Top Banner
Psychological Review 1965, Vol. 72, No. S, 373-384 INFORMATION ABOUT SPATIAL LOCATION BASED ON KNOWLEDGE ABOUT EFFERENCE 1 LEON FESTINGER ANDLANCE KIRKPATRICK CANON Stanford University An experiment was designed to determine whether or not the human organism possessed "outflow" information derived from monitoring nerve impulses in motor pathways. The experiment focused on the extraocular muscles since proprioceptive input to the central nervous system from these muscles is poor. The results show that in the absence of good proprioceptive information, the presence or absence of "outflow" information makes a difference in accuracy of localizing an object in space. The human being continually ac- quires and uses information about him- self and his relation to the environ- ment. We are accustomed to thinking of this information as having been acquired through input to afferent mechanisms. That is, we know about the environment through seeing, hear- ing, touching, and a variety of other means. Not the least of these sources of information is input from proprio- ceptors. For example, if I am led blindfolded into a room and I touch an object in that room with my hand, I know where that object is in relation to my body because, on the basis of proprioceptive feedback, I know where my hand is. There is, however, another possible source of information about one's rela- tion to the environment that has not been adequately explored. If, in the central nervous system, outgoing motor nerve impulses are monitored and re- corded, then information would also exist concerning spatial location on the basis of this record of efferent impulses, 1 This research was supported by Grant No. MH 07835-01 from the National Insti- tutes of Health to the senior author. We wish to thank Douglas H. Lawrence and Gordon H. Bower for their help on the ex- periment. that is, a record of the specific direc- tions given to the musculature. This information, if it exists, need not rely on any current afferent input. To make this clear, let us illustrate by a loose analogy. Imagine there is a per- son who will unconditionally obey your orders. Let us also assume that you and the other person have had suffi- cient previous experience with the en- vironment so that you can give him, and he can follow, clear directions. You tell this person to go to a certain specific place and to wait there for you. Even in the complete absence of any current sensory input you will know exactly where that person is because you know where you told him to go. The question of whether or not such monitored efferent information exists is an old one in psychology although, of late, it has been rarely mentioned. Actually, a closely related speculation was vigorously debated many years ago. James (1950) 2 stated the issue clearly: There must, of course, be a special current of energy going out from the brain into the 2 We give the dates of the later editions from which we have quoted. The book by James was originally published in 1890, and the first edition of the book by Helmholtz was earlier than that. 373
12

Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

Aug 07, 2020

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

Psychological Review1965, Vol. 72, No. S, 373-384

INFORMATION ABOUT SPATIAL LOCATIONBASED ON KNOWLEDGE ABOUT

EFFERENCE1

LEON FESTINGER AND LANCE KIRKPATRICK CANON

Stanford University

An experiment was designed to determine whether or not the humanorganism possessed "outflow" information derived from monitoringnerve impulses in motor pathways. The experiment focused on theextraocular muscles since proprioceptive input to the central nervoussystem from these muscles is poor. The results show that in theabsence of good proprioceptive information, the presence or absence of"outflow" information makes a difference in accuracy of localizing anobject in space.

The human being continually ac-quires and uses information about him-self and his relation to the environ-ment. We are accustomed to thinkingof this information as having beenacquired through input to afferentmechanisms. That is, we know aboutthe environment through seeing, hear-ing, touching, and a variety of othermeans. Not the least of these sourcesof information is input from proprio-ceptors. For example, if I am ledblindfolded into a room and I touchan object in that room with my hand,I know where that object is in relationto my body because, on the basis ofproprioceptive feedback, I know wheremy hand is.

There is, however, another possiblesource of information about one's rela-tion to the environment that has notbeen adequately explored. If, in thecentral nervous system, outgoing motornerve impulses are monitored and re-corded, then information would alsoexist concerning spatial location on thebasis of this record of efferent impulses,

1 This research was supported by GrantNo. MH 07835-01 from the National Insti-tutes of Health to the senior author. Wewish to thank Douglas H. Lawrence andGordon H. Bower for their help on the ex-periment.

that is, a record of the specific direc-tions given to the musculature. Thisinformation, if it exists, need not relyon any current afferent input. Tomake this clear, let us illustrate by aloose analogy. Imagine there is a per-son who will unconditionally obey yourorders. Let us also assume that youand the other person have had suffi-cient previous experience with the en-vironment so that you can give him,and he can follow, clear directions.You tell this person to go to a certainspecific place and to wait there for you.Even in the complete absence of anycurrent sensory input you will knowexactly where that person is becauseyou know where you told him to go.

The question of whether or not suchmonitored efferent information exists isan old one in psychology although, oflate, it has been rarely mentioned.Actually, a closely related speculationwas vigorously debated many yearsago. James (1950)2 stated the issueclearly:

There must, of course, be a special currentof energy going out from the brain into the

2 We give the dates of the later editionsfrom which we have quoted. The book byJames was originally published in 1890, andthe first edition of the book by Helmholtzwas earlier than that.

373

Page 2: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

374 LEON FESTINGER AND LANCE KIRKPATRICK CANON

appropriate muscles during the act; and thisoutgoing current (it is supposed) must havein each particular case a feeling sui generisattached to it, . . . . This feeling of thecurrent of outgoing energy has receivedfrom Wundt the name of the feeling ofinnervation. I disbelieve in its existence,and must proceed to criticise the notion ofit, at what I fear may to some provetedious length [p. 493].

If in this statement we replace thephrase "feeling of" by "informationabout," then this old controversy isexactly germane to our present ques-tion. While we do not intend to en-gage in an exhaustive review of theargument about "feeling of innerva-tion," let us look at the principal dataabout which the disagreement cen-tered.

One major piece of evidence at thattime is summarized by Helmholtz(1925). He states:

For instance, if the external rectus of theright eye is paralyzed or the nerve leadingto it, this eye can no longer be pulled aroundto the right. As long as the patient con-tinues to turn it inwards only it still makesregular movements, and he perceives cor-rectly the directions of objects in the fieldof view. But the moment he tries to turn hiseye outwards, that is, to the right, it ceasesto do his bidding, and remains standing inthe middle, while the objects appear to moveto the right, although the adjustment of theeye and the positions of the retinal imagesin it have not varied [p. 245].

From this Helmholtz concludes thatsince there was absolutely no afferentchange when the eye tried to move tothe right, and since motion was per-ceived as if the eye had moved to theright with the retinal image remainingconstant, there must be a feeling of(information about) innervation.

William James (1950) quotes otherdata in addition. He says:

Partial paralysis of the same muscle, paresis,as it has been called, seems to point evenmore conclusively to the same inference, thatthe will to innervate is felt independently of

all its afferent results. I will quote the ac-count given by a recent authority, of theeffects of this accident: "When the nervegoing to an eye muscle, e.g., the externalrectus of one side, falls into a state ofparesis, the first result is that the samevolitional stimulus, which under normal cir-cumstances would have perhaps rotated theeye to its extreme position outwards, now iscompetent to effect only a moderate outwardsrotation, say of 20 degrees. If now, shuttingthe sound eye, the patient looks at an objectsituated just so far outwards from the pareticeye that this latter must turn 20 degrees inorder to see it distinctly, the patient will feelas if he had moved it not only 20 degreestoward the side, but into its extreme lateralposition, . . . . The test proposed by vonGraefe [1878], of localization by the senseof touch, serves to render evident the errorwhich the patient now makes. If we directhim to touch rapidly the object looked at,with the fore-finger of the hand of the sameside, the line through which the finger moveswill not be the line of sight directed 20 de-grees outward, but will approach more nearlyto the extreme possible outward line ofvision [p. 507]."

The theoretical relevance of this ob-servation is stated succinctly by James:

It appears as if here the judgment of direc-tion could only arise from the excessive in-nervation of the rectus when the object islooked at. All the afferent feelings must beidentical with those experienced when theeye is sound and the judgment is correct.The eyeball is rotated just 20 degrees in theone case as in the other, the image falls onthe same part of the retina, the pressures onthe eyeball and the tensions of the skin andconjunctiva are identical. There is only onefeeling that can vary, and lead us to ourmistake. That feeling must be the effortwhich the will makes, moderate in one case,excessive in the other, but in both cases anefferent feeling, pure and simple [p. 508].

James then proceeds to rebut theinterpretations of these observations.Acknowledging that G. E. Muller wasthe first to propose the rebuttal ex-planation, he states:

Beautiful and clear as this reasoning seemsto be, it is based on an incomplete inventoryof the afferent data. The writers have allomitted to consider what is going on in the

Page 3: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

INFORMATION ABOUT SPATIAL LOCATION 375

other eye. This is kept covered during theexperiments, to prevent double images, andother complications. But if its conditionunder these circumstances be examined, itwill be found to present certain changeswhich must result in strong afferent feel-ings. And the taking account of these feel-ings demolishes in an instant all the conclu-sions which the authors from whom I havequoted base upon their supposed absence [p.508].

James then proceeds to point outthat the covered, healthy eye does ro-tate as directed by the efferent im-pulses and thereby provides the affer-ent stimulation necessary for the per-ception of motion in the Helmholtz(1925) example, or the misperceptionof direction in the Graefe (1878) ex-ample. Although James, in his ex-planation, never copes with the ques-tion of why the afferent impulses fromthe covered eye should completelydominate the afferent impulses from theopen eye, nevertheless his "demolition"of the argument for feeling of (infor-mation about) innervation appears tohave been very effective. So per-suasive was the argument by Jamesthat Mach (1914), who in 1886 hadargued strongly for the "feeling of in-nervation" and presented original ex-periments supporting it, almost com-pletely reversed his stand in the fifthedition of his book, written in 1906.Here he says:

The theory of James and Munsterberg fitsthese facts, as I think, without any strain-ing, and we ought therefore to consider itas correct in essentials. The innervation isnot felt, but the consequences of the inner-vation set up new peripheral sensible stimuli,which are connected with the execution ofthe movement [p. 176],

Rightly or wrongly, James appar-ently won the argument, and the issuehas been a dead one in psychology formany years. Many dead issues do notstay dead, however, and this one hasrecently been revived by physiologists.Recently von Hoist (1954), concern-

ing himself with how the organismdifferentiates between self-generatedmovement of a part of the body andan identical movement generated byexternal forces, proposed the idea of"efference copy." His idea was thatincoming afferent signals were matchedagainst a temporary copy of outgoingefferent signals. If they matched per-fectly, the motion involved was en-tirely self-generated. This, of course,is somewhat different from the ideathat information from a record of ef-ferent impulses is available and used allby itself. Nevertheless, it is relatedand served to revive the issue in othercontexts.

The question has become particularlyimportant to those who are concernedwith understanding the control systemfor eye movements. Probably the ma-jor reason for this is that there is greatdoubt among physiologists that afferentsignals from' the extraocular musclesare used to any significant extent inthe control of eye movements. Ifafferent feedback from the extraocularmuscles is not useful for determiningthe position of the eye, then it becomesconvenient far the theoretician to positthe existence of information obtainedfrom a record of efferent impulses.

Thus, Fender (1964), discussing thepossible role of afferent signals of posi-tion of the eye, says:

There is experimental evidence that thepositioned signal is not used, for if a subjectviews two similar but separately generatedstabilized images, one with each eye, it isfound that for a short period the two visualaxes move in1 conjunction. However, thismotion quickly breaks down, and the visualaxes move independently, sometimes gettingas far apart as 30 deg in the horizontaldirection and 15 deg in the vertical. Thereis, of course, no binocular retinal-image dis-parity to act as a cue in this case, and itappears that any positional signal whichmight arise from the extraocular muscles isquite ineffective in maintaining the paral-lelism of the visual axes [p. 317],

Page 4: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

376 LEON FESTINGER AND LANCE KIRKPATRICK CANON

Fender proceeds to incorporate an"efferent copy" feedback loop into hismodel of the physiological system con-trolling eye movements.

Whitteridge (1962) recently sum-marized the problem as follows:

The role of extraocular afferent impulses inperception is very uncertain. It is selfevident that we are not directly aware ofthe position of our eyes in the same sensein which we are aware of the position ofour fingers even with the eyes shut. Thequestion is whether the position of the eyesenters into judgments of position and move-ment, and if it does, how far proprioceptorsare responsible. The alternative theories arethat information from the volume of out-going motor nerve impulses in the oculomotorpathways is centrally available—the outfloivtheory, or that impulses from proprioceptorsdirectly signal the state of the eye muscles—the inflow theory. The strongest pointagainst the inflow theory is that when a pa-tient with a paralyzed and therefore immo-bile eye tries to turn it to one side, theobserved visual field moves as though hehad succeeded in moving the eye. Thiscannot be due to any conceivable change inproprioceptive discharge [p. 511].

As of 1962, among physiologists, theentire controversy seems to have re-vived. The issue is now more sophis-ticated from a theoretical point of view;but on the empirical side, Whitteridge(1962) seems to be back to Helmholtz(1925). There is, however, more em-pirical evidence on the issue todaythan there was 60 to 70 years ago.Brindley and Merton (1960) report avery direct attempt to settle the ques-tion as to whether or not there is usableproprioceptive feedback from the ex-traocular muscles. They anesthetizedthe surface of the eyes and the innersurface of the eyelids of subjects andcovered the corneas with opaque capsso that the subjects received no visualinformation. They then mechanicallymoved a subject's eyeball by catchinghold of the insertion of either themedial or lateral rectus muscle withtoothed forceps. When the eye was

moved in this manner through rota-tions of 20 degrees or more, sometimeseven backward and forward quiterapidly, the subject did not know thathis eye was moving.

Cognizant of the argument offeredby James, they repeated these observa-tions moving both eyes simultaneouslyand obtained the same result. Mer-ton's (1964) paper, the main purposeof which is ". . . to reinstate theexperiments of Helmholtz, whichproved that no information about theposition of the eyes is derived fromsense endings in the eye muscles [p.315]," comes to the conclusion: "Asubject is only conscious of his inten-tion to move his eye and does not knowwhether the movement has in fact takenplace or not [p. 318]."

Considering these new data, it seemshighly likely that Helmholtz (1925)was correct and that James (1950), inspite of having won the argument inhis day, was wrong. It would be use-ful, however, to have additional dataon the question. After all, the work ofBrindley and Merton (1960) demon-strates the absence of a position sensein the eye based solely on propriocep-tion from the extraocular muscles. Tostrengthen the argument one mightwell desire positive evidence that infor-mation obtained from a record of out-going motor nerve impulses is avail-able and useful.

Let us be specific. If it is true, asseems likely, that we know the positionof the eye mainly in terms of knowingwhere the eye was directed to go, thenit should be possible to show that whenthe eye is directed to go to a specificlocation, a subject knows where his eyeis more accurately than if the eye ar-rived at the same position withoutdirections concerning this specific loca-tion ever having been issued.

The technical problem in doing suchan experiment is, of course, the prob-

Page 5: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

INFORMATION ABOUT SPATIAL LOCATION 377

lem of how to devise a method of hav-ing the subject move his eyes to aspecific location without issuing effer-ent signals concerning that location.A plausible solution to this technicalproblem may be found in the work ofRashbass (1961). He reports a seriesof experiments designed to elucidatethe relationship between the usualsaccadic eye movements and the smootheye movements that occur in trackinga target. Several of his findings areimportant to us here.

Rashbass reports evidence that sac-cadic eye movements and smoothtracking eye movements are controlledand generated independently of oneanother. Barbiturate drugs serve toalmost completely disrupt smooth eye-tracking movements but do not inter-fere with precise saccadic movements.Thus, a subject who watched a targetwhich moved horizontally at a rateof 3.5 degrees per second ordinarilyshowed a smooth eye movement beforethe administration of any drug. Afteradministration of a barbiturate, Rash-bass (1961) states:

The first noticeable effect was the increasein the number of saccadic movements occur-ring during the first second of tracking. Asthe amount of drug given increased, thesaccadic movements increased at the expenseof the smooth movements, until, after 8 min-utes, no smooth tracking movement could bedetected [pp. 333-334].

From this and other data, he con-cludes that barbiturate drugs make thesmooth tracking response inoperativebut do not interfere with accurate sac-cadic eye movements. Hence the twotypes of eye movements must be sepa-rately controlled.

Rashbass also reports data from ex-periments designed to discover whatproduces smooth and saccadic eyemovements. The specific question is"whether smooth movements arebrought about by the position of the

target's image on the retina, or by itsmovement over the retina [p. 331]."He tests this "by imparting to an ini-tially stationary target a displacementto one side, and at the same time be-ginning a movement of uniform veloc-ity toward the opposite side [p.331]." The result is stated by Rash-bass as follows:

. . . after a reaction time during whichthe eye does not move, a smooth movementstarts in the direction in which the target ismoving. When this has been established,a saccadic movement occurs in the directionopposite to the smooth movement to counter-act the lead which the eye has over thetarget. . . . This result indicates that thesmooth movement is stimulated by the move-ment of the target irrespective of its position.The conclusion that the smooth movementsare brought about by the movement of thetarget explains the apparently paradoxicalobservation that the first movement whichthe eye makes may take the point of fixa-tion further from the target than if no eyemovement at all were to occur [p. 332].

From this and other data, the con-clusion is that "the smooth movementis stimulated by the direction of move-ment and the velocity of the target,and the saccadic movement is stimu-lated independently by the position ofthe target [p. 333]."

We have dealt at length with the re-sults obtained by Rashbass becausethey are critical for us. They suggestthat if the eye were brought into agiven position by a saccadic movement,this movement would be a response toefferent signals concerning the positionof the target. If, however, the eyewere brought into that same positionby a smooth tracking movement, theefferent directions would be concernedwith velocity matching and not pre-cisely with target location.

A possible experiment suggests itselfto answer the question concerning theavailability of information based onefference. The experiment would beconducted in a completely dark room

Page 6: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

378 LEON FESTINGER AND LANCE KIRKPATRICK CANON

with the subject's head fixed so thatonly eye movements could occur. Inone variation, a light would suddenlyappear within the visual field of thesubject, then disappear, and the sub-ject would be asked to point to the lo-cation where the light had been. Inthis variation, in order to fixate thelight, a saccadic eye movement wouldoccur, and "directions" would havebeen given to the extraocular musclesto move from "normal frontal" posi-tion to a specific location. If thesedirections to the musculature are moni-tored and recorded so as to be avail-able as information, the person wouldknow the location of the light on thebasis of knowing where he had sent hismusculature in order to fixate the light.

In another experimental variation,the light would appear and moveslowly and smoothly across the visualfield before coming to a stop. Thesubject would fixate the light when itfirst appeared and would then trackthe light across the visual field untilit stopped moving. To the extent thatonly smooth tracking eye movementsoccurred, the musculature would, pre-sumably, simply have been directed to"follow the light." Thus, in this ex-perimental variation, the efferent in-formation that existed would containinformation about the direction ofmovement and the velocity of move-ment, but would not include informa-tion concerning the specifically desig-nated position in which the light hadstopped.

In both of the above variations, ofcourse, there would be the sameamount of proprioceptive informationconcerning where the light was. If thesubject's head is clamped so that onlyeye movements are used to fixate thetarget, the only proprioceptive signalswould come from the extraocularmuscles. Since these signals are notuseful, as Merton (1964) has shown,

then subjects would know the locationof the light more accurately when itsuddenly appeared than when theytracked it across the visual field. Wewould, of course, expect more than zeroknowledge of location in the trackingcondition. The subject would haveknowledge about direction and alsosome knowledge of eye position fromafference from the eyelids. Also, it iswell known that smooth tracking move-ments lag and saccadic movements oc-cur periodically. These would alsoprovide additional information. If,however, information based on effer-ence is available, we would expect adifference between the two conditions.

Along with this, of course, one wouldwant to set up another experimentalcondition in which the subject's headwas not clamped so that head move-ments could be employed in helping tofixate the light. Useful proprioceptiveinput would be expected from the neckmuscles, and to the extent that theposition of the light could be adequatelyknown on the basis of these propriocep-tive signals from the neck muscles, thedifference between the two experi-mental variations would be expectedto vanish.

Such an experimental design, usingtwo manners of presentation of thelight and two degrees of adequacy ofproprioceptive information, should pro-vide data that would answer the ques-tion as to whether or not outflow in-formation is available and is used.

PROCEDURE

Twenty-eight college students, 12 femaleand 16 male, were subjects in the experi-ment. Each subject volunteered and waspaid $1.50 for participating.

The experiment was conducted in a light-proof room. The apparatus consisted of anoverhead boom fastened to the ceiling withits pivot point slightly in front of a pointdirectly over the subject's head. The boomextended 4 feet forward from the pivot point.

Page 7: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

INFORMATION ABOUT SPATIAL LOCATION 379

From the far end of the boom hung an il-luminated rectangle that measured 2 X 3inches. The experimenter, standing to theside of the seated subject, could move theboom noiselessly so that the light was at anydesired lateral position. The height of thelight was fixed approximately at the subject'seye level. Calibration at the pivot point ofthe boom enabled the experimenter to readthe setting of the light in angular deviationfrom straight ahead of the subject.

On a table directly in front of the subjectand at a suitable height was a pointer at-tached to a calibrated turntable. The pivotof the pointer was directly underneath thepivot of the boom. The subject, when point-ing to where the light was, or had been, wasinstructed to lay his index finger along thepointer and move it so that he pointed inthe proper direction. The measuring scalesfor both the boom and the pointer were verydimly illuminated and shielded from the sub-ject. The illumination was sufficient, how-ever, to allow the experimenter to read thescales in an otherwise totally dark room.The target light was also dimly illuminatedso that there were no problems with after-images, and the target light did not makeother things in the room visible.

Fourteen of the subjects, seven male andseven female, were used in the "eye-move-ment-only" condition. These subjects hadtheir head in a rigid clamp throughout theexperiment so that fixating and tracking thetarget light could be done only with eyemovement. The head and body were alwaysin the directly forward position. The other14 subjects, 9 male and 5 female, wereused in the "head-movement" condition.This condition was identical to the other ex-cept that the head was not clamped. Thus,these subjects could and did rotate theirheads, and even their bodies to some extent,in addition to moving their eyes in fixatingand tracking the target light.

Before data collection started, each sub-ject was given practice at using the pointerwith the target light at various positions.This practice was continued until the sub-ject was familiar with the situation and theuse of the pointer. The actual data collectionconsisted of 28 trials, 4 trials at each of7 positions of the light. The positionsused were +30, +20, +10, 0, -10, -20, and—30 degrees (+ referring to positions tothe subject's right, —, to positions to thesubject's left). For one trial in each posi-tion the target light was turned on in thatposition and stayed on. The subject pointedto the light while it was still visible. This

was intended to yield a measure of theaccuracy to be expected with optimal infor-mation. For another trial in each of theseven positions the light was turned on inthat position, stayed on for 3 seconds, andwas then turned off. The subject was askedto point, after the light was turned off, towhere the light had been. In this condition,outflow information would presumably beavailable to the subject. When the lightcame on, the subject would have to directa saccadic movement of his eyes to a specificlocation and would, hence, know this loca-tion at least with respect to a normal frontalreference point.

On the two remaining trials at each of thetarget-light positions, the light moved acrosspart of the visual field. The light wouldappear, move slowly (approximately 10 de-grees per second) across the visual fieldthrough an angle of IS, 20, 25, 30, or 35 de-grees, and come to a halt at the desired posi-tion. The light then remained on in thisfinal position for 3 seconds and was thenturned off. The subject was asked to pointto where the light had been after it wasturned off. For each of the seven positionsthe light moved from right to left on onetrial and from left to right on the othertrials. These trials were, of course, in-tended to be trials on which outflow informa-tion concerning target position would be lessavailable to the subject. To the extent thatsmooth tracking eye movements would havebeen involved, directions concerning targetlocation would not have occurred.

The decision to keep the light on its finalposition for 3 seconds before turning it offwas an arbitrary one. We wanted a periodof time long enough so that in the trackingtrials there would be no ambiguity aboutwhen and where the light had come to astop. On the other hand, we wanted theperiod short enough so as to reduce the like-lihood of blinking or moving the eyes to aforward position and refixating the light.Such eye movements would tend to vitiatethe procedure. Certainly, in 3 seconds suchmovements can occur, but some compromisebetween allowing this and having an un-ambiguous final position was necessary.

The order of trials was arranged in asequence so that the target light was neverin the same position on any two consecutivetrials, and the four different kinds of trialswere distributed evenly through the series.The same order was used for all subjects.After the subject had pointed for a trial, hewas asked to return his hand to his lap. Theexperimenter then recorded the setting of

Page 8: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

380 LEON FESTINGER AND LANCE KIRKPATEICK CANON

the pointer and moved the boom to the ap-propriate position for the next trial. Theinterval between trials was approximately 45seconds. The subject's hand remained in hislap until the experimenter said, "All right,now point to where the light is (was)."

RESULTS

We are interested in the magnitudeof the error made by the subject inpointing to the position of the targetlight. The less adequately the personknows the position of his eyes, or hishead, when fixating the light, the lessaccurate should he be in pointing to itslocation afterwards. The simplest cal-culation is, of course, to take the abso-lute deviation of the pointer positionfrom the target position for each trial.Thus, if the target was in position +20and the subject set his pointer to +16,this would be an error of 4 degrees.Table 1 presents the results from theexperiment based on this simple cal-culation.

Even a cursory look at the data inTable 1 reveals that the obtained dataare of the kind one would expect if,indeed, proprioceptive input from theextraocular muscles is poor and theperson has available, and uses, outflowinformation. When only eye move-ments are allowed, that is, when thehead was clamped, the error of point-

TABLE 1

AVERAGE ABSOLUTE ERROR (IN DEGREES)OF POINTING TO TARGET LIGHT

Condition

Eye move-ment only

Head move-ment

Type of trial

Lighton

3.06

2.13

Light off

At posi-tion

3. 54

3.92

Trackedfromright

5.24

3.69

Trackedfromleft

5.55

3.35

ing when the light suddenly appearedat the designated position was onlyslightly worse than when the light wason while the person was pointing.However, when the subject trackedthe light across the visual field, andthus would not have relevant outflowinformation, the error of pointing isconsiderably greater. It is also clearthat when head movements are al-lowed, the results are very different.The "tracking" trials are then slightlysuperior to the "at position" trials.

We have presented these data be-cause some readers might consider thisthe proper measure to use. We willnot engage in extended discussion ofTable 1, however, nor present statisti-cal analyses, since it seems to us thata more accurate measure should beused. The absolute error of pointingis, of course, affected by constant er-rors. One subject may consistentlypoint somewhat to the right of thetarget, another consistently to the left.Such constant errors are probably dueto coordinating the physical act ofpointing with knowledge of locationand probably should be disregarded inour calculations. Actually, there wasan average constant error of pointingsomewhat to the left of the position ofthe target. Over all types of trials,this average constant error was 1.6 de-grees to the left in the "eye-movement-only" condition and .1 degree to theleft in the "head-movement" condition.The probable reason for the directionof the constant error in the "eye-move-ment-only" condition is that, using theright hand, the hand position wasmore comfortable along the fixedpointer when pointing toward the leftthan when pointing toward the right.Apparently, head movements providedenough additional orientation to elimi-nate this constant error.

There is also another source of con-stant error in the data. Two types of

Page 9: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

INFORMATION ABOUT SPATIAL LOCATION 381

tracking trials, one from the left, onefrom the right, were used because ofthe possibility that the memory ofwhere the light had stopped might beaffected by the direction in which thelight had moved. This, indeed, turnsout to be the case. In the trackingtrials, the subjects tend to point a bitmore in the direction from which thelight had come. Thus, in the "eye-movement-only" condition the con-stant error is —1.2 degrees when thelight came from the right, but —2.2degrees when the light moved from theleft. Similarly, in the "head-move-ment" condition the correspondingconstant errors are +1.1 and —.6.The difference between the two typesof tracking trials is not quite signifi-cant statistically for the "eye-move-ment-only" condition (t = 1.44) butis significant at the 2% level forthe "head-movement" condition (t =2.85).

Clearly, we do not want to have ourmeasure of accuracy of pointing con-taminated by these various sources ofconstant error. We, therefore, com-puted a "corrected absolute error" ofpointing by taking into account foreach subject, for each type of trial, theconstant error in the data. Thus, forexample, a subject may have had a

TABLE 2AVERAGE CORRECTED ABSOLUTE ERROR (IN

DEGREES) OF POINTING TO TARGET LIGHT

Condition

Eye move-ment only

Head move-ment

Type of trial

Lighton

2. 58

1.95

Light off

At posi-tion

3.11

3.64

Trackedfromright

4.38

3.11

Trackedfromleft

4.50

2.79

constant error of 2 degrees to the lefton the seven trials on which the targetwas tracked in from the left. If thissubject set his pointer at —24 degreeswhen the target light had actuallystopped at —20 degrees, his correctedabsolute error on this trial was 2 de-grees. Table 2 presents the data usingthis measure.

These corrected data show the sameoverall pattern of results as the datausing the uncorrected absolute error.We will discuss these data in detail,presenting appropriate statistical anal-yses.

Eye-Movement-Only Condition

It is clear that when only eye move-ments are permitted, localization of thetarget light is better when the lightsuddenly appears at its final positionthan when it is tracked to its finalposition. An analysis of varianceyields a highly significant F value(8.71, d/ = 3/39) for the varianceamong the means of the different typesof trials. The variance among sub-jects is also significant (F — 2.68, dj= 13/39). This latter, of course,simply means that some subjects areconsistently more accurate than othersin pointing to the target light.

The difference in accuracy betweenthe "light-on" and "light-off-at-posi-tion" trials is not significant (t =1.10). The mean for each is, how-ever, significantly different from themean for each of the "tracking" trials,the smallest t value being 3.86 betweenthe "at-position" mean and the"tracked-from-right" mean. In short,with only eye movements permitted,pointing to the target when it sud-denly appeared at its final position isnot materially less accurate than whenthe pointing was done while the lightwas still on. In the tracking condi-tions, however, when relevant outflowinformation was presumably not avail-

Page 10: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

382 LEON FESTINGER AND LANCE KIRKPATRICK CANON

able, accuracy is materially and sig-nificantly worse.

Head-Movement Condition

When head movements are allowed,the data present quite a different pat-tern, although significant differencesstill exist among the different types oftrials. The variance of the means forthe different types of trials and of themeans for subjects both yield highlysignificant P values (7.30, dj = 3/39;and 6.56, dj = 13/39).

With head movements, the accuracyof pointing with the light still on issignificantly better than each of thethree conditions in which the pointingwas done after the light was off. Theimportant differences to us, however,are between the "at-position" trialsand the "tracking" trials. Here wefind that the "at-position" accuracy isno longer better, but is actually worsethan the accuracy of pointing on the"tracking" trials. The two t valuesare 2.01 and 2.19 which, for dj = 13,are each significant at about the 5%level. We had not anticipated this,and we are not certain of the reasonfor it. It may simply be that occa-sional inattention affected accuracy inthe "at-position" trials. There was nowarning of when the light would ap-pear. In the "tracking" trials, the pe-riod of tracking could minimize theeffects of any inattention. It is clear,however, that when head movementsare allowed, thus making availablegood proprioceptive input concerningposition, the availability of relevantoutflow information no longer producesgreater accuracy.

Comparison of the Two Conditions

If we compare the accuracy betweenthe condition in which only eye move-ments were allowed and the conditionin which head movements were also

allowed, we see that in the latter con-dition there is a general tendency tobe more accurate. When the light ison while pointing, the average cor-rected error decreases from 2.58 to1.95, a difference significant at the10% level 0=1.73, rff = 26). Thedata for the "tracking" trials also showmuch less error with head movementallowed. The two t values here are2.11 for "tracking from the right" and2.98 for "tracking from the left," sig-nificant at the 5% and \% level re-spectively.

Only for the "light-off-at-position"trials is there no improvement fromthe "eye-movement-only" to the "head-movement" condition. The actual dif-ference is slightly in the opposite di-rection but is negligible (t = .81). In-deed, it seems as though the presenceof relevant outflow information abouteye position in the "eye-movement-only "-condition is just as good as thepresence of the same outflow informa-tion plus good proprioceptive input inthe "head-movement" condition. It isclear also that, when there is good pro-prioceptive input and no relevant out-flow information, as in the trackingtrials with head movements, accuracyis at least as good as when relevantoutflow information is also present.This would tend to imply that, in thissituation, there is some redundancy ofinformation.

DISCUSSION

The main conclusion we would liketo draw from the results of the ex-periment is that information based onsome kind of record of efferent im-pulses (i.e., outflow information) isavailable to the organism. The majorresult on which we wish to base thisconclusion is the finding that, whenonly eye movements were permitted,target localization was more accuratewhen the target suddenly appeared at

Page 11: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

INFORMATION ABOUT SPATIAL LOCATION 383

its final position than when it wastracked to that position.

Let us review the line of reasoninginvolved in coming to this conclusion.

Accuracy of localization of an ob-ject in space with respect to one's bodydepends on knowledge of body, head,and eyeball position. If the head andbody are fixed, the only variable isposition of the eyeball.

There is evidence that the positionof the eyeball is not adequately knownon the basis of proprioceptive signalsfrom the extraocular muscles. Hence,with head and body in a fixed position,accuracy of localizing an object inspace would be poor if the only infor-mation about eyeball position camefrom such proprioceptive signals.

There is evidence that smooth track-ing movements of the eye are con-trolled and directed by the directionand velocity of movement across theretina and not by target location. Sac-cadic eye movements, on the otherhand, are directed on the basis oftarget location on the retina. Hence,if a target is fixated by means of asaccadic movement, efferent signalsrelevant to target location would havebeen issued. If a target is tracked bya smooth eye movement, however, ef-ferent signals concerning direction andvelocity of movement would havebeen issued—information not optimallyuseful for knowing the target location.

Consequently, if a record of efferentsignals is available, localization inspace of a target should be better fol-lowing fixation by a saccadic eye move-ment than following a smooth trackingeye movement. Having found this re-sult, we regard it as evidence for theexistence of information based on thishypothesized record of efferent signals.

It is, of course, possible that thereare alternative interpretations of thedata we have presented. No suchplausible alternatives occur to us, how-

ever. It does not, for example, seempossible to maintain any alternative in-terpretations in terms of confusion in-troduced by the tracking procedure,since it is clear, in the "head-movementcondition," that the tracking procedure,in and of itself, does not interfere withaccuracy.

Another possible alternative expla-nation could be elaborated as follows.Presumably, during the period ofdarkness between trials, the subject'seyes revert to some "normal" frontalposition. Such a normal position isprobably a reference point for locationin the visual field, and, presumably,directions are issued to the extraocu-lar muscles with respect to some suchreference point. The eyeball thenmoves, in accordance with the effer-ent directions, in a saccadic, ballisticmovement. Under such circumstancesthe initial movement of the eye tofixate the target is not a continuouslycontrolled movement. Once started itproceeds to its destination. The sac-cadic movement, hence, must have acomplete set of directions issued at thebeginning.

It thus becomes clear that, in orderto issue directions that are relativelyaccurate for the initial ballistic move-ment of the eye, information as to thelocation in space of the target mustexist before the directions are issued.And indeed, this information must beobtained on the basis of the stimula-tion of the periphery of the retinawhen, with the eyes in frontal position,the target light suddenly appears. Itis on the basis of this information thatthe initial ballistic eye movement ismore or less accurately directed.

Why, then, is it necessary to saythat the differences obtained betweenthe "eye-movement-only" conditionsare due to a record of the efferent im-pulses actually sent out to the muscles ?Why could we not simply maintain

Page 12: Festinger (1965) Information about spatial location …wexler.free.fr/library/files/festinger (1965) information...I know where that object is in relation to my body because, on the

384 LEON FESTINGEE AND LANCE KIRKPATRICK CANON

that the information the person has asto the location of the target light issimply the information on the basis ofwhich the efferent directions wereissued? After all, on the trackingtrials the subject did not see the targetlight at its final position in peripheralvision. The results of the "head-movement" conditions rule out thisexplanation of the results. If seeingthe target light in peripheral visionwere important, the tracking condi-tions should still be inferior even withhead movements allowed.

One must admit, however, that in-formation based on a record of the ef-ferent signals is not likely to be betterthan the information on the basis ofwhich those efferent signals were sent.Our present data cannot answer ques-tions concerning the relation betweenthese two things. Our experimentdoes, however, confirm the existence,and usefulness, of outflow information.

REFERENCES

BRINDLEY, G. S., & MERTON, P, A. The ab-sence of position sense in the human eye.Journal of Physiology, 1960, 153, 127-130.

FENDER, D. H. The eye-movement controlsystem: Evolution of a model. In R. F.Reiss (Ed.), Neural theory and modeling.Stanford: Stanford Univer. Press, 1964.Pp. 306-324.

GRAEFE, A. VON. Handbtich der gesammtenAugenheilkunde, 1878, 6, 18-21.

HELMHOLTZ, H. VON. Treatise on physio-logical optics. (3rd ed.) (Ed. & trans,by P. C. Southall) Vol. 3. Menasha,Wis.: Optical Society of America, 192S.

HOLST, E. VON. Relations between the cen-tral nervous system and the peripheralorgans. British Journal of Animal Be-havior, 1954, 2, 89-94.

JAMES, W. Principles of psychology. Vol.2. New York: Dover, 19SO.

MACH, E, The analysis of sensations. Chi-cago: Open Court, 1914.

MERTON, P. A. Absence of conscious posi-tion sense in the human eyes. In M. B.Bender (Ed.), The oculomotor system.New York: Harper & Row, 1964. Pp.314-320.

RASHBASS, C. The relationship betweensaccadic and smooth tracking eye move-ments. Journal of Physiology, 1961, 159,326-338.

WHITTERIDGE, D. Afferent mechanisms inthe initiation and control of eye move-ment. In, Proceedings of the InternationalUnion of Physiological Science: XXIIInternational Congress. Amsterdam: Ex-cerpta Medica Foundation, 1962. Pp. 509-512.

(Received June 10, 1964)