UNCLASSIFIED

AD 408 245 LDEFENSE DOCUMENTATION CENTER

FOR

SCIENTIFIC AND TECHNICAL INFORMATION

CAMERON STATION. ALEXANDRIA. VIRGINIA

UNCLASSIFIED

'U .•= l l Ii

NMfC3: -n @ovezunt or other dravingsp inpeci.ticatioL" or other data are used. flor any pulpbogeother than in connection vith a deftnitely relaited.goverinnt procur 1-t operation,, the U. S.Oovevswmnt thereby incurs no responsibility., nor anyobligstion what soevrj and. the fact that the Govern-ment my have tozougated. furnished. or in aOW vymipled. the oadd drwAvnpp specifications,, or otherdata Is not to be reprded. by Implication or other-vise as In any mexcer Uicensiag the holder or anyother person or corporations, or conveyizig azy ri~oor pexzmissiom to numafutues, use or sell uWpobalesed. Isvention that mm In my vW be relted

JPRS: 199100 OTSt 3418

7 )ky 1963

IL

C..)A. CWUARATIVE ANALYSIS CF PSYCHIC CONITROL

ADREUUT1014 -1t AUTOWiAVA

by L. M. Bakker

-USSR'

L' NPMTMIN F COMMWud JUL 9 196

-01*4~Pre. 9.50M~fA W~

IWWI

FORE WORD

This publication was prepared under contract for theJoint Publications Research Service as a translationor foreignalanguage research service to the various!federal government departments.

The contents of this material In no way represent the'policies, views or attitudes of the U. S. Goverue-nt6t of the parties to fty'distribution arradgement.

PROCUREMENT OF JPRS REPORTS

All JPRS reports may be ordered from the Office of TechnicalServices. Reports published prior to I February 1963 can be provided,for the most part, only In photociopy (merox). Those published after1 February 1963 will be provided in printed form.

Details on special subscription arrangements for JPRS socialscience reports will be provided upon request.

No curmulativa subject index or catalog of all JPRS reportshas been couqtiled.

All JPRS reports are listed in the Motl aao fU. S.Government Publications, available on subscription at ý4.50 per year?iire orign).including an annual index, from the SuperintendentOf Documents, U. So Government Printing Office, Washington 25, Do C,

All JPRS scientific and technical reports are cataloged andsaabJect..indexed in Technical Translations, published semimonthly bythe Office of Technical Services, and also -available on-subhcripties(112.00 per year domestic, $16.00 foreign) from the SuperintendentOf DOeimentse Senianmaalindexes to Technical Translations areamailable at additional cost.

JPRSI * 19.100•

A COMPAUMVB ANALXS S, Or PSYCHIC CONTMAND *BOVMTIbK IN 'AVTOM&A-T

- USR -

Following is the translation of an article by L. M.Bekker in the Russian-languag publication Ywpgaz

fllu5= (Problem of, Philosophy), No 2, l963•-pages

The idea of the uniqueness of the psychic is one of thegnoseological foundations of idealism. For the noncomparabilityof a phenomenon with any other reality leads to the idea of itssubstan-tiality. The proposition of the substantiality of thepsychic is inevitably related also with the recognition of intro-spection as the only method of psychological investigation.' Covertforms of these tendencies even now are still persisting in thematerialistic science on the' psyche. The approach realised throughmodern cybernetics to the nature of reflection by its very natureprecludes the idea of the nancomparability of the psychic and leadsto a conclusive surmounting of the traditional errors impeding sci-entific progress. The cybernetic approach to the problem of reflec-tion unfolds new constructive means for its examination mainly byway of the following two interrelated aspects.

I. The general-cybernetic teaching on the production, trans-mission, storage, and utilisation of information, revealing thequantitative e 7id qualitative, the structural characteristics asinformational processes in their various forms and levels, yieldsa system of concepts which can be used in analyzing psychic pro-cesses (in their informational aspects) as a special form of

informational phenomena. In so doing we in fact eliminate thedualistic position of the absolute uniqueness of the psychic,uncover a constructive scientific approach for its examination,revealing the specific character of the particular through thegeneral, and giving concrete expression to the most importantposition of V. I. Lenin on reflection as a general property ofmatter, in relation to which psychic reflection is its higher,though particular form.

II. The general principles of the functioning of controlsystems revealed by cybernetics form the basis for a comparativeanalysis of the 'reAlatiua function of signals of information at

-1- ,

various levels. In its turn$ analysis of how reflection "works"and what are the differences in its working function at its variouslevels affords a fruitful investigation of methods of organizingreflection processes. This approach is essentially important forstill another reason, that it equips investigation with the strictlyobjective, and in addition with a wholly adequate method of study-ing psychic reflection. This leads to the actual eliminationof theremnants of the introspectionist apptk-.i'toward methods of inves-tigating the psychic.

The reflex theory of the psychic developed by I. M. Sechenovrepresented the higher level of voluntary regulation of action as aspecial case of psychic regulation, and psychic regulation of actionitself '-- as a special manifestation of the general..,reflex principleof performance of the nervous system. The general t#I*ory of controlcreated by cybernetics, realizing the next stage Qf. scientific gen-eralization was synthesis, represented this reflex: regulation offunction as a special case of general principle of performance ofcontrol system, in which information circulates along a closedcircuit. The scheme., of the reflex circuit lying at the base ofthe mechaniam of behavior regulation is one of the cases of acontrol circuit working on the principle of feedback.

Discovering a commonness between the structural system andthe functional principle of information circulating in an: obj*cti-val adaptive or transforming (by labor) effort of man also inoperations of automatic systems reveals the real possibility ofmaking a concrete comparative analysis of these systems. The goalof such an analysis is to establish the specific character 'of theprogramming and regulating function of.psychic processes as infor-mation signals in comparison with the regulating function of-infor-mation signals controlling the dynamics of working operations inautomatic systems.

The task of revealing this specific character of psychicregulations of objectival work actions within the limits of anelementary geno rZ level of. regulating systems (sensations, per-ceptions, and concepts) is the subject of the present article.But it is relevant to this starting level to deal here only withthat area of the questions, which is related to the qualitative,structural features of sensory signals and to the role of thesefeatures in the control process.

The principle of programming and the method of regulationrealized in objectival work actions of man finds its ultimateexpression in the dynamics of the executive orgaZrs-acting on theobject being transformed. The organization of the functions ofthe working organs that carry out the program of action servesan the most adequate objective indicator of the psychic processesprogramming and regulating the action. Therefore, the characteristic

of control principles over the objectival action is associated withthe specific function of the executive organs of man, chiefly thehuman hand as an oksan of wqrk.., Vhe main feature of the human handas an OxecutiVl orgas oomiawed to .the executive organs of any auto-matic device (including A1*s a aechalial hand) consists in its highfunctional versatility (teltoe toW the number of degrees of freedom),which enables practically sU iaUat±ted class of motor tasks to besolved with the same morphologically unaltered structure# The moetfanciful special dopiction of a. motion. adequateiy repioducing thestructure of the objest being acted upon can be realised by thUwOrking points of the hand as an executive organ. The versatilityand variability is kinematically limited here only by limits ofreach (extent of the motor field), and dynamcilly__-- by the "resis-tance of materials".. Providing the freedom of accomplishing anunlimted diversity of motor solutions, the very design of the handis not restricted to an-'pelcific objectival program of moveuents.The structure of the hand has neither a program nor individualexecutive operations built in., still less any algorithmic *6quenceof such operations.

Serving as functions of reflex effects carrying out actionare efferent-nerve codes of command information, activating theexecutive organ and controlling its pe'formance. However, as hasbeen shown bys studies of the structure of movements (N. A. Bern-shteyn), effector codes cannot serve as the original program car-riers for the action realized by the working organs. This propo-sition is vividly substantiated by the factual material of studiesof motor habits, indicating that the development of habits doesnot occur through formation of a fixed sequence of movements, asit needs must occur if the program carrier of the action. was theeffector command information.

We must, therefore, conclude that the programming of objec-tival work acticns of man and the specific character of their reg-ulation must be uought for In the structure of the guiding Zoovedom-itel'naya7 inf-: -mation inherent in the original regulating function.

The obj .tival structure of the executive acts of man derivedfrom the objectival structure of the psychic processes, which arethe signal of the guiding information regulating the action. Itis precisely this function of sensory imaging that I. M. Sechenovpointed to when he wrote that "sensation is everywhere a regulatorof motion."

The process of technical progress and mechanization of laborhas interrupted the reflex circuit of elementary objectival actionand has interposed between the executive organs of man and thefinal object of labor activity an, ever-increasing intermediatelinks. Technological, power, and transport functions of humanactivity have been wholly converted to machine links of thecontrol circuit. Left for the actions of man's working organsis the function of control. * an has become the controlling linkof the system. Controlling action is increasingl being redused,

-3-

and in relation to the ultimate object of working activity, is"Increasingly losing its developed objeotival,.struature, beingturned into a signal-code of command information C(uotor reactionsof the operator at control panels)h With the transition fromdirect control by machines to Temote..cfontroli; the* objeotivai

... character of this control and the guiding ofo ton s bein...:Aiplaced and lost,, the guiding .info'atiio: eihg received by

.the operator in coded form as -various -.iids :bf indicatIozw .(instru-ment dials, sign indication6, diagrams oiop crts)'t KoWever, losingits directly objectival 6haractern ra n h t .he.:,u1t.tateWoo of labor, ma's perception and attion who-Iy" r-tfain its

obJectival, structure in relation to their. direct objects •!- theindicating equipment. and the control organ. This obh~ctivalnessof.the structure of .'perception and action in-relation..to theirdirect obSects lies'at the basis of the versatility,.genera4,Aa-bility, and relative universality of the controlling operations of

"man,o which 'are a particular came of the gen.ral variability anduniversality of human objectival.action.

Thus a contradiction has arisen between the code ;.characterof guiding and command information in relation to the tobject of action" and the objectival structure of perception andaction in relation'to their direct objects. This contradiction

. lies at the basis of two opposing trenis in the development ofmodern systems of remote control over production proce*ses. Thefirst trend consists'in the guiding information in co4trol systemsbecoming si•'ilsa to the directly objectival forms (three-d:mensionaldiagrams, tele.- I.and radio-locating representations, etc.) and theimparting of the highest 'possible structural commonness of the sen-sory and motor components of action. This trend of making thecontrolling operation similar tQ the functional scheme of objec-tival action derives from man's requirements as a link in .toecontrol system and is related to the need for preserving thesefunctions for h:L-mi.. The second trend in the development of remotecontrol systems -,s manifest in the effort, to replace the structureof guiding and ..oimand information and operation by, signal'8. ofcoded. informaL .•n based on specific algorithm. Car.rying out thistrend will load to the automation of man's controlling operations,that is, to thp development of systems of automatic control.

This line of development aimed at displacing the guidingand command.information find its high point in the principles ofautomatic system functioning. From this boint of view it is nota chance matter that in technical systems information transmittedin the form of..oodes is decoded only in those comnunication chau-nels at the exit of which the information is reoejms at a sensoryaccess point of man (teievisor, a radio-ldcater). At the exit ofsuch channels the coded information# being decoded, is convertedinto an image..- In systems of automatic control, however., guiding"Information anAergoes only. recoding aid'is nowhere converted intoimages. Ohis Ii the essential feature Of such systems.

-4-

Another fundamentally important characteristic of controlprinciples in automatic tsytems: is that, in spite of the tendencyto be free from the fixed 4a&rAoter of the program in the designand to thus make for the iA1V.i's"1e at? the system, the programof system performance eti ow ie ns. as it. were partially fixedin the deign#, In particular this refer '*to. the program of elemen-tary operations built into the design of the syste'"s working organs.In various control systems there is a different didtribution of pro-gram components between the system design aJd the information-.enter-ing the system. But even in those program parts, 'whi•h are givenby information, the elemental relationship of each value of the

* guiding information signal with the states of the controllingexecutive organ corresponding to it is determined by the design(for example, in automatic control systems by the flash of zenithartillery).

From this point of view it is very indicative that the useof universal computers as components of automatic control systemsis complicated by the fact that they require appropriate equipmentfor intake'and outflow of information, but the design of the latterdepends essentially on the specific character of the automaticsystem. It is also important in this regard to note that in thesingle type of automatic systems with unfixed design (which thusdoes not feature a set program), in tVZ so-called indeterminateself-organizing 3ystems, the control process is accomplished onthe basis of random search or on a "trial and error" method. Thetotal absence of a fixed specific program in the design thus con-demns the system to the wo.v of blind search (finding ways ofoptimizating, which is presently the object of numerous inves-tigations).

The indicated features of principles of control realizedin automatic systems on reasonable grounds considerably restretthe versatility, the interference-proof charapter, and the reli-ability of func 'joning of theses systems.

The goal of a comparative investigation of the main charac-teristics of the sensory regulation of human actions and of astudy of control methods in automatic systems consists in discover-ing the differences between these two categories on the basis ofthose general Principles of control, whose specific particularmanifestation constitutes both kinds of control under comparison.

Analysis allows us to discover an essential difference inthe interrelationship of information and design in the process ofcarrying out a program of operations by man and by an automaticsystem. As was indicated, In its very design the automatic sys-tern has partially built in the program for carrying out itsoperations. In the first place, this refers to the system inherentin the working organs, the very structure of which Is built into

the program. of the corresponding executive operations; The designentails in "itself fixation and other program componpnts in variousautOnatic system, as occurs for e•apjl •n systems usi.ng analoguecomputers, whose design entails in its'prt6graa the transformation ofinformation incoming to the system.

In man the-design of his workin -organs and all the otherlinks of the reflex mechanism of objectivalaction does not intro-duo any in4ividual specific program. The role of design consistshire in that it sets up the morphological foundation for possibil-ities of aarryingout any objectival kinematic structure of action.

.A program of actions or. operations ultimately is determinedonly by two factors -- the design of the system or the. character-istics of the info;mation: circulating through its circuit. There-fore, the differing roles of design in programming operations inboth systems can be related only to the different characteristicsof information in those systems. Quantitative characteristics ofinformational processes occurring in both systems in" themselvescannot serve as the basis for the differences in the role thatdesign plays which are of interest to us. In regard to the passingcapacity and rate of operation performance the advantages #re clearlyheld by automatic machines. More improved controlling functions ofinformational processes not reqcuiring that programming be based onsystem design would scarcely derive from a restricted passing capac-ity and speed of operation that man displays compared to the auto-matic machine. Based on' this, the existence of the fundamentaldifferences referred to between the system. under comparison indistributing the program between information and design inevit-ably leads to the ©okclusion' that the differences in the rolethat design plays in realizing the program can be found only inthe substantial difference In qualitative structural character-istics of the information signals ciroulating in the system andinherent in the specific program itself.-

Analysic of the functioning of all the main kinds of con-trol system. supports the conclusion that in systems of rigidcontrol and in •egulating systems, as also true in self-adjustingautomatic systm., the information signals inherent to given pro-gram components exist in the form of continuous or discrete codes.

This conclusion 'derives directly from the characteristic ofthose transformations of information which occur in sensory coel-'pickups /element-datchi&7, in central amplifying convertor linksof the regulator, and in executive equipment. From the veryessence of the physical characteristics of these processes wefind that coding of the incomin information occurs here intothe alphabet of the communication channel, which is a member ofthe control circuit; Ciroulating along the circuit, the infor-mation undergoes several recodings, leaving it, however, in acoded form.

In the objeotival actions of man the ultimate stage ofthe conversion of command Information s represented by the

-6-

systems of iovemonts carkied "o0 by the "working point" of the execu-tive organs which draws hoar the object and acts on it. From theirvery nature these motor staUiiake are similar to the object and tothe ultimate product of actionn iAd in this sense they are not acode but a motor reproduction of the objectls structure. Such anobjectival character of the motor structures of human action can-not be determined, as has been mentioned, either by the design ofthe executive organ, for they have no concrete program inherent tothem, or by the codes of the executive effector information, forthese codes are not fixed in the establishment of habits with. action.The source of this objectival:character of the motor structure here

* can only be the guiding information. In order that signals of the lat-ter can give and adequately receive objectival programs of action, thesignals themselves must inevitably be the representation of objectsof these actions asnth~y occur and in the final state. No othersource and carrier of the objectival program of action, nor itsadequacy to the object has analysis here been able to discover.

The presence of the necessary conjunction of the code formof information with the partial fixation of the program in thesystem designed, and in particular In the structure of its workingorgans, indicates that in and of itself information circulating inthe, form of codea cannot wholly supply the geometrically objectival,and kinematic characteristics of the working operations. Therefore,the part of the program which cannot be determined by informationmust be fixed by the design. In distinction to this informationsignals having the form of objectival images program and regulatethe entire structural kinematic characteristics of human actionwithout the support of a prouraen built into the design. The designfoundation here consists in revealing the biomechanical and psyco-physiological possibilities for realizing any program of objectivalworking and controlling operations. Such essential differences-.Inthe regular defuanctions of these two forms of signals are obviouslyrelated to the -1ndamental differences between representation andcode in the very form of their organizsation.

Every ,-L..gnal of information is a set- of states or elements,ordered appropriately to the set of elements of the informationsource. It is precisely this preservation of the correspondingorderedness of its states that makes the signal the carrier ofinformation of its source. But the distorting effect of any noisoor interference consists precisely in disarranging the correspondingorderedness of the signal elements.

Based on the nature of the information signals as a def-.-nitely ordered set of elements, there is reason to believe thatthe differences between code and representation as forms of infor-mation signals must lie along the line of the properties of theoriginal elements and the character of how these elements woreordered in the set. From this point of view the essence of thecode form of Information sigpals is clear enough. Zere the onlyrequirement for the set of signal states is an Iso0awphio

-7-

correspRndentce to the set of source elements. The essence of iso-

morphism consists, ts is known, in the pafrwise single-valueddorrespondence of the ;elemonts of one set to the 'elements ofanother set and the pVAz.rwse single-v, a2ued correspondence of thetransformation operations determining the conversion from one ele-ment to another inboth sets. As to the physical properties andspatio-teaporal characteristics of- the elementsuof both note iso-morphism in generil is-indifferent. But as to the method underwhich the elementii.of both sets are' ordered isomorphiam presentsno other requirements b6sides that° of the single-valued corre-spondence referred to. Under the' conditions that this correspond-once is preserved the internal strVc'ture of the set, the spatio-temporal organization of its elements,, can be arbitrary,, sines itschanges do not affect the essence of the isomorphic relationships.Since it .is specifically this single-valued correspondence of the.elements of twos.,ets and their transformations that make .the setof e9imments of the information signal a'code of its source, the con-dition of isomorphism is a necessary and sufficient determinant ofthe code nature of the information signals.

In regard to the information 'signals organized in the formof images, the situation is substantially otherwise. The isomor-phism requirement is unconditionally n,>.essary also in relation tothe image, for it is general for any'ki.'d of in-formation signal.Without.the isomorphic relationship of the sets there is no infor-mation signal. But, though necessary:, this condition is not, heresufficient. In order that the information signal be not only acode, but also an image, its characteristics must include additionalrequirements. Analysis showb (cf L. M. Vekker and B. F. Lomov, "TheSensory Inage as Representation," Voprosy filosofi4, No 4, 19,51)that such additional requirements alto affecting 'the properties ofthe elements of the set constituting the signal, and.methods ofordering these' -cements reduce *to the following. In orderthat the information signal be not only 'a code of its couce, but animage of any of its properties, it le first of all necessary that.the properties '.-f the 'elements of the set forming the signal bereproduced by tkhe corresponding physical charactoristice of theelements of the net comprising the source, and in the second place,that the spatio-temporal orderedness of the elements'of-.the setforming the signal not. only, isomorphously correspond to the order..edness of the set of elements constituting the source but alsoadequately structurally reproduce this orderedness, an occurs, fcrexample, in photographic or television images. Observance of boththese additional requirements also makes the information sigual arepresentation of the object.

Since the representation satisfies the conditions of iso-morphism, it still remains.a 'code, but here the commonness sharedby the requirements additionally, stipulated for the signal whichis a. representation gives rise to substantial differences betweenthe representation and the code. Thui, ismovorphism, a necessary

-8-

and sufficient goneral oond4iibb tOi the code form of the signal,does not stipa•late any roq4ioienta in regard: to the dimension-ality (number of measuir *Rs). ot.the n.ot of elements forming thesignal, or of the set .sof eoemont. ýfo4ig the information source.For example, a three-dimensional volumetrically spat1:a1s:kset..6felements forming a physical. object of specific size and form canbe coded-in a two-dimensional or one-dimensional set of elementsvarying in their features. And change in the dimensionality ofthe signal-code compared to .the source can never-distur~b the

character of the code relationships. The situation changes funda-mentally 'for the signal-representation. Change in the number ofmeasurements of a set forming the signal compared with the set-source attacks the nature of the adequacy of the representation.Thus, if the television exposure, being a two-dimensional .st,

is drawn into a line, then it clearly loses its property of imgejs,and, therefore,. remains a code. Analogously, the code relationsi psof two sets freely permit.conversion from the spatial orderedness ofthe elements to a temporal, and vice versa.

Thus, in distinction to the code form of information signalsthe makeup of a signal-image requires the preservation of invariancyof the general orderedness of the elements in the conversion from thesot-source of information to the set forming its signal. But theinvariancy of the total structure or orderedness of the elementsis related to the preservation of the invariancy of the number ofmeasurements of the correlated sets Z-eoo Note7. superimposingthe additional restrictions indicated on the orderednoss of theimage elements defines another method of organizing the signal,under which the latter contains reproduced not only a measure oforderedness (that is, a quantity of information determined by theisomorphism of the signal and the source), but also its concreteform.

(,NOTJ7 At this point,- further strict analysis of the spe-cific character .)f the image as an information signal compared tothe code will fthad to the formulation of a mathematical problemof finding such-. classes of transformation, in relation to whichthe adequate iaternal ordoredness of the set elements and thedimensionality corresponding to it (number of measurements) wouldremain invariant.)

All these indicated differences between the signal-imageand the signal-code "operate" in the process of carrying out therealizing function of the signal, that is, in the transformationof guiding information Into command.

The requirement of isomorphism extends not only to therelationship of the information signal to its source, but alsoto the relationship of the guiding information to the command.The condition of isonorphism applicable to the code form ofinformation signals is ase uarui. and Sufficient both in regardto their guiding as well a& te their operating, regulating func-'tion. The Isomorphic relationships betwees the guiding and the

-. 9-

command information determine the 4~ement-by-element, paiAwisee cor-respondence of the values of both aignals, But they neither setnor determine the principles of li.tua1 !. orderedze'ss Of' ilement ain the sets forming the signals of riaidiig abd command .nformation.Therefore, the, principle of the orderedness of the signal elemsnts,which signal would determine the objectival structure of the execu-tive reactions, the adequate object Df action, cannot be set by thecode form of the signal. Thise stems from the very nature of the.code form of information, which is indifferent to the internalorderedness of the elements of toe correlated sets and requiresonly their single-valued correspondence. -Since the code form ofinformation signal does not consist in. the integrally objectival,structural characteristicsQof the information source, being at thesame. tize the -object of action, it yields no sort' of spatio-geomet-rio. and spatio-dynami' foundations for programing routes ofmotion of the working organs. Not containing a single-valued'spatio-geometric structure, the code cannot of.i'tself providethe kinematics of motion of the executive organs. It is preciselybecause of this that the realization of the program of' operationsin automatic systems, since they operate with the code form ofinformation signals, iusthbe based on geometrical characteristicsof the program being built into the de1.-gn of the system. In thisway the program components are set in vhe.design of'the executive,working organs of the system and in acuordance with the design ofthe executive elements Of the automatic regulators. As to theregulating function of the i-ageo.sigalsa .the situation turns outsubstantially different.

The nature of the reliationship between command informationand guiding derives from the relationship of guiding informationto its source. Just as for the case of the relationships of image-signals to the; information source, for the relationships of. commandinformation to guiding information existing in the form 6f'imageathe conditibn Oi isomorphism of the sets of elements is a necessary,but an insufficient determinant of the mutual relationships. As wasshown, for the im.aage in'.comparison to the code there is not only apairwine ~corr.er-jondence 'between the elements of the signal 'and thoseof the source, 'ut also a commonness of the principle of their spatio-temporal orderedness. 'Correspondingly, in distinction to the code-signals between command and guiding information we require here notsimply an' element-by-element correspondence, indifferent to theorderedness of the elements within each of the sets, but a com-monness. of the principles of their orderedness. Deserving theproperties of the elements and the adequate source of information,their orderedness, the iiage reproduces the physical and the spatialstructure of the objects. And exhibiting the dynamic and geometricobjectival characteristics, the image can set the spatial character-istiCs of the routes of motion realizing the operation with thisobject. The objeotival-at•rctural dymamic and geometric charac-teristics of the im" determine its kinematics of operations

- 10 -

directed toward the object. Here, the geometry of how the object isrepresented against the background of a specific spatial situationand the kinematics of how the object is dealt with are correlatednot only element-by-eloient, but also ptructurally.

In carrying out the regulating function through code signalsof information, the command and the paiding information are corre-lated to each other as two sets, be i coies of the informationsource -- the object, of action, .

Under an optimal realisatio4••, the1 regula.ting function ofthe image-Aigals, the guiding inforlbtion and the ultimate formof the transformations of the command information -- ,the system ofmovements -- are correlated to each other as two sets,, whose ordet-edness reproduces the structure Of the object of action. And this

means that not only each element of the projectory of movement canbe single-valuedly compared with the corresponding element of theimage, but also that the motor structure as a wholereproduces theorderedness of the image elements, and through the agency.of thelatter, the structure of the object of action as well. The geometryand t of the image of the object and the conditions of actioncontain potentially the routes of the possible movements. realisingthe operation on the given object. Thus, for example, a psychicimage of a perceived or represented ob2ect contains in itself theroutes of movements for reproducing this object (in a drawing, inmarble, etc.). In more complex instances and in the general casethe kinematics of movement derive not only from the geometric otopological, but also from the power components of the image. Butthe power components of'the image, even though they do not coincidewith the elementary geometrical logic of the object (they "do notlead there", from the expression of N. A. Bernshteyn), are organizedin *patio-objectival structures determining the kinematics of thecorresponding motor operations.

This is expressly why, in distinction to code-signals thatinformation-siC&TAs organised in the form of images cnn programand regulate tkt. motor functions of the executive organs withoutdepending on s't.icifically geometric components of the program beingbuilt into the design of the system.

From this derives the principle of the correlation of infor-mation with the design of the system, the principle consisting inthe fact that the design of the controlling system of psychic reg-ulation, not being fixed by any definite program of specificallyexecutive or controlling operation, unfolds within definite limitsoptimal possibilities for realising any objectival program of trans-port, technological, or controlling operations. This is alsorelated to the absence of any transcriptiQn of a program of• xec-utive and controlling operations in the design of the executiveorgans of man, as well as being related to the universality ofhis executive and controlling functions.

The problem consists in how, from the above characterizedprinciple of programing and regulating of actions by image-sigamls,

to derive the main specific features of psychic control in comparison.'to regulation of operations in automatic systems.

Thd present analysis makes such a step:.pos.QJOe.only in a mostgeneral. form, applivcable to severial of only the 4m6st"''saential stages.

1. The absence of azy fixed character of-the.:'program of*action in the executive organs and the fact thatl'tbie .p6gam isderived from the "guididg information organized in the form ofimages is exipeeso d in two 4iterr+:elated pointsi a) the same, ex'ec-utive organ of ma for example, the main executive organ -- thehand) can e treise anrunlimitEd diversity of objectival-motbr pro-grams, b) the same program,s gyild4ing the structure o'f the IAM*g-signals directly forming in the process of perception or being theproduct of its intellectual treatment can be realized by variouseJecutive organs (cf N. A. Bernshteyn, 0 Dostroyenii dvizheniy.1On. the :Structure of Motions7, Moscow, 1947, page 91). Both theu nivsrsality 'of the motor programs realized 'by the same organ,and the"indicated broad replaceability of the executive organscarrying out the same program derive clearly from the absence ofany fixed character of the program in the design of the organ andfrom the method of obtaining the program, under which it is whollydrawn from 'the objectival structure of the image-signal. Exhibit-ing, in distinction to the code-signal. "an orderedness of itselements adequately reproducing the structure of the object ofaction, and containing in itself the routes of movements carry-ing out the corresponding operations, the image-signal formsthe kinematic of functioning of not just one, but within thelimits of any rexecutive organ (to the extent that a sufficientnumber of degrees of freedom exists).

2. The image of the object and the conditions of actionreproduce the geometry of three-dimensional space or two-dimen-sional surface. The route of movement realizing the attainmentof this object -i always a line, forming in soMe way in thestructure of the given region of space or the given surface.But the struct-..re of the given region of space or surface includesnot a single t".-jectory of possible movement, but an entire familyor unlimited b+et of lines lying on the given surface. Even in thecase in which the image of the object is a line, for example, theline of a contour, this single line of the cdntour contains initself a set of various trajectories lying on it, reproducingthe same total contour. Thus, the geometry of the image of theobject and conditions of action, including also energetic, force'components, potentially determine the kinematics of an entirefamily of routes of movements realizing the attainment of thegiven object. From this correlation' between the geometry andthe topology-of the image of the object and (the conditions) ofaction, on the one hand, and the kinematics of the correspondingtrajectories of movement, on the other, derives such an essentialcharacteristic of psychic'c6atrol, its variability.

-12-• e

3. From this same general principle derives another main'feature of psychic control, also intimately related to variability,

its generality. There are two main forms of the generality of motorbehavior:

a). th. generality of motor rpdposes .under an unaltered gen-oral .situation of action (ti,, r~sobon6 of gKneralization),

b) the gen rality of •.4tiohii in a transition from one situ-ation to anQthor (the stimulus of..$eneralization).

The first form of generality, that is, the gKnerality ofmotor responses, consists in that the 8sa motor task under anunchanged general structure of the objectival situation san beaccomplished by various approaches, using various routes of move-ment and procedures of action.

The spatio-temporal structure of the objectival image ofthe situation of action includes, as was indicated, an entirefamily of routes of movements, contained in the general geometrical-dynamic structure of the image. In each individual case of reali-zation of the motor solution, a selection is made of one of theroutes and corresponding procedures contained in this image struc-ture. Since with an unchangod situation of action the regulatingimage remains the same, correspondingly the total structure of thesignal of guiding information from whirh the selection of a givenvariant is made is likewise preserved.

Thus, the generality of reactions here is the constancy ofthe image of the objectival situation with variation of the kine-

matic characteristics of individual motor solutions potentiallycontained in the image structure.

The second form of generality of psychically controlledmotor behavior -- the generalization for stimuli -- is expressedin the transfer of motor habits from a situation to a situationwhere their objectival structure shares a commonness, that is,here a singling out of common components of the image occurs.

Both of -M-.e examined forms of generality of psychic controlcharacterize th- process of building motor effects (spesificallyexcutive or cor,.,:.olling) not as a selection and treatment of afixed order of motor acts, but as the formation of a sonco-motorstructure, in wh..ch the sensory image-signal builds its motorcomponents.

4. From the characterized features of psychic controlstems another of its main characteristics, redunda.. Theessence of the latter consists in that the regulating image-signal determines the possibility of accomplishing not only thatvariant of movement which has actually been selected, but .hoentire family cf these variants, whose kinematic characteristiccorresponds to the geometry and topology of the objectival image.In regard to the actually realized solution, the potential pos-sibility of all its other variants deriving from the generalstructure of the imaoe-sial is redundant.

-13-

I

I"f the realization of the selected variant from some externalor internal cause cannot boeachieved, then the existence of methodsof solution redundant in relation to the given variant provides forthe possibility of its replacement. Here, such a replacement caneasily be achieved (undet the regulating action of the general struc-ture of the image)' both: before. the selected variant of motion beginsto be realized, as well a jin'cht midst of its accomplishment.

5. All these. redundant obsaibilities of replacement and.chang6-over also determines such essential features of psychic con-

, trol, as its interference-resistance and reliability. Interference-resistance derives from the integrity of the iage, and the relia-bility.of realization of a certain operation under changing condi-tions (not allowable with systems operating from a code form ofinformation signals) here stems directly from the divereity of thevariants of motor solutions, determining the general objectivalstructure of the image.

All the enumerated features of psychic control find.theirultimate and general expression in such basic characteristics asrelative universality and versatility, manifested to the sameextent both in specifically executive and controlling operations.

The analysis presented shows that all phenomenalogicallyknown characteristics of psychic contr'.1, in distinction to thecharacteristics of regulation in automatic systems, actually etemfrom differences in the methods under which those signals of infor-mation which carry out the process of control are organized.

The method of ordering the image-signal elements in distinc-tion to those of the code-s&:.gnal, reproducing the objectival struc-ture of the information.scurce -- the object of action, determinesthose features of interrelationship of guiding information with.commind which elements are expressed in the above-listed basicforms of psychic control.

The or-,,izedness of elements of sensory signals in theform of objectý.val images (in distinction to code-signals) revealedby analysis or' the regulating functions of various .levels of vie-nals confronto science with the task of further studying themechanism of reflexive dynamics of analyzers, realizing the for.a- -tion of the sensory image as an image-zignal.

Modern knowledge of the analyzer as a feedback system &ndour understanding of the sensory image as the effect of reflexiveacts of this system enables us to examine the structure of thnimage-signal as a process of the synthesis ani ordering of theset of elements forming the signal into an adequate sourco ofinformation of integral structure.

Particularly urgent problems in this investigatiou relateto that link of the analyzer mechanism which provides for thetransition from a frequency-impulse nerve code to an image-aignal,

that is, which accomplishes the operation of decoding. Discoveryof these mechanisms will enable us to advance to the next stageof analysis -- discovery of the mechanisms by which guiding infor-mation is transformed into commahd information, these mechanismlying at the basis of the concrete realization of the working func-tion of a sensory image.Thus, analysis of the real living function of the process of

reflection will lead to singling out its specific structure, anddiscovery of the latter will again permit, although on a more pro-found basis, investigation of the mechanisms of its working, con-trolling functions. Here, the practical action of man at thegiven phase of analysis emerges as the objective criterion andthe source for comprehending the very process of perception asa process of reflection or representation in the proper and pre- 6

cise sense of this term.

10,123CSO: 1872-S - ED -

- 15 -