN:RELIEF OF CHRONIC PAIN BY PREFRONTAL LEUCOTOMY,

OPIATES, PLACEBOS, AND HYPNOSIS

THEODORE XENOPHON BARBER1

Department of Social Relations, Harvard University

The response to a nociceptive stim-ulus normally includes at least fourcomponents: "the sensation of pain";discomfort; withdrawal movements;and some measurable physiologicalalteration, e.g., a transient or pro-longed increase or decrease in bloodpressure (Nafe & Wagoner, 1938;Goetzl, Bien, & Lu, 1951). Thispaper is concerned with the neuro-logical correlates of this total re-sponse—hereafter termed the pain re-sponse—and how this total responseor some components of this responsecan be mitigated or eliminated byprefrontal leucotomy, opiates, place-bos, and hypnosis.

THE NEUROPHYSIOLOGICAL COR-RELATES OF THE PAIN RESPONSE

Free Nerve Endings: The So-Called"Pain Receptors"

It has generally been assumed thatthe free nerve endings, which arefound widely scattered near the cu-taneous and visceral surfaces, are thespecific receptors for noxious stimuli.However, Sinclair, Weddell, andZander (1952) have shown that 5scan discriminate cold, heat, touch,

1 Postdoctoral research fellow, NationalInstitute of Mental Health, Public HealthService. It is a pleasure to thank E. G. Boring,Asenath Petrie, H. K. Beecher, and D. R.Evans for the many hours they spent criti-cally reading this manuscript. Although theirvaluable advice and suggestions led to thecorrection of many errors, responsibility forthe remaining . faults and the opinions ex-pressed remain with the writer. At presentthe writer is a research associate at the Wor-cester Foundation for Experimental Biologyand Medfield (Massachusetts) State Hospital.

430

and prick just as well from the earpinna, which contains only barenerve endings and a basketlike net-work around the hair follicles, as theycan from the skin of the forearm,which contains all of the encapsu-lated endings which have been de-scribed. Lele, Weddell, and Williams(1954) have demonstrated that thefree nerve endings in the skin, whensuitably stimulated, "give rise to awide range of sensory experiencewhich includes reports of 'cold,''touch,' 'warm,' 'prick,' 'itch,' and'sharp pain.' " Lele and Weddell(1956) have confirmed earlier findings(e.g., Nafe & Wagoner, 1936) that 5sreport not only pain but also touch,warmth, and cold when appropriatestimuli are applied to the center ofthe cornea which contains only freeendings. These and other investiga-tions recently reviewed by Weddell(1955) and Sinclair (1955) indicatethat a wide variety of sensory experi-ences can be evoked by suitable stim-ulation of the free nerve terminalsand that theories of cutaneous sensi-bility postulating specific receptorsfor each sense modality are open toserious objection at the present time.3

Peripheral ConductionA series of earlier investigations,

reviewed by Bonica (1953, pp. 29-

' The possibility remains that the term freenerve endings does not refer to homogeneousunits. If future investigations demonstratespecific biochemical differences between theseendings, the question of specific receptors forthe various sense modalities may require fur-ther investigations at the molecular level.

RELIEF OF CHRONIC PAIN 431

30), had apparently shown (a) thatasphyxia or pressure applied to a per-ipheral nerve blocked the large, my-elinated, fast-conducting A fibersfirst and abolished touch before painand (b) that cocaine blocked thesmall, unmyelinated, slow-conduct-ing C fibers first and abolished painbefore touch. This appeared to besatisfactory evidence that touch iscorrelated with conduction in thelarger A fibers and that pain is cor-related with conduction in the smallC fibers.* Other data, however, indi-cate that noxious stimuli applied tothe cutaneous surface activate many,if not all, of the fiber types presentin the cutaneous nerves, viz., thesmaller A fibers and the C fibers.Since the large A fibers are presentonly in the muscle branches of thenerves and the Group B fibers consistof sympathetic preganglionic axones(Lloyd, 1955; Ruch, 1955, p. 334),

* At the present time, extreme caution isnecessary in drawing conclusions from theearlier experiments on reversible nerve blocksproduced by asphyxia, compression, cocaine,etc. (Jones: 1956, 1958; Schiller, 1956). In aseries of carefully controlled studies of nerveblocks produced by procaine, compression,and cooling, Sinclair and Hinshaw (Sinclair,1955) have demonstrated that it is possibleto obtain almost any order of sensory loss byusing different 5s, by varying the site stimu-lated, and by altering the nature of the stimu-lus.

The question of "double pain" and its re-lation to conduction in A-delta and C fibershas also been opened for further inquiry."Double pain" may be due to inadequate con-trol of the stimulus at the receptor level:Jones (1956) demonstrated that (physio-logically normal) Ss do not report double painif the stimulus is prevented from stimulating thesame receptors more than once. Sinclair (1955,p. 594) has also concluded from his own workand from earlier investigations in this areathat "the question of second pain cannot beregarded as settled and the idea of two sets ofpain fibres rests upon work which is not im-mune to criticism of the experimental findingsas well as the interpretations placed uponthem."

they cannot be directly activated bycutaneous stimuli. Heinbecker,Bishop, and O'Leary (1933) demon-strated with a human subject thatelectric shock applied to an exposednerve in such a manner as to stimu-late the A-delta fibers consistentlyevoked a pain response. Zotterman(1939) reported that a burning stim-ulus applied to the skin of cat evokeda spike composition that includedboth the A-delta and C fibers. Brook-hart, Livingston, and Haugen (1953)demonstrated that stimulation of thetooth pulp (which normally evokes apain response) yields conductioncharacteristics of the A gamma-deltafibers. After reviewing these andother investigations attempting torelate the modalities of sensation toconduction in specific fibers, Living-ston (1943), Bonica (1953), Sinclair(1955), and Schiller (1956) agree withGasser (1943, p. 59) that "the fibersbelonging to different modalitiesmust be widely distributed through-out the various fiber sizes, and thatthere seems to be little possibility ofassociating any one sensation with anelevation in the electroneurogram."

If all cutaneous stimuli activate"fibers widely distributed throughoutthe various fiber sizes," what deter-mines the differential response toeach stimulus? To account for thisdifferential response, investigators inthis area (Bishop, 1946; Weddell,1955; Sinclair, 1955) hypothesizethat each cutaneous stimulus sets offa pattern of nerve impulses whichdiffers from the pattern set off byother stimuli in that the relative num-ber of activated fibers of various sizesdiffer, and the relatively differentsizes carry impulses differing in en-ergy value, frequency, and duration.A light touch, for example, preferen-tially activates the larger fibers.However, this does not mean thatthese large fibers are specific to light

432 THEODORE X. BARBER

touch stimuli. The larger the fiberthe lower its threshold. The slightdisturbance caused by light touch,therefore, activates the largest fiberswith the lowest threshold more readilythan the smaller fibers with higherthresholds. Similarly, nociceptivestimuli applied to the cutaneous sur-face may more readily activate fibersin the C range, but this does notmean that these stimuli do not alsoactivate other cutaneous fibers andit also does not mean that otherstimuli cannot activate the C fibers.

Conduction in the Spinal CordPeripheral nerve fibers, which are

both myelinated and unmyelinatedand which may belong to either ClassA or C, travel along the cranial,spinal, or sympathetic nerves toposterior root ganglia where theysynapse with second order neurons.The generally accepted view is thatnoxious stimuli applied to the visceraand to subcutaneous and cutaneousstructures activate those fibers in thespinal cord which cross through theanterior white commissure to the op-posite lateral funiculus where theyascend cephalad, forming the lateralspinothalamic tract. However, thisis by no means the complete story.True enough, in many cases, cuttingthe lateral spinothalamic tract pre-vents a response to many nociceptivestimuli applied to the contralateralside. However, in some cases, thisloss is temporary; normal pain re-sponsiveness may return after an in-tervening period (Ranson, 1943, p.111). Also, this operation (antero-lateral cordotomy) does not abolishpain discrimination while leavingtemperature, touch, and pressure dis-crimination intact. As Schiller (1956,p. 208) points out, "One modality ortwo are never either completelyspared or abolished to the absolute

exclusion of the others . . . parts de-nervated by anterolateral cordot-omy are reported as feeling 'numb,'discrimination of two points and tex-ture of materials is diminished, and,in addition, there are thermanes-thesia and analgesia." In addition,White and Sweet (1955, p. 45) dem-onstrated that a current at 100 ormore volts applied to the "analgesicside" invariably produced a report ofsevere pain in all (40) patients ex-amined. King (1957) confirmed thisfinding and, in addition, found that(after anterolateral cordotomy) themaximum elevation of the painthreshold on the "analgesic side" didnot exceed 40 to 50%. Since, in thegreat majority of cases, the painthreshold elevation was much lessthan this maximum, and, in somecases, was not significantly differentfrom the pain threshold on the nor-mal side, King concluded that "apolysynaptic relay pathway for pain-ful stimuli in man, aside from thespinothalamic system, seems prob-able."

Not only does anterolateral cord-otomy consistently fail to abolish theresponse to more intense noxiousstimuli, it also fails to affect the painresponse to pinpricks in the majorityof cases. White and Sweet (1955, p.262) found that, after this operation,60% of their patients consistently re-ported pain when multiple rapid pin-pricks were applied to the "analgesicside."

French and Peyton (1948), Voris(1951), and White and Sweet (1955,p. 45) have presented additional evi-dence indicating that nociceptivestimuli activate fibers that do notcross to the opposite side in the spinalcord. Each of these investigators hasreported cases in which the "anal-gesia" was ipsilateral following anter-olateral cordotomy. From these re-


ports White and Sweet (1955, p. 275)conclude that the "pain fibres arediffused over a very wide area of theanterolateral quadrant, and that attimes some centrally conductingfibres must run upwards in the ipsi-lateral as well as in the contralateralcolumns, at least for a considerabledistance."

Even the above account is incom-plete; nociceptive stimuli can acti-vate far more units in the cord thanthose found on both sides of the an-terolateral quadrant. Livingston(1943) and White and Sweet (1955)have found that in many cases bilat-eral anterior cordotomy is insufficientto relieve a pain syndrome andLhermitte and Puech '1946), Pool(1946), and Browder and Gallagher(1948) have demonstrated that somepain syndromes can be relieved byposterior cordotomy. Keele (1957, p.164) has reviewed additional evi-dence which indicates that the "paintracts" may be widely dispersed inthe cord and concludes that "one isinduced to look upon their anatomyas one of statistical probability, andto wonder how closely or how perma-nently the function of transmissionof pain sense is attached to fixedneuronal paths in the cord." Aftersummarizing the evidence in thisarea, Adey (1957) similarly concludesthat the concept of localized fiberpathways in the spinal cord carryingparticular types of sensory impulsesis open to serious question and re-quires revision.

A number of investigators haveinterpreted their data as indicatingthat the same stimulus may activatedifferent neural units in the cord atat different times. Gasser (1937)writes that "a given stream of affer-ent impulses over a peripheral nervefollows one pathway in the centersat one time and another pathway at

another time. The direction of theswitching is conditioned by the situa-tion obtaining at the moment, and isalways consonant with a coordinatedreaction of the whole organism."Along similar lines, Livingston (1943,p. 25) interprets the evidence as indi-cating that "impulses, finding them-selves blocked from their customarypathways, eventually find new or pre-viously unused pathways." Bishop(1944) likewise suggests that whenimpulses along a neural pathwayreach a certain critical frequency,they are "switched" to different con-duction units from those into whichthey normally pass.

A number of other considerationsshould be emphasized. First of all,there is no need to hypothesize spe-cific pathways for pain and othermodalities to understand the altera-tions in sensibility which follow an-terolateral cordotomy. As Sinclair(1955, p. 606) has pointed out, "In-stead of cutting specific fibres, wemay be so altering the sensory pat-terns the spinal cord is capable ofconducting in such a way as to leadto a sensory dissociation." Further-more, even if we assume a "segrega-tion" of "pain-conducting fibers" atthe cord level, we cannot relate this"segregation" to the "sensation ofpain," to discomfort or suffering, orto other components of the pain re-sponse which appear to requirehigher neurological levels. Whatever"segregation" of fibers occurs at thecord level can be related only to re-flex functions at this level and tonothing more (Bishop, 1946). Itshould also be noted that afferentimpulses in the cord can be alteredby impulses descending from thebrain stem and cerebrum. Hagbarthand Kerr (1954) have demonstratedthat afferent volleys in the anteriorcolumns of the spinal cord are re-


duced in amplitude by electricalstimulation of the bulbar and mid-brain reticular formation, of the pre-central and postcentral gyri, and ofvarious other forebrain structures.

Conduction at the Brain Stem Level

There seemed to be general agree-ment, just a few years ago, that spi-nothalamic pathways "carried pain"without interruption through themedulla, pons, and midbrain to theposterolateral ventral nucleus of thethalamus. Recent evidence indicatesthat this also is an incomplete ac-count. First of all, there is littledoubt that the great majority of thefibers from the anterolateral funicu-lus of the cord terminate at levelsbelow the thalamus (Walker, 1940;Walker, 1943; Bowsher, 1957). Fur-thermore, an extensive series of in-vestigations, recently summarized byMagoun (1958), indicate that the"classical sensory pathways" (includ-ing the "pain" pathways) give offcollaterals to the reticular formationof the brain stem (and to the "dif-fusely" projecting thalamic nuclei)and that appropriate electrical stimu-lation of this zone of collateralization—"the reticular activating system"—causes a desynchronization of elec-trical activity throughout wide areasof the cortex such as is seen in the"arousal" reaction in the normal ani-mal. In line with this evidence, ithas been demonstrated that anes-thetics exert their primary effect inblocking the response to noxiousstimuli (as well as other stimuli) bypreventing conduction through thereticular area of the brain stem(French & King, 1955). French,Verzeano, and Magoun (1953) re-ported that both sodium pentobarbi-tal and ether depress conductionthrough the reticular formation whilethe direct afferent pathways con-tinue to conduct impulses in normal

manner. Similarly, Arduini andArduini (1953), Peterson (1955), andHaugen and Melzack (1957) foundthat the potentials in the reticularformation were much more suscepti-ble to procain, nitrous oxide, andother drugs than the potentials inthe direct afferent pathways.

In summary, recent evidence seemsto be consistent with Melzack, Stot-ler, and Livingston's (1958) conclu-sion from their study of brain stemlesions in the cat:

Whatever the nature of pain perceptionmay be, its neural substrates appear to bemuch more complex than that envisaged in asingle ascending system. The patterns of im-pulses subserving pain appear to travel overmultiple pathways at the brainstem level atleast, and the ultimate perceptual event seemsto depend upon activities occurring along allof these pathways (p. 365).

Conduction at the Thalamo- CorticalLevel

It has generally been assumed that,after synapsing at the posterior ven-tral nucleus of the thalamus, "painfibers" project to the postcentral con-volution of the cortex. However, asWalker (1943) has pointed out, theposterior ventral nucleus has numer-ous connections with the adjacentthalamic nuclei and impulses can beconveyed to wide areas of the cere-bral cortex in this indirect fashion.4

Also, as pointed out above, nocicep-tive stimuli applied to visceral,somatic, and cutaneous structuresalso activate neurons in the reticularformation which sends impulses tomany cortical areas over both tha-lamic and extrathalamic pathways.In line with these considerations,

4 Murphy and Gellhorn (1945) report thatstrychninization of this thalamic nucleus alsoleads to firing in the ipsilateral and contra-lateral hypothalamus. Apparently, this is oneof the pathways involved in the generalhypothalamic excitation which follows periph-eral noxious stimulation (Gellhorn & Ballin,1946).


Gellhorn and Ballin (1946) reportthat noxious stimuli applied to theperiphery of narcotized animals alterelectrical activity throughout the en-tire cortex, and Benjamin and Ivy(1949) report that noxious stimuliapplied to the extremities of human5s evoke a nonspecific decrease inamplitude of the waves from theparietal, occipital, temporal, andfrontal areas.6

In general, the evidence summar-ized below indicates that "adequate"stimulation of the cerebral cortexmay elicit reports of pain; and thatdamage to a number of cortical areasmay affect "the sensation of pain"and the withdrawal movementswhich normally follow noxious stimu-lation.

In rare instances, electrical stimu-lation of the cerebral cortex, espe-cially of the precentral, postcentral,and superior parietal gyri, has beenfollowed by "a sensation of pain"localized in the face, limbs, trunk, orother body area (Penfield & Boldrey,1937; Horrax, 1946; Lewin & Phil-lips, 1952). Also, in a few patients,destruction of the postcentral, supe-rior parietal, superior temporal, andinsular convolutions (Davison &Schick, 1935); or tumors in the pari-etal lobe alone or in the parietal plusthe frontal or occipital lobe (Michel-son, 1943); or tumors in the right orleft parietal, frontal, and temporalareas (Bender & Jaffe, 1958); havebeen reported to give rise to "spon-taneous pain" referred to variousbody areas. However, we cannot

6 Although wide ares of the cerebral cortexare normally activated by peripheral nocicep-tive stimulation, it is quite certain that someof the components of the pain response can becarried out by animals lacking this structure.In pontile cats, for example, Bard and Macht(19S8) report that a strong nociceptive stimu-lus elicits growl-like vocalizations, protrusionof claws, running movements, piloerection,and increased respiratory and cardiac activity.

conclude from these reports that thecerebral cortex "subserves" some spe-cial function in the "perception ofpain." Electrical stimulation ofmany other neural tissues also, attimes, evokes referred "pain sensa-tions" (Sweet, White, Selverstone, &Nilges, 1950). Also, a series of in-vestigations, summarized by Bonica(1953, p. 131), White and Sweet(1955, pp. 526-528), and Bender andJaffe (1958), indicate that pathologi-cal processes in the spinal cord, thebrain stem, and the thalamus mayalso produce "spontaneous pain" in-distinguishable from that producedby lesions or tumors in the cortex.Furthermore, the pain response whichfollows electrical stimulation or path-ological processes in the cortex couldpossibly be integrated by subcorticalstructures. Finally, it should be em-phasized that reports of pain follow-ing electrical stimulation or lesionsin the cerebral cortex are so rare thatPenfield and Rasmussen (1950, p. 3)conclude that "the thalamus retainsthe problem of disposing of pain im-pulses without calling on the cortexfor essential help."

In rare instances, cortical lesionshave been reported to prevent "thesensation of pain" or "the recogni-tion of the stimulus" without affect-ing other components of the pain re-sponse. Gilliatt and Pratt (1952)have reported that after a "right-sided cerebral thrombosis" a patientdid not "consciously recognize" noxi-ous stimuli applied to the left side ofthe body, even though the stimuligave rise to general restlessness, in-creased blood pressure, tachycardia,deepening of respiration, and dilata-tion of the pupils. Marshall (1951)has also published 11 cases of left andright parietal lesions which were fol-lowed by a deficient "pain sensation"when pinprick, or intravenous injec-tion of hypertonic sodium chloride,


was applied to areas contralateral tothe lesion. These reports are excep-tional; localized cortical lesions arenot usually followed by alterations in"the sensation of pain." Penfield(cited by White & Sweet, 1955, p.109), after wide experience with corti-cal ablations, states that he has"never seen a patient who had aparietal lesion lose sensation of painexcepting for a few hours or days fol-lowing excision." White and Sweet(1955, p. 63) similarly conclude afterreviewing the evidence that "studiesin man following localized corticalextirpations reveal little reduction ofpain sensation upon peripheral stimu-lation, and confirm the huge extentof the cortex concerned with sensa-tion."

Damage to a number of corticalareas may affect the purposive with-drawal movements which normallyfollow nociceptive stimulation.Schilder and Stengel (1931) pub-lished a study of 3 patients withtumors or lesions of the left parietallobe (with additional lesions, in twoof the cases, in the frontal or tem-poral lobe) who did not withdraw fromnoxious stimuli, threatening gestures,loud noises, or sudden flashes of light.Similarly, Hemphill and Stengel(1940) reported that a patient with aprobable lesion of the left labyrinthfailed to show withdrawal responsesto noxious stimuli and to unexpectedsounds. Although the patient "ad-mitted that he could feel the painfulstimulus" and that he could hear anautomobile horn, he failed to showwithdrawal or defense reactions whena match was struck close to his eyesand when an automobile horn threat-ened his life. Rubins and Friedman(1948) have published a similar studyof four patients with lesions in oraround the supramarginal gyrus ofthe dominant hemisphere who showeda lack of withdrawal to noxious

stimuli and to threatening gestureseven though they "felt" pain andwere aware of the threatening char-acter of the gestures. The latter in-vestigators emphasize that only cer-tain motor withdrawal reactions ap-pear to be normally integrated in oraround the damaged areas.

Although specific unilateral lesions,in some instances, result in deficien-cies in pain responsiveness, it by nomeans follows that the more exten-sive the unilateral lesion, the moredeficient the response. On the con-trary, removing either the right orleft cerebral hemisphere either doesnot seriously affect the response to anoxious stimulus or alters only theresponse to lower intensities of stimu-lation. Dandy (1933) reported twocases of extirpation of the right cere-bral hemisphere; in both patients theresponse to a pinprick on the contra-lateral side was seriously deficient,but movements of joints and com-pression of muscles on either side ofthe body brought forth a pain reac-tion with an intense "feeling" com-ponent. Gardner (1933) found that20 months after right hemispherec-tomy firm pressure with a pin (on thecontralateral or ipsilateral side) wasrecognized as "painful." Zollinger(1935) reported that after removal ofthe left cerebral hemisphere (in aright-handed woman), the patientshowed "acute pain with motion ofthe joints or compression of the deepmuscles." Rowe (1937) stated that,after removal of the right hemi-sphere, his patient responded nor-mally to nociceptive stimuli appliedanywhere on the ipsilateral side andto scattered areas on the contralat-eral side. Somewhat in contrast tothe above are the later reports ofEvans (cited by Walker, 1943), Belland Karnosh (1949), Krynauw (1950)—12 patients—, and Marshall andWalker (1950)—4 patients: a few


months after hemispherectomy, mostof their patients showed normal painresponses and accurate localization ofpinprick applied on either side of thebody.

From our current neurological con-cepts we might assume that afterhemispherectomy the ipsilateral thal-amus integrates the response to anociceptive stimulus on the contra-lateral side. This is by no means thecase. The hemispherectomized chim-panzee shows practically completedegeneration of all the ipsilateralthalamic nuclei which project to thecortex (Walker, 1943). There is noreason to suspect that the same retro-grade thalamic degeneration does notoccur in man. Apparently, the re-maining cerebral hemisphere and thethalamic nuclei on the same side aresufficient to integrate the response tonociceptive stimuli on either side ofthe body.

The above studies appear to indi-cate the following:

1. Noxious stimulation in the peripheryalters electrical activity in many corticalareas.

2. Referred "pain sensations" are at timesevoked by electrical stimulation or pathologi-cal processes at any level of the neuraxis,including the cerebral cortex.

3. In rare instances, localized cortical le-sions abolish "the sensation of pain," i.e., theability to discriminate a noxious stimulus andto differentiate it from other stimuli, withoutaffecting other components of the pain re-sponse. Also, in rare instances, localized cor-tical lesions abolish the avoidance movementswhich normally follow noxious stimulation.

4. Removal of either the right or left cer-ebral hemisphere either does not seriouslyaffect the response to nociceptive stimuli oralters only the response to relatively non-intense stimuli applied to the contralateralside.

5. Although the decorticate animal showssome components of the pain response—e.g.,running movements and increased respiratoryand vasomotor activity—, the intact organ-ism probably utilizes a variety of corticalneuronal mechanisms when carrying out thetotal response to a noxious stimulus. This isfurther exemplified in the following discussion

on the mitigation of the discomfort-sufferingcomponent of the pain response by prefrontalleucotomy.

"RELIEF OF PAIN" BY PREFRONTALLEUCOTOMY

During recent years an extensivegroup of patients has undergone pre-frontal leucotomy (or "lobotomy")for the relief of severe, intractablepain syndromes such as causalgia,postherpetic neuralgia, metastaticcarcinoma, thalamic syndrome, etc.(e.g., Van Wagenen, cited by Walker,1943; Dynes & Poppen, 1949; Free-man & Watts, 1950). Although somepatients died soon after the operationand others were not "relieved ofpain," others were "relieved" (atleast for an extended time period)and further analysis of this effect maygive us an increased understanding ofthe pain phenomenon.

First of all, it is necessary to pointout that intractable pain has beenalleviated in some patients not onlyby bilateral frontal leucotomy, whichsupposedly destroys the thalamo-frontal projections, but also by uni-lateral frontal leucotomy; by bilat-eral lower quadrant frontal leucot-omy; by topectomy (i.e., by remov-ing limited areas, such as Brodmann'sAreas 9 and 10, from the frontallobes); and by a number of otheroperations on the frontal areas whichhave been summarized by Sargantand Slater (1954). Secondly, the"pain relief" which may follow theseoperations does not appear to be re-lated to the specific prefrontal areasaffected; on the contrary, the degreeof "relief" appears to be a nonspecificeffect, closely related to the extent ofthe prefrontal damage (Petrie, 1951;Hardy, Wolff, & Goodell, 1952;Petrie, 1958; Elithorn, Glithero, &Slater, 1958).

It must be further emphasized thatonly some patients have been helped


by these procedures. Walker (1950)estimates that at least one third ofthe patients receiving these opera-tions have not had any "pain relief."A representative report is Hardy,Wolff, and Goodell's (1952) analysisof 38 prefrental leucotomies (25 uni-lateral and 13 bilateral) performedby Dr. Bronson Ray at the New YorkHospital for the relief of pain syn-dromes related to metastatic cancer,Hodgkin's disease, radiculitis, tabes,etc. Of the 25 patients receiving uni-lateral leucotomy, 10 were relieved ofpain and 15 showed no alteration intheir pain syndrome. Of the 13 pa-tients receiving the bilateral opera-tion, 11 were relieved and 2 were nothelped. The term relief of pain, asused by these investigators, impliesthat when the patient was directlyasked if he had pain, he replied eitherthat he no longer had it, or that it wasstill present but of lower intensitythan before, or that it was still pres-ent but no longer "bothered" him.Although some investigators use anadditional criterion—viz., that thepatient no longer asked for drugs—most investigators also use the abovecriteria.

A further point should be empha-sized: postmortem examinations ofleucotomized patients indicate thatin some cases the prefrontal areasare not damaged and the th alamo-frontal projections are not severed.In a postmortem study of 15 pa-tients who had undergone transorbitallobotomy for pain of malignant dis-ease, Freeman and Williams (1951)found that 3 cases were characterizedby massive hemorrhage, 2 cases failedto involve the thalamofrontal pro-jections, and the other cases appar-ently showed destruction of the thal-amofrontal radiations with retro-grade degeneration of the dorso-medial nucleus of the thalamus.Meyer and Beck (1945) also report

from postmortem studies that theprefrontal lobe is at times entirelyuntouched and that severance of thethalamofrontal fibers is often incom-plete.

When prefrontal leucotomy allevi-ates intractable pain it does not nec-essarily elevate the pain threshold oralter "the sensation of pain." Chap-man, Solomon, and Rose (1950)found a lowering of the pain thresh-old immediately after the bilateraloperation followed by a return to pre-operative levels after an interveningtime period. Hardy et al. (1952) re-ported that the pain threshold in 10postleucotomy patients, who wererelieved of their pain syndrome,showed no significant difference fromthe preoperative level. King, Clau-sen, and Scarff (1950) noted a slightlowering or no change in the painthreshold after successful unilateralleucotomy for intractable pain. Also,with few, if any, exceptions, investi-gators report that the "sensation" or"perception" of pain is practicallyunaltered by any of these procedures:e.g., "Prefrontal lobotomy changesthe attitude of the individual towardhis pain, but does not alter the per-ception of pain" (Freeman & Watts,1950, p. 354).

The evidence available at presentalso indicates that, if and when pre-frontal leucotomy relieves a painsyndrome, the relief is secondary toa more generalized effect of the op-eration which, at times, can be con-ceptualized as apathy, i.e., as a de-creased responsiveness to all stimuli—including nociceptive stimuli.Hardy et al. (1952, p. 317) empha-size that postleucotomy patients whowere either partially or totally re-lieved of pain "exhibited in manyways . . . a flattened affect if notactual apathy. . . . They failed notonly to complain of their spontane-ous pain but also of their needs, such


as personal nursing care, need of urinebottle, bedpan, or the adjustment ofan uncomfortable dressing. Whenincontinent of feces they were indif-ferent to the odor it spread abouttheir persons and beds." Bonner,Cobb, Sweet, and White (1952) havealso emphasized that apathy charac-terized their patients immediatelyfollowing bilateral lower quadrantfrontal leucotomy. Although theapathy tended to lessen with the pas-sage of time, it was still a character-istic feature in patients followed upto 36 months postoperatively.

Although many patients who arerelieved of intractable pain by pre-frontal operations do not show theextreme apathy described above, theevidence indicates that all patientswho are helped by these operationsshow a characteristic personality al-teration (Krayenbiihl & Stoll, 1950;Petrie, 1952; Petit-Dutaillis, Mes-simy, & Berges, 1953; Elithorn,Glithero, & Slater, 1958); and thatpatients who are not helped and pa-tients who have undergone other op-erations which do not mitigate in-tractable pain, e.g., temporal lobot-omy, cingulectomy, and orbital un-dercutting, do not show the samechange in personality (Petrie, 1958).This characteristic alteration has re-ceived a wide variety of formulations:Dynes and Poppen (1949) concep-tualize it as a decrease in "worry"and "concern"; Le Beau (1950) termsit the relief of "anxiety"; Elithorn etal. (1958) formulate it as an "im-paired ability to elaborate a persist-ing attitude or mood." These formu-lations are not necessarily in basicdisagreement; they appear to be re-ferring to a common behavioral ma-trix, viz., to a mitigated "readiness torespond" to external and internalstimuli. Summarizing the investiga-tions in this area, Walsh (1957, p.474) writes: "The patient suffering

from pain complains less of his dis-comfort than before. Not . . . a fail-ure to appreciate the situation but afailure to respond to i t . . . . This fail-ure to react is seen when stimuli thatarise within the body itself are con-sidered ; but there may also be a dim-ished response to external situa-tions."

It should be emphasized that theleucotomized patient is able to re-spond normally to nociceptive stimu-lation. Hardy et al. (1952, p. 316)have reported that "some patients,although ostensibly tranquil beforebeing asked about their pain, over-reacted with a show of grimacing andfears when their attention was focusedupon it by a direct question concern-ing its quality and its intensity" (em-phasis added). The same theme is re-peated by other investigators; for ex-ample, Hawkes and Gotten (1948, p.209) report that "when questioned[the leucotomized patients] all indi-cated that they realized some painwas present when they thought aboutit" (emphasis added). Apparently,when the leucotomized patient is di-rectly asked to report on his pain, he'' focuses his attention " on and " thinksabout" the ever-present nociceptivestimulus in his body and, when thusreacting to it, often shows discomfortor suffering and almost always re-ports a "sensation of pain." How-ever, when the patient is not directlyasked to report on the noxious tissuecondition, he does not "attend" to itor "think" about it to the same extentas before the operation and, whennot thus reacting to it, does not ap-pear to be "in pain," i.e., does notshow discomfort.

Apparently, discomfort and suffer-ing can be minimized or eliminatedby preventing a "secondary reaction"to the noxious stimulus. Neurosur-geons have used somewhat differentterms to describe this effect: Freeman


(1949, p. 18) writes after extensiveexperience with prefrontal operationsthat "when the emotion is done awaywith, the pain either becomes nolonger significant or actually disap-pears"; Otenasek (1948, p. 234) sim-ilarly suggests that "when the fearof pain is abolished, the perception ofpain is not intolerable."

Neurophysiological Correlates of Post-leucotomy "Pain Relief"

Since prefrontal leucotomy miti-gates the discomfort-suffering com-ponent of the pain response in somepatients and fails to do so in others,since this effect is often temporary,and since we are rarely certain in anyone such operation which fiber tractswere destroyed, to what extent scarformation and vascular damage oc-curred, and to what extent the opera-tion resulted in physiological dis-turbances in other cerebral areas, it isextremely difficult to formulate anyhypothesis concerning the "pain re-lief" which may follow this operation.Also, as Koskoff, Dennis, Lazovik,and Wheeler (1948) have pointed out:

The relief of suffering in such a wide varietyof patients suggests that the mechanism doesnot involve the interruption of specific painpathways, despite the evidence of thalamicdegeneration following frontal lobotomy.Preservation of the response to painful stimu-lation noted in such patients also suggeststhat the interruption of specific afferent paintracts is not responsible for the relief of suffer-ing (p. 740).

Nevertheless, a number of investi-gators have proposed a variety ofmechanisms which may be directlyor indirectly related to postleuco-tomy "pain relief." Starzl and Whit-lock (1952) have presented evidencethat the "diffuse" thalamic projec-tion system, which exerts a generalcortical "arousal" effect, projectsprimarily, but not exclusively, to thefrontal cortex in monkey. From this

evidence they hypothesize that "painrelief" following leucotomy is due tothe destruction of the afferent fibersfrom this system. Fulton (1951, p.127) has suggested that frontal leu-cotomy relieves pain by removing"large numbers of visceral pain pro-jections from the sphere of conscious-ness," specifically, by destroyingvisceral afferent pathways to theorbitofrontal cortex. However, asWhite and Sweet (1955, p. 64) pointout, "We know of no evidence . . .that stimulus to the central end ofany visceral nerve carrying manynerve fibers, such as the greatsplanchnic nerve, will cause synchro-nous bursts of change of potentialwithin this part of the brain in mam-mals." Also, Fulton's suggestion doesnot explain why the leucotomized pa-tient appears to have a diminishedresponsiveness to many other stim-uli besides noxious stimuli nor doesit explain why the patient may state,when directly asked, that the "painfeeling" is the "same" but does notmatter any more.

The above investigators have em-phasized the destruction of the af-ferent projections to the frontal areasand have neglected the probable ex-tensive destruction of corticofugalfibers. In a postmortem investigationof six lobotomized patients, Yakov-lev, Hamlin, and Sweet (1950, p.328) found that the frontopontinetracts were bilaterally and symetri-cally degenerated and that "thegreat frontal corticofugal pathwaydescending from the entire anteriorpre-Rolandic half of the cerebralhemisphere was deprived of a largeand important component." Theyconclude with a statement that can-not be too much overemphasized:

On the basis of this study it seems to us thatin the attempts made thus far to correlate the


behavioral changes following frontal lobotomywith anatomy . . . the degeneration of an-terior thalamic radiations and nuclei has beenstressed to the exclusion of the obvious de-generation of the far greater mass of efferentprojections which connect frontal lobes to allthe levels of the neuraxis . . . (p. 328).

In line with this suggestion, a fewworkers have attempted to formu-late the effects of leucotomy in termsof the destruction of efferent projec-tions. Bonner et al. (1952) speculatethat if the connections between neo-cortex and archicortex are severed byprefrontal leucotomy "there wouldbe less activation of the archicorticalcircuits which probably subserveemotional reactions and thus per-petuate suffering." Arnold (1955, p.154) hypothesizes that since thedorsomedial nucleus of the thalamusdegenerates after prefrontal leucot-omy and since the frontal lobes ac-tivate the sympathetic centers in theposterior hypothalamus by connec-tions through this nucleus, "Anxietyis reduced because the excitation ofsympathetic effectors is now pre-vented and with it a prolongationand intensification of the emotion."However, since sympathectomizedanimals (Cannon, Newton, Bright,Menkin, & Moore, 1929) and sym-pathectomized human patients (Ray& Console, 1949; Crimson, Orgain,Anderson, Broome, & Longino,1949) apparently respond with nor-mal "emotion" and "anxiety" tomany stimuli, it is doubtful that theprevention of sympathetic excitationis the mechanism involved in the re-duced "anxiety" or diminished reac-tivity of the leucotomized patient.

In summary, although a number ofneurophysiological mechanisms areapparently nonfunctional after pre-frontal leucotomy, we cannot statewith any degree of certainty which ofthese mechanisms are necessary forthe maintenance of intractable pain.

Nor are we certain that destructionof one or more specific nuclei or nervepathways is closely correlated withthe effects of this operation. In fact,the evidence at present suggests thatprefrontal leucotomy has differenteffects on different patients evenwhen apparently similar neural tissueis destroyed. To account for thesedifferential effects we must have notonly (a) much more preoperativedata on each patient (e.g., the pa-tient's personality characteristics, hisgeneral level of reactivity, the dura-tion of his pain syndrome) but also(b) much more specific postoperativedata such as the specific tracts de-stroyed and the extent of postopera-tive hemorrhage, and (c) a betterunderstanding of a number of phe-nomena which at present are not wellunderstood, such as the "reintegra-tion" of function which may occurafter destruction of neural tissue, thespecific functions of the afferent andefferent fibers from the frontal areas,etc. When such specific data areavailable, we may be able to accountboth for the patient who is not helpedby this operation and for the pa-tient who is not only relieved of painbut also of worry and concern aboutmany situations including, in manycases, forthcoming death.

THE PROBLEM OF CONGENITALINSENSITIVITY TO PAIN

A theory of pain must account forthe "normal" response to noxiousstimulation, for the alterations inthis response by analgesics, place-bos, hypnosis, neurosurgical and otherprocedures, and for the antithesis of"normal" pain responsiveness, i.e.,the problem of "congenital insensi-tivity to pain." At the present timewe are far from a complete under-standing of the latter phenomenon.Nevertheless, within the limits of the


evidence as it now stands, certain sig-nificant factors stand out that shouldbe emphasized.

McMurray (1955) and Critchley(1956) have recently reviewed thehandful (ca. 16) of well-documentedcases of "congenital insensitivity."McMurray's (1950) case can bebriefly summarized to indicate themore or less typical findings in thesepatients:

A 22-year-old female college stu-dent, IQ 128, with no apparent per-sonality disorders. A history of con-sistent lack of pain responsivenessdating at least since early childhood.Extensive burns, frostbite, deep cuts,and other serious tissue damage "hadgone unnoticed or been looked on in-differently." Her medical history in-cluded the incision of a large abscessover the occipital bone, osteomyelitisof the right calcaneus and of the leftfemur, tonsillectomy and adenoidec-tomy, and acute pyelitis, with nocomplaints of pain or tenderness.When subjected in the laboratory tosuch noxious stimuli as cold waterat a temperature of 0° to 2° C., hotwater at 49° to 51° C., and electricshock from an inductorium, she didnot report pain, did not show wincing,withdrawal, or other indications ofdiscomfort, and did not show anysignificant alterations in blood pres-sure, heart rate, or respiration. Ex-tensive neurological examination didnot reveal any evidence of organicneurological disease.

Although the other reported casesgenerally follow the above pattern,there are some differences: 2 patients(Ford & Wilkins, 1938, Case 2;Kunkle & Chapman, 1943) showedepileptic tendencies; 3 patients(Kunkle & Chapman, 1943; Arbuse,Cantor, & Barenberg, 1949; Cohen,Kipnis, Kunkle, & Kubzansky, 1955)showed increased diastolic and sys-

tolic blood pressure and increasedheart rate when their hands wereimmersed in water at 0° C.; and atleast 4 patients were of borderlinenormal or below normal intelligence(Ford & Wilkins, 1938, Case 2;Arbuse et al., 1949; Farquhar & Sut-ton, 1951; Madonick, 1954).

In many of the reported cases, theinsensitivity to pain is not an all-or-none phenomenon. Many, and possi-bly all, of these 5s have at some timein their life responded in the normalmanner to some noxious stimuli.Three of Jewesbury's (1951) casesillustrate this exceptionally well. Hisfirst case showed no response to pin-prick or to laboratory pain tests suchas muscle ischemia and histamine in-jection, he was able to pick up glow-ing coals without discomfort, and hehad teeth drilled and extracted with-out any report of pain; however, thepatient did, at times, show the nor-mal response to noxious stimulation,for example, when he had smashedhis fingernail and when he had beenkicked on the testicles. Jewesbury'ssecond and third cases also did notshow pain responses in the labora-tory tests of muscle ischemia, hista-mine injection, and electric shock at40 milliamperes (the maximum avail-able); however, the second patienthad frequent frontal headaches andshowed normal pain responsivenessduring appendicitis and pyelitis andthe third patient had reported painfrom headaches and from retention ofurine due to an enlarged prostate.Similarly, Kunkle and Chapman's(1943) patient had complained of"moderate toothache"; Rose's (1953)patient complained of pain after avascular accident in his right leg;Madonick's (1954) patient and two ofFord and Wilkins' (1938) patientshad "abdominal pain"; the patientof Cohen's et al. (1955) had reported

RELIEF OF CSRON1C PAIN 443

a "throbbing headache" after spinalanesthesia; and Je'quier and Deller's(1956) patient reported "a littlepain" when stimulated with a veryhot object.

As Critchley (19S6) has noted,these 5s are not actually "insensi-tive" to noxious stimulation; theycan detect, identify, and localizenoxious stimuli and can easily dif-ferentiate them from other stimuli.McMurray's (1950) 5 states that,when a hypodermic needle is insertedinto her skin, she feels it penetratingthe tissue layers but does not "feelpain." Stimuli such as pinprick andcutaneous shock and heat producethe report of a pricking or sharp qual-ity, but she does not describe thisquality as "painful." In fact, sincethis 5 can discriminate the sharpquality of heat stimulation, McMur-ray was able to establish in the pa-tient a "threshold" close to the nor-mal heat pain threshold. Similarly,Ford and Wilkins (1938), Kunkleand Chapman (1943), Boyd and Nie(1949), Jewesbury (1951), Westlake(1952), and J6quier and Deller (1956)have reported that their 5s had nodifficulty differentiating and localiz-ing a nociceptive stimulus; theycould, for example, easily discrim-inate between the blunt and pointedend of a pin and had no difficultylocalizing the pinprick.

The available evidence indicatesthat many, if not all, of these 5s havenormal peripheral neural apparatus.Biopsy specimens from McMurray'spatient showed "nerve fibers and freenerve endings present. . . . No mor-phological features that would distin-guish them from the pain endings ofnormal subjects" (Feindel, 1953, p.402). Other investigators who at-tempted histological studies (Girard,Devic, & Garin, 1953; Madonick,1954; Cohen et al., 1955) also found

nerve fibers in apparently normalpattern.

In many, if not all, of these cases,the evidence indicates that no dis-tinct localized damage exists in thecentral nervous system. Investi-gators who performed extensive neur-ological examination of their 5s(Boyd & Nie, 1949; Arbuse et al.,1949; McMurray, 1950; Jewesbury,1951; Rose, 1953; Madonick, 1954;Jequier & Deller, 1956) report thatall tests were essentially normal—normal reflexes, normal skull andspine X ray, normal pneumoenceph-alogram, normal electroencephalo-gram, etc. Arbuse et al. (1949) haveemphasized that there is no indica-tion in their case, or in any other re-ported case, of a lesion in any specificpart of the brain. Most investigatorswho have examined these 5s appearto be in agreement with De Jong's(1949, p. 411) conclusion that the de-fective reaction is more likely due toa "generalized or diffuse develop-mental anomaly" and that it ishighly doubtful that any "local le-sions" exist.

In at least three of the reportedcases the pain insensitivity was notdue to an irreversible "anomaly."Ford and Wilkins1 (1938) first caseappeared to be insensitive to painand readily submitted to many seri-ous noxious stimuli in the laboratorywithout signs of discomfort; later,however, he seemed to be afraid of"getting hurt," refused to have atooth extracted without an anes-thetic, and generally appeared to bebecoming more concerned about po-tentially pain-producing stimuli.Similarly, during the first 2| years oflife, Jewesbury's (1951) fourth casedid not show any signs of pain re-sponsiveness to a wide variety of in-juries—serious burns, bruises, bleed-ing fingers, etc. At two years of age


he had been reported in the press as"the child who knows no pain."However, when examined at 3| yearsof age he showed normal pain re-sponses to all nociceptive stimuli.Rose's (1953) case also followed asimilar pattern; Rose reports that"his sensitivity to pain is becomingprogressively nearer the normal andhe now feels the minor injuries ofboy's life as well as any other child."

Since each of the reported cases ap-pears to differ in some way fromevery other reported case, we cannotgeneralize from the above data. How-ever, we are probably safe in tenta-tively concluding that some of these5s are able to respond to at least somenociceptive stimuli in the normalmanner, i.e., with the "sensation ofpain," discomfort, and alterations insome physiological functions, eventhough they almost always fail to doso. Also, many, if not all, of these 5scan discriminate and localize noxiousstimuli and easily differentiate thesestimuli from heat, warm, pressure,and touch stimuli. But this "sens-ing" of a noxious stimulus is not"painful"; very rarely is it associatedwith unpleasantness or discomfort.As Critchley (1956, p. 742) haspointed out: "The most remarkablefeature in this syndrome is a typicallack of conformity between the feel-ing of pain as a discriminative qualityof sensation, and the registration ofdistress, either overtly or automati-cally." In fact, the available evi-dence suggests that some, if not many,of these 5s resemble in their pain re-sponsiveness the hypnotic "anal-gesic" 5 and the restricted and iso-lated animals studied by Melzackand Scott (1957) more than they re-semble those rare patients with le-sions of the afferent apparatus whoare unable to discriminate a nocicep-tive stimulus. The former phenom-

ena, summarized below, also indicatethat "pain" in the sense of discom-fort and suffering is not necessarilypresent when noxious stimuli are dis-criminated, differentiated, and local-ized.

THE EFFECT OF EARLY ISOLATIONON THE PAIN RESPONSE IN

THE ADULTMelzack and Scott (1957) have

provided much needed data concern-ing the effect of early isolation onpain responsiveness in the mature or-ganism. These investigators reared10 dogs in isolation from puppyhoodto maturity in special cages whichdrastically limited both their over-allexperience and their specific experi-ence with nociceptive stimuli. Com-paring the behavior of these re-stricted dogs with the behavior of 12normally reared dogs, they report thefollowing:

(a) In general, the 10 restricteddogs failed to show adaptive and in-telligent responses to noxious stimuli.Many of the dogs made no attempt toavoid a pinprick, a flame, or an elec-tric shock stimulus. Although someof the restricted dogs did learn toavoid these stimuli, they requiredmany more trials than the controlanimals. As long as two years afterrelease from isolation, many of therestricted dogs continued to showmaladaptive behavior when givennoxious stimuli. The investigatorsconclude that "it appears that therequisite experience must come at thecorrect time in the young organism'slife. During later stages of develop-ment, the experience necessary foradaptive, well-organized responses topain may never be properly ac-quired" (p. 159).

(&) The restricted animals ap-peared to be unable to localize thesource of the noxious stimulus. Not


only were the stimuli "not 'per-ceived' as coming from the experi-menter" but the dogs also appearedto be "unaware that they were beingstimulated by something in the en-vironment" (p. 158).

(c) Although the restricted ani-mals may have "felt" the nociceptivestimuli "in some way," they rarelyshowed discomfort or suffering:

Their reflexive jerks and movements dur-ing pinprick and contact with fire suggestthat they may have "felt something" duringstimulation; but the lack of any observableemotional disturbance apart from these reflexmovements in at least 4 of the dogs followingpinprick and in 7 of them after nose-burningindicates that the perception of the event washighly abnormal in comparison with the be-havior of the normally reared control dogs. . . .The results suggest that the restricted dogslacked awareness of a necessary aspect of nor-mal pain perception; the "meaning" of physi-cal damage or at least threat to the physicalwell-being (p. 159).

Additional investigations areneeded to determine the validity ofthe following hypothesis suggested bythis study: some components of thenormal pain response (local reflexmovements and "the sensation ofpain") do not require prior experi-ence with noxious stimuli; other com-ponents of the pain response (localiz-ing the stimulus, purposive with-drawal movements, and discomfort-suffering) require previous experiencewith such stimuli.

HYPNOTICALLY-INDUCED"ANALGESIA"

The experimental evidence, sum-marized by Weitzenhoffer (1953), in-dicates that, when given appropriatesuggestions to induce "analgesia,"some "good" hypnotic 5s do not showa pain response to some noxious stim-uli, that is, they do not give a verbalreport of pain, they do not withdrawfrom the stimulus, they do not show

discomfort by wincing, tremor, orrestlessness, and they do not showsignificant alterations in blood pres-sure, heart rate, pulse rate, or res-piration. Dynes (1932) reported thatfollowing pinprick during hypnoti-cally-induced "anesthesia" seven"trained" hypnotic 5s denied thatthe stimuli were painful, did not showwithdrawal or facial flinch, andshowed little or no disturbance in thenormal rate and rhythm of respira-tory and cardiac activity. Subse-quently, Dynes' 5s were asked (bysomeone other than the experi-menter) to "fake a trance" during thefollowing experiment but not to"enter hypnosis." In this situation,pretending they were in trance, theyshowed all of the normal responsesto the nociceptive stimuli. In a sim-ilar study, Sears (1932) recorded theresponses of seven "good" hypnotic5s to a sharp steel point pressedagainst the leg for 1 sec. with a pres-sure of 20 oz. Suggestions of anal-gesia were given for the left leg andthe right leg was employed as a con-trol. When the stimulus was appliedto the "analgesic" left leg, the 5s didnot show facial flinch or variations inrespiration and the increased pulserate, which normally follows nocicep-tive stimulation, was significantlydecreased. However, they did showthese responses when the stimuluswas applied to the "normal" right leg.In further control experiments, whenthe 5s were asked to inhibit all reac-tions to the noxious stimulation, all5s showed alterations in pulse andrespiration. Doupe, Miller, and Kel-ler (1939) have in general confirmedthese findings, reporting that theirhypnotic 5s showed a slight altera-tion in respiratory rhythm, no sig-nificant change in pulse rate, and nofacial grimace when multiple pin-pricks were applied to the "anal-

446 T&E&DORE #. BARKER

gesic" arm.6 Brown and Vogel (1938)also found that three 5s showed lessvariability in blood pressure, pulse,and respiration when nociceptivestimuli (lancet, thumb tack, andwater at 49° C.) were applied to the"anesthetic" limb than when thesame stimuli were applied to the"normal control" limb. Althoughthey conclude that "physiologicalreactions to moderate and mild sen-sory stimuli may be affected by sug-gestion in the hypnotic state and byimagination in the waking state," itis not clear from their report to whatdegree these responses were affected.

Although the experimental studiesgenerally report either a completelack or a significant decrease in vaso-motor and respiratory alterations fol-lowing nociceptive stimulation dur-ing hypnotically-induced "anal-gesia," they report completely con-tradictory results with the galvanicskin response (GSR). Some investi-gators (Georgi, 1921) found that theGSR to noxious stimulation was com-pletely eliminated during hypnotic"anesthesia"; others (West, Niell, &Hardy, 1952) concluded that it is attimes significantly decreased over thenormal and at other times completelyeliminated; and still others (Levine,1930; Barber & Coules, 1959) re-ported that the GSR to noxiousstimuli is not significantly alteredafter hypnotic suggestions of anal-gesia. However, an extensive groupof investigations indicate that theGSR is the least specific of all thephysiological responses which may

6 Doupe et al. also found that, in compari-son with the normal limb, the hypnotically"anesthetic" limb showed a reduced vaso-constrictor response to pinprick. They areuncertain, however, whether this "residualresponse" is "of the nature of a spinal reflex"or due to "sub-conscious or co-conscious ac-tivities" on the part of the 5.

follow noxious stimulation. Althoughblood pressure alterations, for ex-ample, are at times present when Sis responding to nonpainful stimuli,some variation in blood pressure ap-pears to be always present when Sdoes "feel pain" (Nafe & Wagoner,1938; Goetzl, Bien, & Lu, 1951). Thisis not true of the GSR, however. An5 may show a GSR when he does not"feel pain" and he may not show aGSR when he does "feel pain."Brown and Vogel (1938) demon-strated that hypnotic 5s often showeda GSR when there was no doubt thatthey did not "feel pain," that is, whennoxious stimuli were applied to anarea made insensitive by novocainblock. They write that "light appli-cation of the pin point [to the area inwhich novocain had been injected]. . . appreciated as touch, caused

large galvanometer deflections" (p.419). Along similar lines, Levine(1930) and Barber and Coules (1959),using hypnotic 5s, and Sattler (1943),using nonhypnotic 5s, found that5s often show a GSR when they aretold they are to be given a pain-ful stimulus but are not given thestimulus. West et al. (1952) foundthat (a) the GSR showed a significantdecrease over the control levels forall seven of their hypnotic 5s evenwhen "there was no alteration inpain perception, according to subjec-tive reports," and (6) during the con-trol periods a stimulus "evoking apain of 6 or 7 dols" at times failed toproduce a GSR. After a careful,long-term investigation designed todetermine the relationship of theGSR to the pain response, Purer andHardy (1950) concluded that theGSR is directly related to the"threat-content" of the stimulus andis not related to the "sensation ofpain" as such. Following Purer andHardy's interpretation, we can con-


elude that studies which have re-corded the GSR during hypnotically-induced "analgesia" indicate that tosome hypnotic "analgesic" 5s thenoxious stimuli are "threatening,"to others they are less "threatening"than during the control period, andto still others they are not "threaten-ing" at all.

However, we cannot draw any con-clusions from the above studies asto the effectiveness of hypnotic pro-cedures when the stimuli are moresevere and of longer duration. Forthis type of report we must turn tothe clinical investigations. In gen-eral, the clinical reports suggest thathypnotic methods, with some patients,may be as effective as morphine andother opiates in minimizing patho-logical pain syndromes and in miti-gating or totally eliminating the dis-comfort-suffering component of thepain response during a variety ofsurgical procedures. A typical reportof the surgical use of hypnotic tech-niques is Mason's (1955) discussionof a case of mammaplasty: during theoperation, which consisted of exci-sion of breast tissue, skin, fat, andcomplete reshaping of the breast, thepatient "never showed signs of painor seemed distressed" and the pulseand blood pressure showed very lit-tle, if any, alteration. Kroger (1957)has also reported four cases whichare more or less typical of the surgicalfindings. The first case, a 20-yr.-oldfemale, had "a fairly large tumor" re-moved from the right breast withoutpreoperative or operative medica-tion. She showed "no indication of apain reflex at any time" and she was"fully aware of the entire surgicalprocedure." Another patient, whounderwent Caesarean section andhysterectomy with hypnotic "anes-thesia," "experienced no subjectivediscomfort and conversed with every-

body in the operating room. She wasfully conscious and was able to watchthe birth of her baby. There was nodiscomfort when the baby was de-livered by forceps, or when the uteruswas extirpated." Other studies, re-cently summarized by Barber(1958b), also report that hypnoticmethods are successful with some pa-tients in minimizing or completelyeliminating the discomfort-sufferingcomponent of the pain response dur-ing childbirth, terminal cancer, fatalburns, dysmenorrhea, and other painsyndromes.

It should be emphasized, however,that in the more severe and intracta-ble pain syndromes, such as terminalcancer and spinal cord injuries, hyp-notic methods are reported to mini-mize discomfort and suffering; rarely,if ever, are these procedures reportedto completely eliminate the total painresponse to the ever-present noxiousstimulus in the patient's body. Dor-cus and Kirkner (1948) found that al-though hypnotic methods could min-imize discomfort in five cases of spinalcord injury—i.e., the patients re-ported less pain and requested asmaller amount of drugs—thesemethods were by no means effectivein entirely eliminating the pain re-sponse. Similarly, Butler (1954) re-ported that hypnotic methods wereeffective with some patients in min-imizing discomfort during terminalcancer—the patients either requiredhalf of their usual amount of mor-phine or, in a few cases, did not re-quire any drugs for a period of time.

The evidence available at presentindicates that two objections whichhave been raised concerning the ef-fectiveness of hypnotic procedures in"relieving pain" are not valid. Hull(1933) was of the opinion that hyp-notic 5s may state, after the experi-ment, that they did not "feel pain"


during the experiment because am-nesia has been suggested and theysimply do not remember. However,in a number of recent studies (Rosen,1951; Mason, 1955; Kroger, 1957)posthypnotic amnesia was not sug-gested and the patients continued toinsist that they had not "felt pain"even when they were perfectly ableto recall the entire procedure. Othershave objected that hypnotic 5s ac-tually "feel pain" but deny it (whenquestioned by the hypnotist) be-cause of their rapport or strong"transference" relationship with thehypnotist. This also seems doubtful.Whenever any of the above patientswere questioned afterwards by dis-interested observers, they continuedto vehemently deny "feeling pain"during the procedure (Marcuse &Phipps, 1956; Kroger, 1957).

Before we can state what are thenecessary and sufficient conditionsfor hypnotic "analgesia," we needmore extensive, controlled experi-ments utilizing a wide variety ofnoxious stimuli applied to visceral,somatic, and cutaneous structures.Recent investigators, however, haveemphasized three conditions whichmay be among the necessary condi-tions for this phenomenon.

First of all, it seems that the 5 mustbe a certain type of person who isable to become "deeply hypnotized."With few, if any, exceptions, investi-tors agree that these individuals (us-ually termed somnambulists) are asmall minority—5 to 25%—of thepopulation, at least in our culture(Weitzenhoffer, 1953, p. 59; Butler,1954; Mason, 1955; Kroger, 1957).The limited evidence available atpresent suggests that these indi-viduals are characterized by a num-ber of distinct "abilities." Young(1928, p. 372) found that one or moreof the following characteristics

showed themselves in all of his "best"5s long before they were "hypno-tized": "deep abstraction, reverieamounting almost to ecstasy, puttingoneself to sleep at will, actually hyp-notizing one's self." Similarly, Bar-ber (1958b, 1958c) found that all ofhis somnambulistic 5s had been ablesince childhood to go to sleep easilyand quickly at anytime—day ornight—and to concentrate on theirwork or studies by "blocking-out"irrelevant stimuli.

What appears to be a second neces-sary condition for hypnotic "anal-gesia" has been formulated by Leuba(1957) as follows:"There must be concentration on the ideaspresented by the hypnotist and with a mini-mum of counter or critical thoughts; and abelief that what the hypnotist says will hap-pen, can actually happen, and will happen. Inother words, there must be a set or attitude toaccept the hypnotist's statements completelyand uncritically" (p. 37).

Along similar lines, Kroger (1957, p.xi) has concluded from his extensiveexperience with hypnotic proceduresthat "when one wishes to performmajor surgery under hypnoanesthesia. . . it is very important to get the

patient to believe in the actuality ofthe trance state." Recent evidenceindicates that these statements maybe valid. Barber (1957b) reportedthat a somnambulistic hypnotic 5(who quickly and easily carries outall of the "complex" hypnotic behav-iors such as analgesia, age-regression,negative and positive hallucinations,etc.) becomes "unhypnotizable"when he no longer "believes" in hyp-nosis, i.e., when he concludes fromhis own reading, or from training inautohypnosis,7 that the hypnotistdoes not possess any special power orability and that whatever occurs dur-

7Shor, R. E.October, 1957.

Personal communication.


ing the hypnotic situation is broughtabout primarily by the subject him-self. A number of investigators havealso reported that somnambulistic 6*8(in the "deepest stage of hypnosis")do not carry out "complex" hypnoticbehaviors such as color blindness(Erickson, 1939), age-regression(Orne, 1951), immoral or dangerousbehavior (Young, 1952), and nega-tive hallucinations (Barber, 1958a)when the hypnotist simply givesthem the appropriate suggestions;however, they do carry out such be-havior when the hypnotist manipu-lates the situation in such a way as tolead the 5s to "believe" that the sug-gestions are literally true statements.

A third factor which seems to beclosely related to the above, has beenrecently emphasized by physiciansattempting to relieve the pain ofterminal cancer or childbirth by hyp-notic methods; the patient must haveconfidence in his physician-hypnotistand the hypnotist must "give of him-self" to the patient. In treating thepain of terminal cancer by hypnoticprocedures, Butler (1954) saw his pa-tients at least daily and often two tofour times a day. Whenever hypno-therapy was terminated for anylength of time, the patients all showeda return of the original pain syn-drome. However, in the few caseswhen hypnotic procedures were dis-continued but the patient received thesame amount of personal attentionfrom the physician, the patients didjust as well for one or two days asthey did during "hypnosis." Butleremphasizes that in treating the painof terminal cancer by hypnotic meth-ods the hypnotist-physician "gives ofhimself to the patients. . . . Even anhour's treatment with a very sick pa-tient can produce an appreciable tir-ing of the hypnologist, and, as thesympathetic bond between the two

grows stronger, the hypnologist mayeven 'feel' the symptoms he is tryingto eradicate from the patient" (p. 6).Along similar lines, Winkelstein (1958,p. 154) concluded, after using hyp-notic methods over a 2-yr. periodwith 200 of his obstetrical patients,that "the mental attitude of the pa-tient, the patient-obstetrician rap-port, and the confidence of the pa-tient in the procedure as well as inthe accoucheur, seemed to be as im-portant factors as was the hypno-suggestion itself."

In summary, the evidence availableat the present time indicates thatwhen the hypnotist properly manipu-lates the situation some "good" hyp-notic 5s show a mitigated pain re-sponse to some noxious stimuli,8 thatis, (a) they do not show withdrawalor avoidance, (b) they report that thestimuli are not painful, (c) they donot show discomfort and (d) theydo not show physiological responsessuch as vasomotor and respiratoryalterations (although they may ormay not show galvanic skin re-sponses) . The evidence also suggeststhat 5s who are able to carry out theabove are "set" to accept the hyp-notist's suggestions as literally truestatements and have complete confi-dence in the hypnotist and in theefficacy of hypnotic procedures.9

8 It should be emphasized that the hypnotic"analgesic" S, like the leucotomized, nar-cotized, or congenitally insensitive patient, isable to discriminate, differentiate, and localizethe noxious stimulus when asked to do so(Rosen, 1951), Although he can "sense" thestimulus, it does not arouse discomfort.

9 As will be pointed out below, the "painrelief" which at times follows the administra-tion of a placebo is also closely related to the5's belief or conviction that the "drug" hascurative properties.. Apparently, at least someof the effectiveness of hypnotic "analgesia"is due to a "placebo effect."

However, the hypnotic "analgesic" 5 alsoresembles the patient who has received mor-


THE INCONSTANT PAIN THREHSOLDRecent studies indicate that both

morphine and placebos can eliminatediscomfort and suffering without al-tering the pain threshold and withoutaffecting "the sensation of pain."Since many of the studies on placebosand analgesic drugs are intimately re-lated with the pain threshold studies,we shall first review the latter investi-gations.

When the subject of pain was lastreviewed in this journal (Edwards,1950) it appeared that Hardy, Wolff,and Good ell (1940) had establishedthat the pain threshold was relativelyconstant in the same 5 at differenttimes. Using what was later to be-come known as the Hardy-Wolff-Goodell radiant heat technique10 andusing themselves as 5s, these investi-gators reported that when pain thresh-old measurements were taken almostdaily for nearly a year, the averagethreshold value was 232 mc./sec./cm.2 with a standard deviation ofonly +9 millicalories. In addition,they reported that all observationswere within ± 12% -of the mean. Italso appeared that the same workers(Schumacher, Goodell, Hardy, &Wolff, 1940) had established that alarge group of untrained 5s had ap-proximately the same pain threshold.They reported that the average thresh-

phine or other opiates. As pointed out below,morphine apparently gives "pain relief" bybringing about "freedom from anxiety" or "abemused state." These terms also appear ap-plicable to the "good" hypnotic 5 who be-comes relatively unconcerned about and"relatively inattentive to all stimuli except thewords of the operator and stimuli to whichthe operator specifically directs his attention"(Barber, 19S7a).

To what extent hypnotic "analgesia" is dueto each of these seemingly different mecha-nisms is a subject for further research.

10 For a detailed descrption of this methodsee Hardy et al. (1952, pp. 67-85).

old for 150 untrained 5s was 206mc./sec./cm.2 with a standard devi-ation of only +21 millicalories and arange extending only from 173 to232 millicalories. Subsequent investi-gations, however, have failed to con-firm both of the above conclusions;it now appears that there is a widevariation in pain threshold among agroup of 5s and that the thresholdis by no means consistent in thesame 5 over time.

Using the Hardy-Wolff-Goodell ra-diant heat technique, Chapman andJones (1944) found that the painthresholds of 200 5s varied between— 40 to +50% of the mean andKuhn and Bromiley (1951) reportedthat the pain thresholds of 37 5sranged from 169 to 296 millicalorieswith a standard deviation of 31.9.Hall and Stride (1954), using a modi-fied Hardy-Wolff-Goodell technique,found that the pain threshold of 400psychiatric patients (neurotics, de-pressives, and schizophrenics) ex-tended "over almost the whole rangeof stimulus intensity" with the meanat 260 millicalories and a standarddeviation of 72 ±45. The depressivesand schizophrenics reported pain ata uniformly high level of stimulusintensity while the anxiety neuroticsconsistently reported pain at lowstimulus intensities. Since the painthreshold, but not the warmth thresh-old, could be easily altered by vary-ing the instructions, Hall and Stridesuggest that pain threshold varia-tions are due to "central attitude orpain-conceptualization and not todifferences in peripheral sensitivity."Five additional studies, using the Har-dy-Wolff-Goodell technique, whichalso failed to find consistency in thepain threshold have been recently re-viewed by Beecher (1957).

Other workers using other methodshave also found wide variability in the


pain threshold. Although Hardy etal. (1940) had reported that allthreshold measurements of their 3 5s(themselves) fell within +12% ofthe mean, Lanier (1943), using anelectric shock stimulus, found thatthe threshold of 15 college womenshowed a variation around the meanof -80 to +300%. He also foundthat some 5s showed a relativelyconstant threshold while othersshowed wide variations in their painthreshold at different times. Clarkand Bindra (1956), using thermal,electrical, and mechanical stimuli,have demonstrated wide individualdifferences in the pain threshold of46 untrained Ss. They attribute thesevariations to "attitudinal" variablessuch as the definition of pain, set,anxiety, and timidity.

After reviewing the many investi-gations in this area, Beecher (1957,p. 128) writes that "a survey of theabundant literature on the subjectpresented above forces one to con-clude that the pain threshold is notconstant from one individual to an-other nor even in a given individualfrom one time to another." Simi-larly, Kutscher and Kutscher (1957)conclude, after reviewing the litera-ture, that the pain threshold varieswidely among human beings, pro-vided that a sufficiently large groupof 5s is tested.

The second conclusion that appearsto have been established by theseinvestigations is that the pain thresh-old can be easily influenced by vary-ing the instructions (Hall & Stride,1954), by a wide variety of "dis-tractions," and by placebos, anal-gesics, and hypnosis. Wolff and Good-ell (1943) had earlier demonstratedthat placebos, in some cases, elevatedthe pain threshold as much as 95%,that the distraction caused by retain-ing and repeating from 5 to 9 digits

raised the threshold as much as 45%and that "shallow hypnosis" ele-vated the pain threshold by 40%.Subsequent work on the effects ofanalgesics and placebos on pain thresh-old will be discussed in the followingsection of this paper.

Kutscher and Kutscher (1957)have noted that the pain thresholdcan be significantly influenced bythe operator administering the test.A report by Denton and Beecher(1949) indicates that this observationmay be valid. Having failed to findany consistent effect of analgesicagents on pain threshold in trainedsubjects, these investigators re-quested the service of an individualwho had had wide experience withthe Hardy-Wolff-Goodell method.They found that this operator re-ported consistent elevations in thethreshold, after the administrationof an analgesic drug, when he knewwhich drug—a placebo or an anal-gesic—had been administered; how-ever, when he did not know whetheran effective drug or a placebo hadbeen administered, he was unable toreport any consistent threshold ele-vation.

That the pain threshold can bereadily influenced by a wide varietyof factors is not surprising if we stopto consider that determination of thehuman pain threshold does not evenremotely resemble the determinationof threshold responses of nerve fibers,nerve trunks, or other isolated physi-ological units; determination of thehuman pain threshold obviously re-quires judgment or interpretation onthe part of the 5. The 5 must inter-pret the stimulus in accordance withhis concept of pain and interpreta-tion clearly depends on S's previouslife-history and especially his specifichistory in responding to similar or re-lated stimuli. In fact, the Hardy-


Wolff-Goodell method requires morethan that S simply judge when hefirst becomes aware of a stimulus;he is required to determine when thestimulus first undergoes a qualitativechange. Operationally, the Hardy-Wolff-Goodell "pricking pain thresh-old" refers to the 5's judgment thatthe feeling of warmth and heat has"swelled" and "drawn together" intoa "very small" and "barely percepti-ble prick" at the "exact end of the3-sec. exposure to the stimulus"(Hardy et al., 1952, p. 81). This"pricking" feeling must be inter-preted by 5 as different not onlyfrom the warmth and heat which pre-cede it but also from the "burning"which may be simultaneously pres-ent. It would indeed be suprising ifsuch an intricate judgment could notbe influenced by a wide variety offactors.

THE EFFECT OF OPIATESON THE PAIN RESPONSE

As pointed out above, morphineand other opiates give "pain relief"without necessarily altering the painthreshold. Although Wolff, Hardy,and Goodell (1940) reported that thepain threshold is consistently ele-vated after morphine, subsequent in-vestigations failed to confirm thisconclusion. Andrews (1943), Chap-man and Jones (1944), Denton andBeecher (1949), Isbell (cited byWikler, 1950), Javert and Hardy(1951), and Kuhn and Bromiley(1951) found that after an analgesicdose of morphine the pain thresholdmay be elevated, may be lowered,or may remain unchanged.

A related line of research com-paring placebos and analgesic drugsarrived at similar results. Denton andBeecher (1949) found, using theHardy-Wolff-Goodell method, thata placebo had the same effect on

pain threshold as an analgesic doseof morphine. Similarly, Birren,Schapiro, and Miller (1950) reportedthat a placebo (lactose) had the sameeffect on pain threshold as 0.6 gm.of acetylsalicylic acid and sodiumphenobarbital. Isbell (cited by Wik-ler, 1950) also found no significantdifference in the effect on the painthreshold when 5s received morphineand when they received a placebobut were told they were being givenmorphine. Beecher (1957) has re-viewed 10 additional investigationswhich also indicate that morphineand other opiates (a) do not neces-sarily elevate the pain threshold whenthey give "pain relief" and (b) if andwhen they do elevate the pain thresh-old, they do so to the same extentand possibly in the same way asplacebos.

The evidence also indicates thatopiates can give "pain relief" with-out altering "the awareness of pain"or "the pain sensation." Cattell(1943) has summarized the data in-dicating that "awareness of pain" isnot necessarily altered by narcotics.Wolff et al. (1940, p. 677) have em-phasized that after morphine admin-istration "the pain sensation is per-ceived and is recognized as pain withno difficulties." Apparently, "thesensation of pain," in itself, is notnecessarily "painful." "The sensa-tion of pain" may be completely Un-affected by morphine (and placebos,hypnosis, prefrontal leucotomy, etc.)and yet discomfort and suffering areno longer present.

The "pain relief," i.e., the mitiga-tion of discomfort and suffering,which follows the administration ofmorphine and other opiates appearsto be one component of a more gener-alized effect on the patient which hasbeen variously conceptualized as"freedom from anxiety," "content-


ment," and "a bemused state." Thisviewpoint is perhaps best epitomizedby Beecher (1957, p. 152) who writesafter extensive clinical and experi-mental experience with analgesicdrugs that "perhaps one can con-clude that the narcotics really alterpain perception very little but doproduce a bemused state, comparableto distraction, which they can be'alerted out of and will then reporton the little altered pain perception."Along similar lines, Hill, Kornetsky,Flanary, and Wikler (1952a) havehypothesized that the "pain relief"following morphine administration isa consequence of a more generalizedeffect which they term relief of"anxiety" or "fear of pain." Theytested this hypothesis by studyingthe effect of subcutaneous injectionof 15 mg. of morphine on 5"s abilityto judge the intensity of electricshock stimuli under two conditions:(a) when 5s were made "anxious"by not "familiarizing them with thepotentially fear inspiring experimentalsituation," and (&) when "anxiety"was allayed by "reassurance, demon-stration, and explanation." Theyreported the following:

(a) Under conditions which promote anx-iety or fear of pain, subjects tend to overesti-mate the intensities of painful stimuli; (b)morphine reduces such anxiety; (c) underconditions in which anxiety is largely elimi-nated, little if any overestimation of the in-tensities of painful stimuli occurs; (d) mor-phine does not affect the ability of subjects toaccurately estimate the intensities of painfulstimuli when anxiety is dissipated (p. 479).

Corroborative data were obtainedin another study by the same groupof investigators (Hill, Kornetsky,Flanary, & Wikler, 1952b). In anadditional follow-up experiment, us-ing thermal stimuli, Kornetsky (1954)also confirmed these results and con-cluded that morphine appears to beeffective as an analgesic agent only

when "anxiety" is present.In summary, the investigations on

narcotics suggest a similar conclusionas the investigations on prefrontalleucotomy and hypnosis which weresummarized in an earlier section ofthis paper: discomfort and sufferingare not inevitably associated withnoxious stimulation; they appear tobe components of a secondary "reac-tion to" the stimulus (which hasbeen conceptualized as "anxiety" or"fear of pain") which can be mini-mized or eliminated by opiates hyp-nosis, placebos, prefrontal leucotomy,and other procedures.

THE PLACEBO EFFECTThe effect of placebos on the pain

response deserves further comment.Jellinek (1946) reported that 60% of199 patients with chronic headachesreceived "relief" from a placebo onone or more occasions. In extensivestudies of severe, steady, postopera-tive wound pain, Beecher (1955) andhis collaborators (Lasagna, Mosteller,von Felsinger, & Beecher, 1954) foundthat about 35% of their patients re-ceived "satisfactory" relief from aplacebo.11 ("Satisfactory relief" isdefined by these workers as "50 percent or more relief of pain at 45 and90 minutes after the administrationof the agent.") Houde and Wallen-stein-(1953) and Keats (1956) havecarried out similar studies and ,haveconfirmed the findings of the Beechergroup.

How does a placebo relieve chronicheadache or minimize the suffering

11 This finding does not indicate that pla-cebos are only 35% as effective as morphine.Morphine, in maximum safe dosages, resultsin "satisfactory" postoperative pain relief inonly 75% of the same group of patients(Lasagna & Beecher, 1954). The placebo,therefore, is about half as effective as mor-phine in the same situation and among thesame patients.


associated with a postoperativewound? As a first approximation toan answer, it seems difficult to dis-agree with Wolf's (1950, pp. 106-108)conclusion:

The above "placebo" actions depended fortheir force on the conviction of the patientthat this or that effect would result. . . . Thefact that "placebo effects" occur depends, ofcourse, on the generalization established re-peatedly by numerous workers that themechanisms of the human body are capable ofreacting not only to direct physical and chemi-cal stimulation but also to symbolic stimuli,words and events which have somehowaquired special meaning for the individual.

In general, the above studies andthe many other studies on the effectsof placebos on physiological func-tions and in psychotherapeutic situa-tions, reviewed by Beecher (1955),Rosenthal and Frank (1956), andKurland (1957) indicate that theplacebo reactor is responding to a"drug" which he believes has cura-tive properties. This belief appearsto be a function of many factors:what the physician specifically tellsthe patient about the "drug," thepatient's previous experience withdrugs, his previous experience withphysicians, his specific experiencewith the physician giving him the"drug," etc. The placebo responsemay be viewed as a direct functionof "the stimulus"; however, "thestimulus" is not the ineffective, inertcompound but the entire situationwhich includes the "drug," the wordsof the physician, and the patient'sprevious experience with physiciansand drugs.

Placebo research is still in its in-fancy. As Kurland (1957) has pointedout, the effect of the placebo isusually stated in general terms, theduration of reactivity is usually notspecified, and specific physiologicalmeasures are rarely reported. Studiesof the differential effect of placebos

are also rare. Are some individualsmore prone to respond to placebos?Lasagna et al. (1954) studied 27 post-operative patients with the Ror-schach, the TAT, the Wechsler-Belle-vue Vocabulary Subtest, and a ques-tionnaire filled out by the nurses onthe wards. The 11 consistent placeboreactors differed from the 16 patientswho never received pain relief froma placebo in a number of character-istics, among which were the follow-ing:

The reactors were more productive of re-sponses, more anxious, more self-centered andpreoccupied with internal bodily processes,and more emotionally labile. They are indi-viduals who seem more dependent on outsidestimulation than on their own mental proc-esses. These processes tend to be less maturethan in the case of the non-reactors. Thereactors are in general individuals whose in-stinctual needs are greater and whose controlover the social expression of these needs isless strongly defined and developed than inthe non-reactors ... (p. 775).

However, Wolf, Doering, Clark,and Hagens (1957) contradict theseconclusions: finding that intra-indi-vidual variations in response to place-bos are as great as interindividualvariations, they conclude that theplacebo reactor cannot be predictedfrom a knowledge of the S's char-acteristics. An interesting field ofresearch has been opened for furtherinquiry.

CONCLUSIONSThe investigations summarized

above suggest the following conclu-sions which may be significant for atheory of pain:

1. The generally accepted view,that "pain" has its "own" peripheralreceptors and its "own" pathwaysin the central nervous system, is mis-leading. Nociceptive stimuli activatevarious types of nerve fibers whichtravel in more than one pathway in


the spinal cord and brain stem andwhich project by thalamic and extra-thalamic pathways to wide areas ofthe cortex.

2. The response to a nociceptivestimulus is apparently brought aboutwhen a spatiotemporal pattern ofneural activity set off by the noxiousstimulus reaches segmental and supra-segmental centers. The pattern ofneural impulses set off by noxiousstimuli differs from the neural pat-tern set off by other stimuli in thatthe relative number of fibers of differ-ent sizes activated differ, and therelatively different fibers activatedcarry impulses of different energyvalue, of different frequency, and ofdifferent duration.

3. "Pain" in the sense of discom-fort and suffering is not necessarilypresent when noxious stimuli arediscriminated, differentiated, and lo-calized. The few cases which havebeen reported of "congenital insensi-tivity to pain" suggest that an indi-vidual may be able to "sense" anoxious stimulus—i.e., may be ableto discriminate and localize the stimu-lus and differentiate it from otherstimuli—and yet not show with-drawal movements, physiological al-terations, or discomfort. Also, dis-comfort and suffering can be mini-

mized or totally eliminated in some5s by placebos, opiates, prefrontalleucotomy, and hypnotic procedureswithout necessarily altering the "sen-sation of pain" or elevating the painthreshold.

4. The mitigation of discomfort-suffering by prefrontal leucotomy,opiates, and, to some extent, hyp-nosis, appears to be secondary toa more generalized effect of theseprocedures. Prefrontal leucotomy"alleviates worry and concern" and"relieves anxiety"; morphine gives"freedom from anxiety" and bringsabout "contentment" and "a be-mused state"; the hypnotic 5 is re-lieved of pain when he becomes"relatively inattentive and uncon-cerned about all stimuli to which thehypnotist does not specifically directhis attention." These terms appearto refer to a common behavioral ma-trix: a mitigated "readiness to re-spond" to stimulation. Apparently,discomfort and suffering follow noci-ceptive stimulation when the 5 "at-tends to" and "reacts to" the stimu-lus. Minimize this readiness to re-spond and "the sensation of pain" isno longer "painful"; it can become anisolated "sensation" unaccompaniedby discomfort.

REFERENCES

ADEY, W. R. Somatic aspects of the nervoussystem. Annu. Rev. Physiol., 1957, 19, 489-512.

ANDREW, H. L. The effect of opiates on thepain threshold in post-addicts. /.• din. In-vest., 1943, 22, 511-516.

ARBUSE, D. I., CANTOR, M. B., & BAREN-BERG, P. A. Congenital indifference topain. J. Pediat., 1949, 35, 221-226.

ARDUINI, A., & ARDUINI, G. Effects of drugsand metabolic alterations on brain stemarousal mechanism. /. Pharmacol,, 1954,110, 76-85.

ARNOLD, MAGDA. The status of emotion incontemporary psychology. In A. A. Roback

(Ed.) Present-day psychology. New York:Phil. Lib., 1955. Pp. 135-188.

BARBER, T. X. Hypnosis as perceptual-cog-nitive restructuring: I. Analysis, of con-cepts. /. din. exp. Hypn., 1957, 5, 147-166.(a)

BARBER, T. X. Hypnosis as perceptual-cog-nitive restructing: III. From somnambu-lism to autohypnosis. J. Psychol., 1957, 44,299-304. (b)

BARBER, T. X. Hypnosis as perceptual-cog-nitive restructurting: IV. "Negative hal-lucinations." /. Psychol., 1958,46,187-201.(a)

BARBER, T. X. The concept of "hypnosis."


J. Psychol., 1958, 45, 115-131. (b)BARBER, T. X. The "good" hypnotic subject.

Set. Dig., 1958, 43 (1), 36-41. (c)BARBER, T. X., & COULES, J. Electrical skin

conductance and galvanic skin responseduring "hypnosis." J. din. exp. Hypn.,1959, 7, 79-92.

BARD, P., & MACHT, M. B. The behavior ofchronically decerebrate cats. In Ciba sym-posium: Neurological basis of behavior. Bos-ton: Little, Brown, 1958. Pp. 55-71.

BEECHER, H. K. The powerful placebo. J.Amer. Med. Ass., 1955, 159, 1602-1606.

BEECHER, H. K. The measurement of pain:Prototype for the quantitative study of sub-jective responses. Pharmacol. Rev., 1957, 9,59-209.

BELL, E., JR., & KARNOSH, L. J. Cerebralhemispherectomy: Report of a case tenyears after operation. /. Neurosurg,, 1949,6, 285-293.

BENDER, M. B., & JAFFE, R. Pain of cerebralorigin. Med. Clin. North Amer., 1958, 42,691-700.

BENJAMIN, F. B., & IVY, A. C. Electroen-cephalographic changes associated withpainful and non-painful peripheral stimula-tion. Proc. Soc. exp. Biol. Med., 1949, 72,420-421.

BIRREN, J. E., SHAPIRO, H. B., & MILLER,J. H. The effect of salicylate upon painsensitivity. J. Pharmacol., 1950,100,67-71.

BISHOP, G. H. The peripheral unit for pain.J. Neurophysiol., 1944, 7, 71-80.

BISHOP, G. H. Neural mechanism of cuta-neous sense. Physiol. Rev., 1946, 26, 77-102.

BONICA, J. J. The management of pain. Phil-adelphia: Lea & Febiger, 1953.

BONNER, FRANCES, COBB, S., SWEET, W. H.,& WHITE, J. C. Frontal lobe surgery in thetreatment of pain. Psychosom. Med., 1952,14, 383-405.

BOWSHER, D. Termination of the central painpathway in man: The conscious apprecia-tion of pain. Brain, 1957, 80, 606-622.

BOYD, D. A., & NIE, L. W. Congenital uni-versal indifference to pain. Arch. Neural.Psychiat., 1949, 61, 402-411.

BROOKHART, J. M., LIVINGSTON, W. K., &HAUGEN, F. P. Functional characteristicsof afferent fibers from tooth pulp of cat. J.Neurophysiol., 1953, 16, 634-642.

BROWDER, J., & GALLAGHER, J. P. Dorsalcordotomy for painful phantom limb. Ann.Surg., 1948, 128, 456-469.

BROWN, R. R., & VOGEL, V. H. Psychophysio-logical reactions following painful stimuliunder hypnotic analgesia, contrasted withgas anesthesia and novocain block. /. appl.Psychol., 1938, 22, 408-420.

BUTLER, B. The use of hypnosis in the care ofthe cancer patient. Cancer, 1954, 7, 1-14.

CANNON, W. B., NEWTON, H. F., BRIGHT,E. M., MENKIN, V., & MOORE, R. M. Someaspects of the physiology of animals surviv-ing complete exclusion of sympatheticnerve impulses. Amer. J. Physiol., 1929,89, 84-107.

CATTELL, M. The action and use of anal-gesics. Res. Publ. Ass. nerv. ment. Dis.,1943, 23, 365-372.

CHAPMAN, W. P., & JONES, C. M. Variationsin cutaneous and visceral pain sensitivity innormal subjects. /. din. Invest., 1944, 23,81-91.

CHAPMAN, W. P., SOLOMON, H. C., & ROSE,A. S. Measurements of motor withdrawalreaction in patients following frontal lobot-omy. In M. Greenblatt, R. Arnot, & H. C.Solomon (Eds.) Studies in lobotomy. NewYork: Grune & Stratton, 1950. PP. 386-392.

CLARK, J. W., & BINDRA, D. Individual dif-ferences in pain threshold Canad. J. Psy-chol, 1956, 10, 69-76.

COHEN, L. D., KIPNIS, D., KUNKLE, E. C., &KUBZANSKY, P. E. Observations of a per-son with congenital insensitivity to pain.J. abnorm. sac. Psychol., 1955, SI, 333-338.

CRITCHLEY, M. Congenital indifference topain. Ann. intern, Med., 1956, 45, 737-747.

DANDY, W. E. Physiological studies followingextirpation of the right cerebral hemispherein man. Bull. Johns Hopk. Hasp., 1933, 53,31-51.

DAVISON, C., & SCHICK, W. Spontaneouspain and other subjective sensory disturb-ances. Arch. Neural. Psychiat., 1935, 34,1204-1237.

DE JONG, R. N. Discussion of "Congenitaluniversal indifference to pain," by D. A.Boyd and L. W. Nie. Arch. Neural. Psy-chiat., 1949, 61, 411-412.

DENTON, J. E., & BEECHER, H K. New anal-gesics. I. Methods in the clinical evaluationof new analgesics. /. Amer. Med. Ass., 1949,141, 1051-1057.

DORCUS, R. M., & KIRKNER, F. J. The use ofhypnosis in the suppression of intractablepain. J. abnorm. soc. Psychol., 1948, 43,237-239.

DOUPE, J., MILLER, W. R., & KELLER, W. K.Vasomotor reactions in the hypnotic state./. Neural. Psychiat., 1939, 2, 97-102.

DYNES, J. B. An experimental study of hyp-notic anesthesia. /. abnorm. soc. Psychol.,1932, 27, 79-88.

DYNES, J. B., & POPPEN, J. L. Lobotomy forintractable pain. J. Amer. med. Ass., 1949,140, 15-19.


EDWARDS, W. Recent research on pain per-ception. Psychol. Bull., 1950, 47, 449-474.

ELITHORN, A., GLITHERO, E., & SLATER, E.Leucotomy for pain. /. Neural. Neurosurg.Psychiat., 1958, 21, 249-261.

ERICKSON, M. H. The induction of color-blindness by a technique of hypnotic sug-gestion. /. gen. Psychol., 1939, 20, 61-69.

FARQUHAR, H. G., & SUTTON, T. Congenitalindifference to pain. Lancet, 1951, 1, 827-828.

FEINDEL, W. Note on the nerve endings in asubject with arthropathy and congenital in-difference to pain. /. Bone Jt. Surg., 1953,35-b, 402-407.

FORD, F. R., & WILKINS, L. Congenital uni-versal insensitiveness to pain. Johns Hopk.Hasp. Bull., 1938, 62, 448-466.

FREEMAN, W. Discussion of "Lobotomy forintractable pain" by J. B. Dynes and J. L.Poppen. /. Amer. Med. Ass., 1949,140,18.

FREEMAN, W., & WATTS, J. W. Psychosur-gery—in the treatment of mental disordersand intractable pain. (2nd ed.) Springfield,111.: C.C Thomas, 1950.

FREEMAN, W., & WILLIAMS, J. M. The le-sions of transorbital lobotomy. Trans.Amer. Neural Ass., 1951, 76, 236-237.

FRENCH, J. D., & KING, E. E. Mechanismsinvolved in the anesthestic state. Surgery,1955, 38, 228-238.

FRENCH, J. D., VERZEANO, M., & MAGOUN,W. H. Neural basis of anesthetic state.Arch. Neural. Psychiat., 1953, 69, 519-529.

FRENCH, L. A., & PEYTON, W. T. Ipsilateralsensory loss following cordotomy. J. Neuro-surg., 1948, 5, 403-404.

FULTON, J. F. Frontal lobotomy and affectivebehavior: A neurophysiological analysis.New York: Norton, 1951.

PURER, M., & HARDY, J. D. The reaction topain as determined by the galvanic skin re-sponse. Res. Publ. Ass. Nerv. Ment. Dis.,1950, 29, 72-89.

GARDNER, W. J. Removal of the right cere-bral hemisphere for infiltrating glioma. J.Amer. Med. Ass., 1933, 101, 823-825.

GASSER, H. S. Control of excitation in thenervous system. Bull. N. Y. Acad. Med.,1937, 13, 324-348.

GASSER, H. S. Pain-producing impulses inperipheral nerves. Res. Publ. Ass. Nerv.Ment. Dis., 1943, 23, 44-62.

GELLHORN, E., & BALLIN, H. M. The effectof afferent impulses on hypothalamic poten-tials. Amer. J. Physiol, 1946, 146, 630-635,

GEORGI, F. Beitrage zur Kenntnis des psy-cho-galvanischen Phanomens. Arch. Psy-chiat., 1921, 62, 571.

GILLIATT, R. W. & PRATT, R. T. C. Disorders

of perception and performance in a case ofright-sided cerebral thrombosis. J. Neural.Neurosurg. Psychiat., 1952, IS, 264-271.

GIRARD, P. F., DEVIC, M., & GARIN, A. Apropos d'une observation nouvelle d'in-diffeYence congdnitale universelle a la dou-leur. Rev. Neural., 1953, 88, 198-201.

GOETZL, F. R., BIEN, C. W., & Lu, G.Changes in blood pressure in response topresumably painful stimuli, /. appl, Phy-siol., 1951, 4, 161-170.

CRIMSON, K. S., ORGAIN, E. S., ANDERSON, B.,BROOME, R. A., JR., & LONGINO, F. H. Re-sults of treatment of patient with hyperten-sion by total thoracic and partial to totallumbar sympathectomy, splanchnicectomyand celiac ganglionectomy. Ann. Surg.,1949, 129, 850-871.

HAGBARTH, K.-E., & KERR, D. I. B. Centralinfluences on spinal afferent conduction. /.Neurophysiol., 1954, 17, 295-307.

HALL, K. R. L., & STRIDE, E. The varyingresponse to pain in psychiatric disorders: Astudy in abnormal psychology. Brit. J. med.Psychol., 1954, 27, 48-60.

HARDY, J. D., WOLFF, H. G., & GOODELL,HELEN. Studies on pain. A new methodfor measuring pain threshold; observationson spatial summation of pain. /. din. In-vest., 1940, 19, 649-657.

HARDY, J. D., WOLFF, H. G., & GOODELL,HELEN. Pain sensations and reactions.Baltimore: Williams & Wilkins, 1952.

HAUGEN, F. P., & MELZACK, R. The effectsof nitrous oxide on responses evoked in thebrain stem by tooth stimulation. Anes-thesiology, 1957, 18, 183-195.

HAWKES, C. D., & GOTTEN, N. Prefrontallobotomy vs. cordotomy for relief of intract-able pain. Trans. Amer. Neural. Ass., 1948,73, 208-210.

HEINBECKER, P., BISHOP, G. H., & O'LEARY,J. Pain and touch fibers in peripheralnerves. Arch. Neural. Psychiat., 1933, 29,771-789.

HEMPHILL, R. E., & STENGEL, E. A studyof pure word-deafness. -A. Neural. Neuro-surg. Psychiat., 1940, 3, 251-262.

HILL, H. E., KORNETSKY, C. H., FLANARY,H. G., & WILKER, A. Effects of anxiety andmorphine on discrimination of intensities ofpainful stimuli. /. din. Invest., 1952, 31,473-480. (a)

HILL, H. E., KORNETSKY, C. H., FLANARY,H. G., & WILKER, A. Studies on anxietyassociated with anticipation of pain. I.Effects of morphine. Arch. Neural. Psy-chiat., 1952, 67, 612-619. (b)

HORRAX, G. Experiences with cortical ex-


cisions for relief of intractable pain in ex-tremities. Surgery, 1946, 20, 593-602.

HOUDE, R. W., & WALLENSTEIN, S. L. Amethod for evaluating analgesics in pa-tients with chronic pain. Drug Addict.Narcot. Bull, 1953, App. F, 660-682.

HULL, C. L. Hypnosis and suggestibility—anexperimental approach. New York: Apple-ton-Cenutry, 1933.

JAVERT, C. T., & HARDY, J. D. Influence ofanalgesics on pain intensity during labor(with a note on "natural childbirth").Anesthesiology, 1951, 12, 189-215.

JELLINEK, E. M. Clinical tests on compara-tive effectiveness of analgesic drugs. Bio-met, Bull., 1946, 2, 87.

JEQUIER, M., & DELLER, M. L'indifferenceconge'nitale &. la douleur. Conf. Neural.,1956, 16, 207-215.

JEWESBURY, E. C. O. Intensitivity to pain.Brain, 1951, 74, 336-353.

JONES, MARGARET H. Second pain: Fact orartifact. Science, 1956, 124, 442-443.

JONES, MARGARET H. Reply to Bishop andLandau. Science, 1958, 128, 713-714.

KEATS, A. S. Postoperative pain; researchand treatment. /. chronic Dis., 1956, 4, 72-83.

KEELE, K. D. Anatomies of pain. Springfield,111.: C. C Thomas, 1957.

KING, H. E., CLAUSEN, J., & SCARFF, J. E.Cutaneous thresholds for pain before andafter unilateral prefrontal lobotomy. /.nerv. ment. Dis., 1950, 112, 93-96.

KING, R. B. Postchordotomy studies of painthreshold. Neurology, 1957, 7, 610-614.

KORNETSKY, C. Effects of anxiety and mor-phine on the anticipation and perception ofpainful radiant thermal stimuli. /. camp,physiol. Psychol, 1954, 47, 130-132.

KOSKOFF, Y. D., DENNIS, W., LAZOVIK, D., &WHEELER, E. T. The psychological effectsof frontal lobotomy for the alleviation ofpain. Res. Publ. Ass. Nerv. Ment. Dis.,1948, 27, 723-753.

KRAYENBUHL, H. & STOLL, W. A. Psycho-chirugie bei unertraglichen Schmerzen.Acta Neurochirug., 1950, 1, 1-41.

KROGER, W. S. Introduction and supplemen-tal reports. In J. Esdaile Hypnosis in medi-cine and surgery. New York: Julian, 1957.Pp. i-xxxvii.

KRYNAUW, R. A. Infantile hemiplegia treatedby removing one cerebral hemisphere. J.Neural. Neurosurg. Psychiat., 1950, 13, 243-267.

KUHN, R. A., & BROMILEY, R. B. Humanpain thresholds determined by the radiantheat technique and the effect upon them ofacetylsalicylic acid, morphine sulfate, and

sodium phenobarbital. /. Pharmacol., 1951,101, 47-55.

KUNKLE, E. C., & CHAPMAN, W. P. Insensi-tivity to pain in man. Res. Publ. Ass. Nerv.Ment. Dis., 1943, 23, 100-109.

KURLAND, A. A. The drug placebo—its psy-chodynamic and conditional reflex action.Behav. Sci., 1957, 2, 101-110.

KUTSCHER, A. H., & KUTSCHER, H. W.Evaluation of the Hardy-Wolff-Goodellpain threshold apparatus and technique:Review of the literature. Int. Rec. Med,,1957, 170, 202-212.

LANIER, L. H. Variability in the pain thresh-old. Science, 1943, 97, 49-50.

LASAGNA, L., & BEECHER, H. K. The optimaldose of morphine. /. Amer. Med. Ass.,1954, 156, 230-234.

LASAGNA, L., MOSTELLER, F., VON FEL-SINGER, J. M., & BEECHER, H. K. A studyof the placebo response. Amer. J. Med.,1954, 16, 770-779.

LE BEAU, J. Experience with topectomy forthe relief of intractable pain. /. Neurosurg.,1950, 7, 79-91.

LELE, P. P., & WEDDELL, G. The relationshipbetween neurohistology and corneal sen-sibility. Brain, 1956, 79, 119-154.

LELE, P. P., WEDDELL, G., & WILLIAMS, C. M.The relationship between heat transfer,skin temperature, and cutaneous sensibility.J. Psychiol, 1954, 126, 206-234.

LEUBA, C. The reality of hypnotic phenom-ena: A critique of the role playing theory ofhypnosis. J. din. exp. Hypn., 1957, 5, 32-38.

LEVINE, M. Psychogalvanic reaction to pain-ful stimuli in hypnotic and hystericalanesthesia. Butt. Johns Hopk. Hasp., 1930,46, 331-339.

LEWIN, W., & PHILLIPS, C. G. Observationson partial removal of the post-cental gyrusfor pain. /. Neural. Neurosurg. Psychiat.,1952, IS, 143-147.

LHERMITTE, J., & PUECH. L'algo-hallucinosedes amputfe. Traitement par la r&ectiondu ndvrone, 1'infiltration de la chaine sym-pathique, une double my&otomie pos-t&ieure, la r&ection du lobule parietal su-perieur. Rev. Neural., 1946, 78, 33-35.

LIVINGSTON, W. K. Pain mechanisms. NewYork: Macmillan, 1943.

LLOYD, D. P. C. Special physiology of nervesand tracts. In J. F. Fulton (Ed.) A text-book of physiology. Philadelphia: Saunders,1955. Pp. 43-58.

McMuRRAY, G. A. Experimental study of acase of insensitivity to pain. Arch. Neural.Psychiat., 1950, 64, 650-667.

McMuRRAY, G. A. Congenital insensitivity


to pain and its implications for motiva-tional theory. Canad. J. Psychol., 1955, 9,121-131.

MADONICK, M, J. Insensitiveness to pain.Neurology, 1954, 4, 554-557.

MAGOUN, H. W. The waking brain. Spring-field, 111. :C.C Thomas, 1958.

MAUCUSE, F. L., & PHILLIPS, G. T. A demon-stration of dental extraction with hypnoticanesthesia. /. din. exp. Hypn., 1956,4, 2-4.

MARSHALL, C., & WALKER, A. E. The electro-encephalographic changes after hemi-spherectomy in man. EEG din. Neurophys-iol., 1950, 2, 147-156.

MARSHALL, J. Sensory disturbances in corti-cal wounds with special reference to pain.J. Neural. Neurosurg. Psychiat., 1951, 14,187-204.

MASON, A. A. Surgery under hypnosis. Anes-thesia, 1955, 10, 295-299.

MELZACK, R., & SCOTT, T. H. The effects ofearly experience on the response to pain./. comp. physiol. Psychol., 1957, 50, 155-161.

MELZACK, R., STOTLER, W. A., & LIVINGSTON,W. K. Effects of discrete brainstem lesionsin cats on perception of noxious stimula-tion. J. Neurophysiol., 1958, 21, 353-367.

MEYER, A., & BECK, E. Neuropathologicalproblems arising from prefrontal leucotomy./. went. Sci., 1945, 91, 413-425.

MICHELSON, J. J. Subjective disturbance ofthe sense of pain from lesions of the cere-bral cortex. Res. Publ. Ass. Nen. Ment.Dis., 1943, 23, 86-89.

MURPHY, J. P., & GELLHORN, E. J. Furtherinvestigations on diencephalic cortical re-lations and their significance for the prob-lem of emotion. /. Neurophysiol., 1945, 8,431-447.

NAPE, J. P., & WAGONER, K. S. The sensi-tivity of the cornea of the eye. J. Psychol.,1936, 2,433-439.

NAPE, J. P., & WAGONER, K. S. The effect ofpain upon systemic arterial blood pressure.Amer. J. Psychol., 1938, 51, 390-397.

ORNE, M. T. The mechanism of hypnotic ageregression: An experimental study. /,abnorm. soc. Psychol., 1951, 46, 213-225.

OSTENASEK, F. J. Prefrontal lobotomy forthe relief of intractable pain. Johns Hopk.Hasp. Bull., 1948, 83, 229-236.

PENFIELD, W., & BOLDREY, E. Somatic motorand sensory representation in the cerebralcortex of man studied by electrical stimula-tion. Brain, 1937, 60, 389-443.

PENFIELD, W., & RASMUSSEN, T. The cere-bral cortex of man: A clinical study of lo-calization of function. New York: Mac-millan, 1950.

PETERSON, C. G. Neuropharmacology ofprocaine. II. Central nervous actions.Anesthesiology, 1955, 16, 976-993.

PETIT-DUTAILLIS, D., MESSIMY, R., & BER-GES, L. La psycho-chirurgie des algies ir-reVluctibles. Sent. Hip. Paris, 1953, 29,3893-3903.

PETRIE, ASENATH. A comparison of the psy-chological effects of two different types ofincision on the frontal lobes. Proc. 13th Int.Cong. Psychol., Stockholm, 1951.

PETRIE, ASENATH. Personality and the frontallobes. London: Routledge & Paul, 1952.

PETRIE, ASENATH. Effects of chlorpromazineand of brain lesions on personality. InH. D. Pennes (Ed.) Psychopharmacology.New York: Hoeber-Harper, 1958. Pp. 99-115.

POOL, J. L. Posterior cordotomy for relief ofphantom limb pain. Ann. Surg., 1946, 124,386-391.

RANSON, S. W. The anatomy of the nervoussystem. Philadelphia: Saunders, 1943.

RAY, B. S., & CONSOLE, A. D. Evaluation oftotal sympathectomy. Ann. Surg., 1949,130A, 652-673.

ROSE, G. K. Arthropathy of the ankle in con-genital indifference to pain. /. Bone Jt.Surg., 1953, 35-B, 408-410.

ROSEN, H. The hypnotic and hypnothera-peutic control of severe pain. Amer. J.Psychiat., 1951, 107, 917-925.

ROSENTHAL, D., & FRANK, J. D. Psycho-therapy and the placebo effect. Psychol.Bull., 1956, 53, 294-302.

ROWE, S. N. Mental changes following theremoval of the right cerebral hemispherefor brain tumor. Amer. J. Psychiat., 1937,94, 604-612.

RUBINS, J. L., & FRIEDMAN, E. D. Asym-bolia for pain. Arch. Neural. Psychiat.,1948, 60, 554-573.

RUCH, T. C. Neural basis of somatic sensa-tion. In J. F. Fulton (Ed.) A textbook ofpsysiology. Philadelphia: Saunders, 1955.Pp. 328-357.

SARGANT, W., & SLATER, E. An introduction tophysical methods of treatment in psychiatry.(3d ed.) Edinburgh: Livingstone, 1954.

SATTLER, D. G. Absence of local sign in vis-ceral reactions to painful stimulation. Res.Publ. Ass. Nerv. Ment. Dis., 1943, 23, 143-153.

SCHILDER, P., & STENGEL, E. Asymbolia forpain. Arch. Neural. Psychiat., 1931, 25,598-600.

SCHILLER, F. The cutaneous sensory modali-ties: A critique of their specificity. Arch.Neural. Psychiat., 1956, 75, 203-219.

SCHUMACHER, G. A., GOODELL, HELEN,


HARDY, J. D., & WOLFF, H. G. Uniformityof the pain threshold in man. Science, 1940,92, 110-112.

SEARS, R. R. Experimental study of hyp-notic anesthesia. /. exp. Psychol., 1932, IS,1-22.

SINCLAIR, D. C. Cutaneous sensation and thedoctrine of specific energy. Brain, 19SS, 78,584-614.

SINCLAIR, D. C., WEDDELL, G., & ZANDER, E.Relationship of cutaneous sensibility toneurohistology in human pinna. /. Anat.,1952,86,402-411.

STARZL, T. E., & WHITLOCK, D. G. Diffusethalamic projection system in moneky. /.Neurophysiol., 1952, IS, 449-468.

SWEET, W. H., WHITE, J. C., SELVERSTONE,B., & NILGES, R. Sensory responses fromanterior roots and from surface and interiorof spinal cord in man. Trans. Amer. Neural.Ass., 1950, 75, 165-169.

VORIS, H. C. Ipsilateral sensory loss followingchordotomy: Report of a case. Arch. Neu-rol. Psychiat., 1951, 65, 95-96.

WALKER, A. E. The spinothalamic tract inman. Arch. Neural. Psychiat., 1940, 43,284-298.

WALKER, A. E. Central representation ofpain. Res. Publ. Ass. Nero. Ment. Dis.,1943, 23, 63-85.

WALKER, A. E. The neurosurgical treatmentof intractable pain. J. Lancet, 1950, 70,279-282.

WALSH, E. G. Physiology of the nervous sys-tem. London: Longmans, 1957.

WEDDELL, G. Somesthesis and the chemicalsenses. Ann. Rev. Psychol., 1955, 6, 119-136.

WEITZENHOFFER, A. M. Hypnotism—an ob-jective study in suggestibility. New York:Wiley, 1953.

WEST, L. J., NEILL, K. C., & HARDY, J. D.Effects of hypnotic suggeston on pain per-ception and galvanic skin response. Arch.Neural. Psychiat., 1952, 68, 549-560.

WESTLAKE, E. K. Congenital indifference topain. Brit. med. J., 1952, 1, 144.

WHITE, J. C., & SWEET, W. H. Pain:Its mechanism and neurosurgical control.

Springfield, 111.: C. C Thomas, 1955.WIKLER, A. Sites and mechanisms of action

of morphine and related drugs in the cen-tral nervous system. Pharmacol. Rev., 1950,2, 435-506.

WINKELSTEIN, L. B. Routine hypnosis forobstetrical delivery: An evaluation of hyp-nosuggestion in 200 consecutive cases.Amer. J. Obstet. Gynec., 1958, 76, 152-160.

WOLF, S. Effects of suggestion and con-ditioning on the action of chemical agentsin human subjects: The pharmacology ofplacebos. /. din. Invest., 1950, 29, 100-119.

WOLF, S., DOERING, C. R., CLARK, M. L., &HAGANS, J. A. Chance distribution and theplacebo "reactor." J. lab. din. Med., 1957,49, 837-841.

WOLFF, H. G., & GOODELL, HELEN. The re-lations of attitude and suggestion to theperception of and reaction to pain. Res.Publ. Ass. Nerv. Ment. Dis., 1943, 23, 434-446.

WOLFF, H. G., HARDY, J. D., & GOODELL,HELEN. Studies on pain. Measurement ofthe effect of morphine, codeine, and otheropiates on the pain threshold and an analy-sis of their relation to the pain experience.J. din. Invest., 1940, 19, 659-680.

YAKOVLEV, P. I., HAMLIN, H., & SWEET,W. H. Anatomical studies of lobotomy. InM. Greenblatt, R. Arnot, & H. C. Solomon(Eds.) Studies in lobotomy. New York:Grune & Stratton, 1950. Pp. 309-329.

YOUNG, P. C. The nature of hypnosis, asindicated by the presence or absence ofpost-hypnotic amnesia and rapport. /. ab-norm. sac. Psychol., 1928, 22, 372-382.

YOUNG, P. C. Antisocial uses of hypnosis. InL. M. LeCron (Ed.) Experimental hypnosis.New York: Macmillan, 1952. Pp. 376-409.

ZOLLINGER, R. Removal of left cerebralhemisphere: Report of a case. Arch. Neural.Psychiat., 1935, 34, 1055-1064.

ZOTTERMAN, Y. Touch, pain and tickling:An electro-psysiological investigation oncutaneous sensory nerves. /. Physiol.,1939, 95, 1-28.

(Received January 14, 1959)

toward a theory of painelevation in the electroneurogram." if all cutaneous stimuli activate...

Documents