Document Type: 
Document Number (FOIA) /ESDN (CREST): 
Release Decision: 
Original Classification: 
Document Page Count: 
Document Creation Date: 
November 4, 2016
Document Release Date: 
June 18, 1998
Sequence Number: 
Case Number: 
Publication Date: 
February 1, 1982
Content Type: 
PDF icon CIA-RDP96-00788R001100010004-6.pdf5.28 MB
Approved For Release 2000/08/07 : CIA-RDP96-00788R001100010004-6 February 1982 Vol. 70 No. 2 NE Scanning the Issue Adjustable Speed AC Drives-A Technology Status Review (Invited Paper), Bose, page 116-After nearly a decade and a half of intensive development, adjustable speed ac drives have assumed a position of great importance in a large variety of industrial applications. This has led to a pressing need in the industrial and academic communities for an understanding of the basics and state of the art of this growing technology. This invited paper, believed to be the first of its kind on this subject, is a tutorial-review of ac drives intended to be useful both to the reader who has a general interest in industrial control using ac drives and to the specialist who wishes to extend and expand his or her knowledge. The ac machines which constitute the principal subsystem of these drives are first concisely but thoroughly reviewed in the paper. Following this, the technical features of different classes of drives with converters, such as voltage-fed converters, cur- rent-fed converters, cycloconverters, static Kramer, and static Scherbius drives, are discussed. The paper then describes modeling and simulation of ac drives, and finally the subject of microcomputer control is reviewed in some detail. Through- out the paper, the current trends of development of power semiconductor devices, machines, converters, and controls are indicated wherever possible. The Persistent Paradox of Psychic Phenomena: An Engineering Perspective (Invited Paper), Jahn, page 136-The persistent, fragmentary indication of the possible existence of psychic phenomena requires us, in the interests of scientific integrity and intellectual honesty, to apply the same spirit of objective inquiry that we have focused on the more traditional and accepted branches of science and technology. It was nearly six years ago-in March 1976, to be exact-that the PROCEEDINGS took the unusual step of publishing a paper on the psychic phenomenon called remote viewing. That paper was received with great interest, considerable skepticism, some hostility, and even a bit of shock. In the intervening years, no breakthroughs have occurred to prove or disprove the existence of this and other psychic phenomena, but interest and research have continued unabated. Where does research on these topics n stand now? It was with this question in mind that the paper in this issue was invited. This paper reviews the history, nomenclature, conceptual organization, and status of the generic field of psychic phe- nomena; presents a few detailed examples of contemporary research that could have ultimate technological implications; and attempts to offer a balanced view of the viability and value of continued study of this fascinating and controversial field. The paper is believed to represent the first attempt at a comprehensive engineering survey of the field. Most readers will likely approach it with a healthy blend of curiosity and skepticism. It is hoped all will find it informative and thought provoking. Performance Evaluation of Data Communication Systems (Invited Paper), Reiser, page 171-Great strides have been taken in recent years in developing the architecture of message or packet-based communication networks. Much has been learned about organizing, controlling, and operating such net- works, but performance and design principles were developed in parallel, and networks were built before a complete under- standing of their performance was achieved. Major progress has been made in evaluating the performance of such commu- nication systems, and this invited paper presents a tutorial survey of the state of the art of this field. The majority of the literature on performance evaluation emphasizes protocol evaluation, and the author systematically develops this ap- proach. Following a general introduction to performance evaluation, the paper gives considerable attention to the mathematical tools involved, using the language of the communication engi- neer. The major mathematical techniques are Markov chains applied to discrete time systems and queueing theory. Through- out emphasis is given to how these tools are applied in practice. The mathematical methods do not map on a one-to-one basis to the system, and the leading modeling ideas are highlighted. The embedding environments in which protocols are analyzed are discussed in some detail, and the paper follows the frame- work of layered architecture. Separate sections are devoted to the performance analysis of data-link protocols, path-con- trol protocols, and end-to-end protocols. Approved For Release 2000/08/07 : CIA-RDP96-00788R001100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 1196 PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 11321 M. Schwartz, "Performance analysis of the SNA virtual route pacing control," IBM Res. Rep. RC 8490, Thomas J. Watson Research Center, Yorktown Heights, NY, Sept. 1980. (133] F. Tobagi and L. Kleinrock, "Packet switching in radio channels: Part IV-Stability considerations and dynamic control in carrier sense multiple access," IEEE Trans. Commun., vol. COM-25, pp. 1103-1120, Oct. 1977. 1134] F. A. Tobagi, M. Gerla, R. W. Peebles, and E. G. Manning, "Modeling and measurement techniques in packet communica- tion networks," Proc. IEEE, vol. 66, no. 11, pp. 1423-1447, Nov. 1978. 11351 F. A. Tobagi and V. B. Hunt, "Performance analysis of carrier sense multiple access with collision detection," Comput. Net., vol. 4, pp. 245-259, 1980. (1361 V. L. Wallace, "Algebraic techniques for the numerical solution of queueing networks," in Mathematical Methods in Queuing Theory, A. B. Clarke, Ed. Berlin, Germany: Springer, 1974, pp. 295-305. Contributors Bimal K. Bose (S'59-M'60-SM'78) received the B.E. degree from Calcutta University, Calcutta, India, in 1956, the M.S. degree from the Uni- versity of Wisconsin, Madison, in 1960, and the Ph.D. degree from Calcutta University, in 1966. He was a Member of Faculty at Calcutta Uni- versity where he was awarded the Premchand Roychand scholarship and the Mouat gold medal for research contributions. In 1971, he joined Rensselaer Polytechnic Institute, Troy, NY, as an Associate Professor of Electrical Engineering. He was responsible for organizing undergraduate and graduate power electronics programs, including an advanced course on adjustable speed ac drives. He was a consultant for several industries including General Electric R&D Center, Bendix Corporation, Lutron Electronics, and PCI Ozone Corporation. Since 1976 he has been with General Electric, Corporate Research and Development, Schenectady, NY. His research interests are power conversion systems and micro- computer-based performance optimization of ac drives. Dr. Bose has published 42 research papers and holds a number of U.S. patents. Recently, he edited the IEEE press book Adjustable Speed AC Drives. He is Chairman of IEEE Transactions Review of Static Power Converter Committee, a member of the Power Electronics and Micro- computer Control Committee, the IEEE Industrial Electronics and Control Instrumentation Group, the Scientific Committee of the Inter- national Conference on Digital Control of Electrical Machine, and of the Program Committee of the International Static Power Converter Conference, a member of the Steering Committee for the Workshop on Automotive Applications of Microprocessors, a member of the Fellow Nominating Committee for the IEEE Static Power Converter Committee, and a member of the Program Committee of the Tokyo International Power Electronics Conference. Robert G. Jahn received the B.S. degree in mechanical engineering in 1951, the M.A. de- gree in physics in 1953, and the Ph.D. degree in physics in 1955, all from Princeton University, Princeton, NJ. He has taught at Lehigh University, Bethle- hem, PA, at the California Institute of Tech- nology, Pasadena, and, since 1962, at Princeton University, where he is Dean of the School of Engineering and Applied Science and also serves as Professor of Aerospace Sciences. His principal (137] C. H. West, "General technique for communications protocol validation," IBM J. Res. Devel., vol. 22, no. 4, pp. 393-404, July 1978. (138] J. W. Wong and M. S. Unsoy, "Analysis of flow control in switched data networks," in Proc. IFIP Conf. on Information Processing. Amsterdam, The Netherlands: North-Holland, 1977, pp.315-320. [139] J. W. Wong, "Distribution of end-to-end delay in message switched networks," Comput. Net., vol. 2, pp. 4 -49,1978 [140] J. W. Wong, J.A.B. Moura, and J. A. Field, "Hierarchical model- ing of local area computer networks," In Proc. Nat. Telecom- mun. Conf. '80 (Houston, TX, 1980). 1141] J. W. Wong and S. S. Lam, "Queueing network models of packet- switching networks, Part 1: Open networks," Performance Evaluation, to be published. 1142) P. Zaflropulo, C. H. West, H. Rudin, D. D. Cowan, and D. Brand, "Towards analyzing and synthesizing protocols," IEEE Trans. Commun., vol. COM-28, pp. 651-66 1, Apr. 1980. areas of scholarship include high temperature gasdynamics, ionized gas kinetics, plasmadynamics and plasma propulsion, and his book, The Physics of Electric Propulsion remains the standard text in its field. His interest in psychic phenomena traces to his supervision of a student project begun in 1977, and although his present program in this area remains a very minor part of his professional activity, it has achieved international prominence in the field. Dr. Jahn received the Curtis W. McGraw Research Award of the ASEE, and he is a Fellow of both the American Physical Society and the American Institute of Aeronautics and Astronautics. He has served on several NASA Advisory Committees, and is a member and past-chairman of the Board of Trustees of Associated Universities, Inc. Martin Reiser was born in Switzerland in 1943. He received his education from the Swiss Federal Institute of Technology (ETH) where he obtained the Ph.D. degree in 1971. One year later, he switched from the IBM Zurich Research Laboratory, Riischlikon, Swit- zerland, to the IBM Thomas J. Watson Research Center, Yorktown Heights, NY, where he joined the Performance Evaluation group. He specialized in the numerical solution of queueing networks, and made contributions to the theory of multichain networks and their numerical solution. Together with J. Buzen and H. Kobayashi, he laid the foundation for today's fast algorithms in queueing networks and created one of the first modeling software packages, known as QNET 4. He is author or co-author of numerous publications in the field. After spending about 1 year at the IBM Research Laboratory in San Jose, CA, he returned to Yorktown Heights where he was soon appointed Manager of Performance Evalua- tion. He led efforts in analysis, hierarchical modeling, and system measurement. In late 1977, he started a new group in the area of Computer Communication Network Modeling and Architecture. This activity led to a new assignment in 1978 as Senior Manager of Com- munications and Computer Science back at the IBM Zurich Research Laboratory, a position he still holds. His latest technical interests are in mean-value analysis, a theory he co-founded with S. S. Lavenberg and which he applied to computer networks. Dr. Reiser is a member of ACM, IFIP WG.7.3, and IFIP WG 6.4. Approved For Release 2000/08/07 : CIA-RDP96-00788R001100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 136 PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 The Persistent Paradox of Psychic Phenomena: An Engineering Perspective ROBERT G. JAHN Invited Paper Abstract-Although a variety of so-called psychic phenomena have at- tracted man's attention throughout recorded history, organized scholarly effort to comprehend such effects is just one century old, and systematic academic research roughly half that age. Over recent years, a sizeable spectrum of evidence has been brought forth from reputable laboratories in several disciplines to suggest that at times human consciousness can acquire information inaccessible by any known physical mechanism (ESP), and can influence the behavior of physical systems or processes (PK), but even the most rigorous and sophisticated of these studies display a characteristic dilemma: The experimental results are rarely rep- licable in the strict scientific sense, but the anomalous yields are well beyond chance expectations and a number of common features thread through the broad range of reported effects. Various attempts at theoret- ical modeling have so far shown little functional value in explicating experimental results, but have served to stimulate fundamental re-exam- ination of the role of consciousness in the determination of physical reality. Further careful study of this formidable field seems justified, but only within the context of very well conceived and technically im- peccable experiments of Inge data-base capability, with disciplined at- tention to the pertinent aesthetic factors, and with more constructive involvement of the critical community. Prologue The world of psychic phenomena might be likened to a vast, fog-shrouded swamp, wherein are reported to dwell a bewilder- ing array of bizarre phenomenological creatures, all foreign to our normal perceptual and analytical catalogs. Some scholars who have explored this clouded domain have returned to an- nounce categorically that all such life is illusory-mere sunken stumps and swirling subsurface shadows, inviting misperception by the gullible and misrepresentation by the purveyors. But others of comparable conviction have described in minute de- tail their observations of a variety of extraordinary beings of awesome dimensions and capability. Some of these are claimed to appear unexpectedly, erupting from the roily depths to flash momentarily in the sunlight of human experience, only to disappear again before any systematic calibration of their characteristics can be taken. Others are reportedly enticed to more replicable and controlled behavior, but only by persons of special talent or extensive training. Much invalid, even fraudulent evidence of such activity has been touted by ex- ploiters of these mysteries, thereby casting deep suspicion on all other testimony. When fully sifted, only a very few legiti- Manuscript received July 15, 1981; revised October 26, 1981. This work was supported in part by the McDonnell Foundation, Inc., and by the John E. Fetzer Foundation, Inc., The Explorers Club, the Insti- tute of Noetic Sciences, and the Little River Foundation. The author is Dean of the School of Engineering/Applied Science, Princeton University, Princeton, NJ 08544. mate specimens seem to have been captured, by tediously de- liberate trolling of the brackish domain, or by more incisive invasion of its turbid interior, and even these have proven so incomprehensible and so delicate to exposure, and the imposed criteria for their credibility have been so severe, that they have not been fully persuasive. Yet the goal remains alluring, and the search continues. INTRODUCTION ITH THIS unlikely bit of allegorical musing, I venture to begin the most extraordinary writing task I have yet attempted: to respond to the request of the Editors of this journal for a critical review of the status and prognosis of scientific research into so-called psychic phenomena. I do so with some trepidation, first because the topic is far from my principal line of scholarship and my involvement with it has been brief and tightly circumscribed, and second, because of the intensity of reactions any commentary on this subject tends to call forth from many quarters. For these reasons, it may be well at the outset to specify my perspective on the field and the purpose that I hope this article will serve. My formal training is that of an engineer and applied physicist, and the bulk of my research has concerned a sequence of topics in the broad domain of the aerospace sciences: Fluid mechanics, ionized gases, plasmadynamics, and electric propul- sion. In my present position as Dean of the School of Engineer- ing and Applied Science of Princeton University, I have occa- sion to be involved with an even broader selection of topics selected for undergraduate independent projects, and it was in that context some four years ago that I was requested by one. of our very best students to supervise a study of psychic phenomena. More specifically, this young lady proposed to bring her talents and background in electrical engineering and computer science to bear on some experiments in controlled, low-level psychokinesis. Although I had no previous experience, professional or personal, with this subject, for a variety of pedagogical reasons I agreed, and together we mapped a tenta- tive scholarly path, involving a literature search, visits to appro- priate laboratories and professional meetings, and the design, construction, and operation of simple experiments. My initial oversight role in this project led to a degree of personal involve- ment with it, and that to a growing intellectual bemusement, to the extent that by the time this student graduated, I was persuaded that this was a legitimate field for a high technologist to study and that I would enjoy continuing to do so. Approved For Release 2000/0$/07 : CIA-RDP96-00788RO01100010004-6 de- sive i so >sed !ave and Lure yet tors osis do 'om h it use ject my icle fled nce uid )ul- ,er- ca- )ics was by chic to rnd ed, ice, of ita- 'ro- gn, tial ive- +nt, was .fist Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 JAHN:rPSY IC PHENOMENA: ENGINEERING PERSPECTIVE I have since assembled a small professional staff, secured the requisite funding from a few private sources, and undertaken a modest experimental program in selected aspects of the field that could ultimately have some engineering implications. I should emphasize that my fractional involvement with this program remains quite minor in comparison to my other re- sponsibilities, and that the work is still very preliminary and tentative, but it provides the base of cognizance for my broader observations on the field as a whole. The intention of this article is to provide some balanced per- spective on the modem status of this conceptually and logis- tically difficult subject. Certainly no field of scholarly endeavor has proven more frustrating; nor has been more abused and misunderstood, than the study of psychic phenomena. Dealing as it does as much with impressionistic and aesthetic evidence as with analytical substance, and carrying by its nature strongly subjective and numenistic overtones, it has been incessantly prostituted by charlatans, lunatics, and sensationalists, catego- rically rejected by most of the scientific establishment, and widely misunderstood by the public at large. Interspersed with this, and greatly encumbered by it, a pattern of legitimate effort to comprehend and utilize the purported phenomena has evolved to a point where some dispassionate assessment of its accomplishments can be attempted. The questions addressed by this review are whether, once the overburdens of illegitimate activity and irresponsible criticism are removed, there remains sufficient residue of valid evidence to justify continued research and, if so, how this research might most effectively be styled, facilitated, and evaluated. Before addressing these issues directly, it may be helpful to review briefly the historical evolution of the field, its contem- porary nomenclature and conceptual organization, and the dimensions of current activity. This can then be followed by a general overview and critique of the modem research, and that in turn by more detailed description of a few specific efforts, drawn primarily from our own work. Toward the close, we shall attempt to survey several theoretical approaches to modeling of psychic processes and comment briefly on po- tential implications and applications of the phenomena. In all of this, no tone of advocacy is intended, other than for objec- tive assessment of the evidence in hand. HISTORY In a sense, the study of psychic phenomena is one of the oldest of human endeavors [11-M. As far back as can be traced, mortal man has pondered the supernatural in one form or another. Cave drawings at Lascaux and Altamira, circa 20 000 B.C., reflect this preoccupation, and the religious rites of early societies of both the eastern and western worlds were heavily loaded with psychic formalisms. The classic civilizations of Egypt, Greece, and Rome dealt extensively in psychic process. The Delphic Oracle was politically important from the earliest Hellenic times to the age of Alexander the Great, and was consulted on problems as diverse as the proper measures to stop a plague, the constitutions of Greek city-states, and the best locations for new colonies. Even Aristotle, one of the most empirical of the classical philosophers, examined the causal links in prophetic dreams. Virtually every form of organized religion practiced by man has been thoroughly laced with various forms of psychic mech-, anism. The Bible, like most other basic theological texts, treats Psychic process as a central ingredient, in a tone so matter-of- fact that one is inclined to believe. that people of those times accepted such events rather routinely. Indeed, the Bible is an excellent catalog of psychic phenomena; virtually every cate- gory of effect identified today is illustrated there in one form or another. Christian writers and philosophers, from Augustine to the Reformation, recount many purported instances of psychic phenomena, usually attributed to visitations of divine grace or demonic posession. Secular medieval writing also abounds with supernatural and mystical reference, and even in the Renaissance period it is still difficult to separate psychic allu- sion from religious dogma, although both were then translated into more organized forms in art and literature. Early in the 16th century the celebrated Swiss physician and philosopher Paracelsus wrote extensively on psychic capabilities and po- tentialities. In his words: The mind of man is the microcosmic counterpart of the uni- versal mind ..... ...One man may communicate his thoughts to another with whom he is in sympathy, at any distance however great it may be, or he may act upon the spirit of another person in such a manner as to influence his actions .... [8]. Perhaps the first major scientific commentaries on the topic were offered near the turn of the 17th century by Sir Francis Ba- con, widely regarded as the originator of the scientific method. In The Advancement of Learning he suggested that "supersti- tions and the like" should not be excluded from scientific study, and in his posthumous book, Sylva Sylvarum, he proposed deliberate investigation of telepathic dreams, psychic healing, and the influence of "imagination" on the casting of dice [9). Some years later, a group of British intellectuals including Henry More and Joseph Glanvill met regularly to discuss paranormal topics, and in 1681 Glanvill published the substance of these studies in a book entitled Saducismus Triumphatus [10]. Meanwhile, some four centuries of public and church hysteria over sorcery and witchcraft, as manifested in a sequence of trials, tortures, and executions, had begun to subside, and by the mid-18th century, the Roman Church authorized Prospero Lambertini, who later became Pope Benedict XIV, to carry out a scholarly investigation of reports of psychic events. His conclusions, recorded in De Canonizatione [ 11 ], were surpris- ingly unecclesiastical: namely, that 1) psychic experiences were not necessarily divine miracles, but could occur to "fools, idiots, melancholy persons, and brute beasts"; 2) apparitions had little to do with sanctity or demonic entities; 3) prophesy occurs more often in sleep than in waking; 4) it is difficult for a prophet to distinguish his own thoughts from extrasensory messages; and 5) predictions frequently take. symbolic forms. In all of these, he presaged to some degree modern thoughts on these topics. At roughly the same time, Anton Mesmer's discovery of hyp- nosis opened an alternative route to demonstration and study of unconscious psychic process that has continued to this day. Early reports of hypnotized subjects performing telepathic or clairvoyant tasks were common [ 12] [ 13 1, and although much of this evidence might now be discounted on the basis of in- adequate experimental control, interest in hypnosis specifically} and in various altered states of consciousness generally, as facilitators of psychic experience persists into some of the modern experimentation. Also in this mid-18th century period, a spiritualist move- ment focused on extrasensory contact with the dead, possibly influenced by the work of Emanuel Swedenborg [ 14 J, [IS], germinated in this country as well as in England, and by the Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 138 - PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 19th century had reached the dimensions of an organized religion. Symbolic of the popular preoccupation with the-topic, Mary Todd Lincoln was-reported to have held seances in the White House in the early 1860's [ 161 . A classic two-volume work by F.W.H. Myers, entitled Human Personality and the Survival of Bodily Death 117), brought the topic to its acme of sophistication, but eventually the fanaticism the movement attracted and its fraudulent exploitation created, a negative attitude in the scholarly community which prevails yet today. Despite these millennia of human concern with the paranor- mal, orderly and organized scholarly search for verification and understanding of psychic phenomena began only a century ago, with the establishment in London in 1882 of the Society for Psychical Research, in whose Proceedings appeared the first formal publication of controlled experiments in telepathy and clairvoyance [ 131, [181, [ 191. Three years later the counter- part organization in this country, the American Society for Psychical Research, was founded in Boston by several distin- guished scientists and philosophers. Because of financial difficulty, this shortly merged with the British group, but re- emerged in 1905 as a separate entity with its own professional journal, and has continued as such to the present 1201. Although the SPR attracted a barrage of criticism from the scientific and intellectual communities, it also attracted significant participation of eminent scholars from established fields. Numbered among its presidents are three Nobel Lau- reates, ten Fellows of the Royal Society, one Prime Minister, and a substantial list of physicists and philosophers, including Henry Sidgwick, Frederic W. H. Myers, Lord Rayleigh, Sir J. J. Thomson, William McDougall, Edmund Gurney, Sir William Crookes, Sir William Barrett, Henri Bergson, Arthur, Earl of Balfour, Gardner Murphy, G.N.M. Tyrell, Charles Richet, Gilbert Murray, and one of the most articulate contributors to the evolution of critical thought on this topic in this period, the Harvard psychologist and philosopher, William James. One of the founders of the ASPR, James wrote extensively and eloquently on behalf of objective and disciplined study of psychic phenomena [211-[251: Any one with a healthy sense for evidence, a sense not methodi- cally blunted by the sectarianism of `science,' ought now, it seems to me, to feel that exalted sensibilities and memories, veridical phantasms, haunted houses, trances with supernormal faculty, and even experimental thought-transference, are natural kinds of (phenomena) which ought, just like other natural events, to be followed up with scientific curiosity [ 25 1. Entering the 20th century, a new perspective on psychic phe- nomena was provided by the emergence of psychology .as a scholarly discipline, and especially by the early efforts in clinical psychology and psychoanalytic therapy. The patriarch of this evolution, Sigmund Freud, was a member of the SPR and con- tributed, albeit somewhat reluctantly, to its publications [261, [271. His recognition and exploration of the unconscious mind and of the function of dreams prompted Myers to suggest a possible explication of various psychic effects which is still of theoretical value [ 17 1. Freud's interest in parapsychology increased toward the end of his life, and he is reported to have conceded informally that were he to begin his career anew, he would focus on this topic. Freud's former protege, Carl Jung, who had written his Ph.D. thesis on the psychology of occult phenomena, pursued ex- ploration of the unconscious to deeper dimensions of para- normal experience, publishing widely on such subjects as telep- athy, mediumship, synchronicity, the collective unconscious, and theoretical models of psychic process. [281-[30]. In Memories, Dreams, Reflections, he stated: . the relationship between doctor and patient, especially when a transference on the part of the patient occurs, or a more or less unconscious identification of doctor and patient, can lead to parapsychological phenomena. I have frequently run into this [30]. Jung's collaboration with the eminent physicist Wolfgang Pauli on the topic of synchronicity clearly influenced the subsequent evolution of both careers and of fundamental concepts in both disciplines [ 311. Although much of the established psychologi- cal community has since rejected parapsychology as a valid discipline, some interest has been retained by a few clinical practitioners, presumably because of the demonstrated con- comitance and similarities of apparent psychic experiences with certain psychological processes [ 32 ], [ 331. It was also early in this century that the first organized academic studies of psychic phenomena were mounted. One of the more visible of these devolved from gifts and a bequest from Thomas W. Stanford, brother of the founder of Stanford University, to endow psychic research at that institution, and to this day the university provides support of a "psychic research fellow" and retains a collection of so-called "apports" indicative of the donor's long personal involvement with the field. Modest research programs were also undertaken at Harvard and a few European universities in the first decades of this century, as evidenced by occasional publications in various established journals. The benchmark academic effort, however, germinated at Duke University in the late 1920's, when William McDougall, who had been James' successor at Harvard, arrived to chair the department of psychology and appointed J. B. Rhine and Louisa Rhine, "to study the claims to scientific value of the field known as psychical research." Their early tentative efforts in the study of postmortem survival gradually evolved into a laboratory for controlled research in "extrasensory per- ception," as they first termed the process. In this laboratory were established many of the basic concepts and protocols of modem psychic research, as well as the first extensive and sys- tematic data bases of several types of psychic experimentation. The professional and personal history of the Rhines and their laboratory is a fascinating saga in its own right, but would take us too far afield here 1341-[371. A few excerpts from a 1967 address of J. B. Rhine to the American Psychological Associa- tion, in which he attempted to summarize his first two decades of intensive study, give hint of the inherent attractions and frustrations of this field, and of the man's optimistic vision: The phenomena that were being studied began to show lawful interrelations and even a degree of unity. One by one the major claims, based originally only upon spontaneous human experi- ences, were subjected to laboratory test and experimentally veri- fied.... Certain general characteristics of the psi process became clear during this period. The most revealing of these is the sub- ject's lack of conscious control over any type of psi ability, a characteristic which accounts for its elusive nature. It was new methodological ground, even for psychology.... Also, we were surprised to find that psi ability is widespread, probably even a specific human capacity rather than a capability possessed by a few rare individuals as had been the popular belief. Evidence that psi is not linked with illness or abnormality was another welcome advance.... By 1951... a healthy young science was emerging [381. In 1937, the Rhines began publication of the Journal of Parapsychology, which remains a leading journal in the field today. A professional organization calling itself the Parapsy chological Association was formed in 1957, and in 1969 was . Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 JAHN: PSYCHIC Pro phi' ?#vfOrE~ &rQ$6/08/07: CIA-RDP96-00788RO01100010004-6139 accepted as an affiliate by the American Association for the Advancement of Science. At the present time, there are eight English language publica- tions covering this field [39], supplemented by numerous less formal magazines and countless books of widely varying quality and relevance. Research activity is reported from some twenty U.S. universities and colleges and at least as many institutions in Western Europe [40], but in most cases it is of very small scale. There are very few academic programs of study, although some fifty M.A. and Ph.D. theses have been accepted on psychic topics at reputable universities over the past forty years [41 ] . Some ten research institutes and private corporations in the United States have also authorized publications and reports in the field [42] . The extent of Eastern Bloc and Oriental efforts [43]-[541 and of classified research in this country are matters of considerable speculation on which I cannot comment with authority. Further review of contemporary programs will be attempted in subsequent sections, following an outline of modern nomen- clature and conceptual organization of the topic. In closing this historical overview, we might simply observe that in many re- spects the growth pattern of this field resembles that of the natural sciences in their earliest days, or perhaps even more the incubation of classical psychology, in terms of the ab- sence of replicable basic experiments and useful theoretical models, the low level of financial support and internal profes- sional coordination, and the low credibility in the academic establishment and public sectors. Also like those fields, the survival and early growth of psychic research can largely be attributed to the efforts of a few scholars of sufficient convic- tion, stature, and courage to withstand the rejection of the orthodox communities. NOMENCLATURE AND CONCEPTUAL ORGANIZATION Before turning to an assessment of contemporary research, it may be useful to specify some notation and delineation of the field, to an extent consistent with the present limited comprehension of the phenomena. First, let us agree to a global definition of "psychic phenomena" (frequently denoted by "psi" or "0") to include all processes of information and/or energy exchange which involve animate consciousness in a manner not currently explicable in terms of known science. Similarly, let "psychic research" imply any scholarly study of such phenomena employing scientific methodology, as opposed to any dogmatic, ritualistic, or theological approaches. Within these definitions, the field may then be roughly divided into two major categories: extrasensory perception (ESP) and psychokinesis (PK). ESP refers generally to the acquisition of information from sources blocked from ordinary perception. Under this category are included such subdivisions as telepathy, which refers to detection of another person's thoughts; clairvoyance, which refers to contemporary perception of remote physical objects or events; precognition and retrocognition, which refer to perception of future events and events in the past not accessible by normal recollection; and animal ESP, which encompasses a variety of seemingly inexplicable capabilities, such as homing, psi-trailing, collective behavior, communication, etc. PK (occasionally termed telekinesis, or psychoenergetics) re- fers to a palpable influence of consciousness on a physical or biological system. The interaction may be deliberate or spon- taneous, and the energy transfer involved may range from microscopic disturbance of atomic-level processes, through macroscopic distortion or levitation of objects, up to some very drastic "poltergeist" effects. Psychic healing and man- plant interactions would be two examples of PK in biological systems. Note that in its major subdivision into ESP and PK, the field conforms to two of the main categories of present-day science and high technology, i.e., that encompassing the extraction, con- version, transmission, storage, and utilization of information, and that comprising the same sequence of processing of energy. Other domains of psychic research can be identified which do not fit into these major categories of ESP and PK and with which we shall not be further concerned in this article. Examples would include research into survival of death, and the family of "out-of-body experiences (OBE)," including astral projec- tion, autoscopy, and bilocation. The following table attempts to summarize the subdivisions in a concise form. Categories of Psychic Phenomena 1. Extrasensory Perception (ESP) A. Telepathy B. Clairvoyance C. Precognition/Retrocognition D. Animal ESP II. Psychokinesis (PK) A. Physical Systems B. Biological Systems III. Survival A. Reincarnation B. Apparitions C. Mediumship IV. Out-of-Body Experiences (OBE). Clearly this particular arrangement is neither unique nor orthogonal. Elements of one category frequently appear in the context of another, e.g., precognitive clairvoyance; tele- pathic healing, etc., and occasionally an assignment is ambigu- ous, e.g., a particular effect may be regarded as precipitated by PK, or simply to be forecast by precognition. Notwith- standing, the table may aid in keeping the subsequent illustra- tions in some order. PATTERN OF CONTEMPORARY RESEARCH By its nature and heritage, modern psychic research remains rather diffuse and lightly structured, making any attempt to catalog the work by institution or laboratory, or by tracing developments of given lines of effort, rather ineffective and pre- mature. Instead, it may be more useful to comment on the pattern of attention to this field by academic disciplines, noting the variations in emphasis, methodology, representation, and interpretation brought to bear, using specific projects only as illustrations with no implications. of hierarchy or attempt at completeness. Even in this format, no recitation of specific research results or conclusions will be attempted, since these can be misleading or incomprehensible when extracted from the detailed context of their experimental arrangements and protocols. In later sections, an effort will be made to follow a few sample experiments through to their particular results and conclusions. By far the most sustained and broadcast attention to this field has been given by a cadre of scholars with professional backgrounds in classical psychology, comprising a controversial subdiscipline termed "parapsychology." This group has tended to approach the field with the traditional psychological proto- cols and vocabulary, and to interpret results in the context of Approved For Release 2000/08/07 : CIA-RDP96-00788R001100010004-6 Approved For Release 2000/08/07 CIA-RDP96-00788RO01100010004-6 140 their clinical, cognitive, and behavioral psychological heritages, with the natural consequence that they have concentrated mainly on the ESP category of psychic tasks, although some PK work dots their recent literature. Perhaps the most exten- sive class of parapsychological research has attempted to correlate psychic performance with personality variables. The age, sex, creativity, openness, hostility, extroversion, motiva- tion, and intelligence of the participants as indices of ability to perform ESP tasks have been explored at length, and some sig- nificant correlations, most notably with positive a priori atti- tudes toward the tasks and with outgoing, creative personalities, have been reported from several laboratories. Other studies have searched for connections between psychic performance and dream recall, learning and response strategies, memory, and feedback [ 5 51- [ 611. A more aggressive style of parapsychological research has in- voked a variety of altered states of consciousness in attempts to enhance psychic process. These have included various natu- ral and traditional practices, such as sleep, meditation, and pro- gressive relaxation [62]-[67] ; more mechanical sensory in- hibition strategies such as hypnosis, isolation and "ganzfeld" [66], [681-[701; and a few controversial efforts with drug- induced states [ 711. Physiological correlates have also been sought, using conventional EEG, GSR, and plethysmographic equipment to monitor neurological, cardiovascular, and mus- cular response to psy is effort [351, [72]-[75'. The diffi- culty of obtaining successful replications of previously positive results and an observed common tendency of participant per- formance to deteriorate over time ("decline effect") have led to systematic study of the role of the experimenter in eliciting results, i.e., to consideration of the possible influence of the experimenter's personal attitudes, expectations, and style of interaction with his subjects, as well as the overall environmental ambience of his laboratory, on the experimental yield [ 761 - [81]. Despite its present recalcitrance toward more systematic study of psychic phenomena, the richly diverse, rapidly matur- ing parent field of psychology continues to offer an expanding array of modern methodologies and models which could be brought to bear on increasingly sophisticated study of this sub- ject. Computer-assisted linguistic analyses; psychoneurological studies of attention, perception, and concept formation; social learning theory and similar approaches to human interactions; and the emerging formulations of transpersonal and humanistic aspects of human consciousness, all bear possible relevance to comprehension of various aspects of this ultra-difficult step- child, but at the moment, the low level of financial support, and negative professional peer pressures have discouraged such enterprise. The involvement of physicists in psychic research, while con- siderably less extensive, has been no less dedicated and no less controversial. Since the days of Sir Francis Bacon, a number of noted physicists have made excursions into this field, usually to the bemusement and ridicule of their colleagues of the day. One of the most notable of these was Sir William Crookes, discoverer of the element thallium and pioneer in the physics of low pressure discharges, whose broadside pro- fessional and personal battles with the scientific establishment over this issue make entertaining, and possibly enlightening, reading [821. Sir Isaac Newton was intensely involved in the study of alchemy, including some of its more metaphysical aspects [831, and as already mentioned, Lord Rayleigh and J. J. Thomson were active members of the S.P.R. PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 In more recent years the attention of physicists has influenced development of the field in at least three ways. First, their in- terests have focused more on the PK.category of problems, i.e., the interaction of human consciousness with physical systems, to balance the predominant ESP interests of the parapsycholo- gists. Second, more sophisticated experimental equipment than has typically been available to the psychological com- munity has been brought to bear on the identification and correlation of very low-level physical effects. Third, the tradi- tional theoretical physics formalisms have been directed to the proposition of various models of psychic phenomena, from whence has arisen some hope of establishing the traditional dialogue between critical experimentation and theoretical hypothesis essential to any ultimate comprehension and appli- cation of such phenomena. Typical of the modern physicist's specific contributions to the field have been the development and application of a variety of electronic random event generators (REG) for the purpose of identifying and correlating PK abilities in human subjects [841-[931, and similar application of magnetometers [941, torsional pendula [ 95 1, lasers [ 96 ], interferometers [ 97 1, and electronic strain gauges [98], [99] to a variety of other PK tasks. On the theoretical side, a number of applications of quan- turn mechanics, statistical thermodynamics, electromagnetic theory, and other formalisms to the representation of psychic process have been proposed [1001-[1131, and attempts at some philosophical correlation of the phenomena with other previously or presently obscure physical processes have been suggested [1141. Again, despite the open identification of a few distinguished personalities with such efforts, a more broadly held categorical rejection of the field has inhibited much collaborative or systematic attention to it. Up to this time, the involvement of engineers with psychic research has been very recent, very sparse, and very much along the lines of the experimental physicists. Beyond our own program, which will be outlined in some detail below, I am aware of only a very few engineering laboratories addressing any aspects of the field in any substantial and deliberate way [115]. These have so far tended to concentrate on applied physics types of experimentation and on aspects of informa- tion processing, rather than on more empirical technological applications. Another community of scholars to influence the pattern of psychic research comprises the statisticians and other applied mathematicians and logicians who have been concerned with the proper evaluation and interpretation of the research data. In the absence of any experiments displaying rigid causal replicability, all of the inferences and hypotheses about psy. chic phenomena have necessarily been based on either anec dotal or statistical evidence. The former defy any systematic representation; the latter are vulnerable to alternative inter pretations and hence to impressionistic bias and argument. Early in the emergence of mathematical statistics as at integral discipline, S. S. Wilks found himself involved in controversy over the validity of the statistical procedures o early psychic researchers, and published some recommenda tions for methods that could be applied to telepathy experi ments [ 1161. Since that time, much of the commentary fron the critical community has addressed perceived flaws in th statistical methodology underlying the experimental evident [1171, [1181, and the advocate community has reacted b; paying disproportionate attention to this aspect of their logic Most of the encyclopaedic references in the field contai Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 MEN,: Ps`t'CHICA11pg90ovesdaR R S2OOO 8/07 : CIA-RDP96-00788RO01100010004-6 141 substantial components on the statistical methods [119], and the leading journals regularly display intense dialogues on specific statistical issues [1201-[124]. At least one of these journals routinely refers all articles submitted for publication to a consultant statistician as a part of their review process. A few illustrations of the statistical questions that can arise in psychic experiments appear in the detailed examples presented below. A number of other disciplines have played roles in the evolu- tion of the study of psychic process and continue to contribute, albeit somewhat more peripherally. A succession of philoso- phers, from Aristotle through James and Bergson to C. D. Broad in the present era, have mused on these topics [211-[251, 11251-[1291. The intersections of the field with anthropology, theology, and history have been approached from many per- spectives ranging from aesthetic to analytic, and from dogmatic to scholarly [1301-[1341. Its relevance to the study and practice of medicine has been an enduring and intense debate, focusing in the present day on the propriety and efficacy of holistic health strategies and psychic healing. Isolated instances of interest on the part of chemists, biologists, geologists, and archaeologists can be found, and the application of psychic techniques in criminology and law enforcement, while less rigorous than most of the academic efforts, contributes further anecdotal evidence to the overall data base. In the arts and humanities, the topic continues to provide stimuli for a variety of creative compositions. Still other areas could be cited, but these become progressively more satellite than central to the task of this paper. Rather than pursuing this disciplinary survey further, it seems prefer- able next to focus in greater detail on a few contemporary studies that can serve to illustrate more specifically the bizarre phenomena, the awkward and tedious protocols, and the un- conventional theoretical concepts that arise in this class of research. Although these will provide a better sense of the status of the field than any attempt to summarize results from the diffuse multidisciplinary pattern of effort sketched above, one general assessment may be useful at this point. Namely, throughout all of the work just skimmed, and that sharpened somewhat below, I am aware of no reputable investigator who has claimed, let alone demonstrated, any psychic experiment approaching classical scientific replicability. What have been put forward are a varied assortment of observations, currently inexplicable in terms of established science, which display cer- tain common phenomenological and psychological features, and which could have substantial implications for basic physical theory and ultimate practical applications. The following ex- amples are presented in that spirit. PSYCHOKINESIS The first group of experiments selected as more detailed illus- trations of contemporary psychic research are drawn from the general subdomain of PK. As defined above, this broad cate- gory of purported psychic phenomena encompasses the possible influence of human consciousness on the behavior of physical or biological systems or processes, and comprises several loosely related classes of effect characterized by different scales of energy, forms of manifestation, replicability, and statistical be- havior. Confining attention to interactions with physical objects or systems, the most popularly publicized class features the de- formation, levitation, or other macroscopic disturbance of ob- jects, as commonly propounded by professional performers, mediums, and various Eastern practitioners [135]-[137]. Al- though a number of serious efforts have been made to submit such demonstrations to rigorous scientific testing, these have tended to yield only equivocal confirmations, fodder for the critical community, and some embarrassment and frustration to the investigators. Of a yet more bizarre nature are the family of very rare and spectacular "poltergeist" effects, more technically termed "recurrent spontaneous psychokinesis" (RSPK), wherein are reported specific major events of levitation, vibration, telepor- tation, and breakage of a wide range of objects, a variety of acoustical and electromagnetic phenomena, and various optical aberrations [138], [139]. , For years these phenomena were naively attributed to manifestations of the spirit world, or return of the dead to "haunted" houses, and inspired countless horror movies and pulp-magazine articles. Recently, some order has been brought to this weird business by systematic surveys of documented poltergeist cases undertaken by A. R. G. Owen, W. G. Roll, J. G. Pratt, and others [138]-[142]. In one of these surveys, 116 cases of reported poltefgeist activity, rang- ing back to the year 1612, were re-examined. Of these, 92 were found to be associated with particular individuals living in the affected dwelling, most of whom were adolescents, and most of whom were affected by some neurological/emotional ailment, most commonly epilepsy. Often a precipitating traumatic event could be identified which seemed to initiate the activity, after which the general pattern involved a period of relatively mild precursor events, a sequence of major dis- turbances, and a period of "after shocks," extending as much as several weeks beyond the main events. Controlled experimen- tation on poltergeist phenomena has proven virtually impossible because of their infrequent and unpredictable occurrence pattern and because of the delicate physiological and emotional situations prevailing in most cases, which have taken precedence over the technical investigations. Nevertheless, these processes have retained some fascination because of the magnitude of the energy transfer involved and because of their apparent correlation with specific types of individuals and neurological disorders. The most systematic and persuasive studies of PK, however, have dealt with much more modest scales of physical distur- bance, in some cases reaching down to the atomic level. This somewhat more viable domain has been addressed by numerous investigators in various ways, but basically one of two strategies is followed. In one approach, relatively simple physical systems are employed-mechanical, electrical, optical, thermal, etc.,- each of which involves a particular component or process that is ultra-sensitive to disturbance. The experiment is arranged to signify such disturbance by a relatively large change in some dis- play which provides feedback indication to the operator, much in the spirit of a biofeedback instrument, and simultaneously to provide some form of permanent data record. Examples of this class of experiment would include the use of magnetometers, torsional pendula, optical interferometers, electronic strain gauges, glow discharges [ 143 1, and sensitive thermistors [ 144 ] . In the second approach, attempt is made to distort the nor- mal statistical patterns of various random physical processes on either a microscopic or macroscopic scale. In a sense, these ex- periments deal with energy rearrangement within the systems, i.e., with their information content or entropy, rather than with energy transfer to the system per se. The earliest versions of this class employed dice, or other simple mechanical implements of well-known statistical behavior [37], [92], [145]-[148], but more recent studies have tended to employ more sophisti- Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 , Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 FRINGES ON SCREEN RAMP DRIVE cated apparatus, such as electronic REG's like those described in detail below. Although the first two categories of PK feature more dramatic effects and thus have enjoyed greater popular attention, the smaller scale phenomena seem more amenable to controlled experimentation and theoretical interpretation, and the re- mainder of this section deals only with such. Perhaps the most efficient means of elaborating on this type of research would be to review briefly the spectrum of such experiments in progress in our own laboratory, and then to display and attempt to interpret data from one of them. In so doing, we intend no neglect of other work noted in the references, but simply deal from greater familiarity. As examples of the first class of low-level PK experiment mentioned above, we have in operation a Fabry-Perot optical interferometer, a dual-thermistor bridge, and a photoelastic strain detector. In the interferometer experiment, shown in the photograph and schematic of Fig. 1, a Coherent Optics Instrument #360-370, using a diffuse sodium lamp as source, is adjusted to produce circular fringes on a screen visible to the operator (Fig. 2). Small changes in the separation of the inter- ferometer plates cause the fringes to migrate radially inward or outward. By visual observation of the fringe movement, plate motions of less than 0.1 wavelength can be readily detected. Via a. pinhole in the screen, the brightness of the central fringe is recorded by a photomultiplier/chart recorder system at an order of magnitude higher sensitivity, thereby preserving quan- titative output data while the operator simultaneously sees an the fringe pattern in a stated direction relative to the normal baseline drift of the instrument. The protocols involve rigid control and monitoring of the environment of the instrument and surrounding laboratory, and the interspersing of baseline responses with active PK efforts obtained under otherwise identical conditions, including the position of the operator and any other personnel relative to the instrument. In pilot studies with this device, a variety of fractional-fringe responses were observed, using several different operators and various initial interferometer settings. A more formal procedure has since been developed which provides more precise conditions for an ongoing series of trials. In this protocol, the central fringe is set initially on a maximum gradient position between a bright and dark fringe, and its progress monitored for subse- quent periods of baseline or PK effort. Encouragingly repli- cable data have been obtained from a number of different operators, in the form of chart recordings of 5-min PK trials with interspersed 5-min baseline drifts of the instrument. Using computerized graphic, regression, and spectral analyses of the data, it is possible to discern characteristics in the hierarchy of trace derivatives and the Fourier spectra which, while not definitive, display certain recurrent features [97] . No physical interpretation has been attempted other than to acknowledge that the observed fringe migrations could also be indicative of slight changes in the index of refraction of the air in the plate gap or in the wavelength of the light source, as well as of a displacement of the plates. The dual-thermistor experiment comprises a much more sen- attractive optical display of his progress for use as feedback. sitive version of a multiple-thermistor arrangement on which ~l@ latRceleea8iaiQ4O/QZritro4Af RDR9 U7c88RN4 OO S Schmeidler [ 1441. mal igid ent line use and lies ere tial ce for ge is se- ?li- 3nt 'als ng he by lot ge we he of en- ich 4]. JAIaN: PSAp row"i c. R Ie 29LWtQ J4 : CIA-RDP96-00788RO01100010004-6 Fig. 2. Interferometer fringe patterns. As shown in Fig. 3, two Omega Engineering thermistors, Model UVA 3254, each with its own electronic bridge and volt- age source, are connected differentially to a Tektronix l A7A oscilloscope preamplifier and to a visual feedback display. With suitable ground planes and cable shielding, sensitivities greater than 0.001 K are obtainable, and by subtracting the two output signals the major portion of spurious electrical and mechanical interference is eliminated. The effects of ambient thermal vari- ations in the laboratory are essentially excluded by enclosing each of the thermistors in identical Pyrex flasks immersed in a large liquid reservoir, in which configuration the undisturbed system exhibits stable baselines over long periods of time. Using the same interspersed baseline protocol as in the inter- ferometer experiment, the task of the operator is to achieve an increase in the reading of one thermistor with respect to the other or in some more subtle fashion to alter the PK response relative to the baseline. Some such effects have indeed been ob- served, but little systematic data have so far been accumulated on this experiment. Also in a preliminary stage is an experiment to monitor in- ternal strain in a solid specimen via photoelastic optical tech- niques. Several studies have been reported on the PK deforma- tion of solids, but most of these have employed conventional engineering strain gauges or microacoustic sensors as detectors [98], [99], [135], [149], [150], both of which require sub- stantial interface electronics before a feedback signal reaches the operator, leaving unclear the role of the sensor in any pos- sible PK influence. - Although less sensitive than the electronic methods, photoelastic techniques have the advantage of relating the operator more directly to the sensitive element of the ex- periment via an attractive optical fringe pattern much like that of the interferometer (Fig. 4). This equipment and technique may also be applied to a sensitive levitation experiment wherein Fig. 3. Dual-thermistor apparatus. the object is suspended on a photoelastic lever arm of suitable dimensions. Within the second category of low-level PK experiment, we are employing or are now constructing several devices based on random physical processes, some macroscopic in scale, others deriving from atomic-scale processes. The largest of these in- volves a 6 X 10 ft apparatus, shown in Fig. 5, which drops some 10 000 - in spheres through a "quincunx" array of 336 nylon pegs in about 12 min. As a consequence of the multitudinous collisions with the pegs and with each other, the spheres are dis- persed into a good approximation of a Gaussian distribution as they fall into 19 collecting compartments at the bottom. The goal of the operator is to distort the distribution in some prescribed fashion to a significant degree compared to empirical baseline experience. Photodiode counters mounted in funnels at the entrance to each bin provide real-time digital displays of the bin populations to supplement the more qualitative feedback of the growing ball stacks seen by the operator and to provide quantitative data for on-line statistical analysis. Fig. 5 shows a typical baseline distribution for this device and a distorted distribution obtained in a particular PK effort. Full statistical analysis of the significance of any particular achieved pattern is a challenging problem in its own right, since it must deal with a combination of 19 bin populations, each of which has its own empirical baseline mean and standard deviation, all constrained by total ball count. A somewhat similar experiment, not yet refined, employs a device which allows small metallic or dielectric spheres to bounce on an optically flat, precisely horizontal circular plate of glass, which is oscillated by a vibration coil at frequencies from 10 Hz to 20 000 Hz. In the absence of any external dis- turbance, a sphere started at the center of the plate executes a random walk toward the outside edge, arriving with equal prob- ability at any azimuth. Since the sphere may make as many as I0$ collisions in the process, it is vulnerable to statistical distor- tion of its trajectory and consequent terminus. The task of the operator is preferentially to direct the sphere to a prescribed terminal quadrant. In an attempt to intervene with a random physical process at the atomic level, we have constructed a large glow-discharge device whose luminous patterns are indicative of the mean free path of the current-carrying electrons against inelastic excita- tion collisions with the background gas. This device, shown in Fig. 6, presents a 36-in X 2-in diam cylindrical glow marked by a sequence of bright and dark zones along its positive column Approved For Release 2000/08/07 : CIA-RDP96-00788 R001,100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 ___ i2 " "fit1Lltlt ...:~"~t9Ittl ; . Fig. 5. Gaussian analog device and distributions, typical of dc discharges in a given range of gas pressure and terminal voltage. The number and locations of these striations are sensitive to the electron inelastic mean free path, which in turn depends on the gas type and density, the electron tempera ture, and the local electric field. Striation position is monitorec by photoelectric detectors, and the goal of the operator is t( expand or contract the pattern on demand, to a significan extent compared to the normal background jitter and drifi Protocols are much the same, output data take the same ger eral form, and are analyzed by the same algorithms as in th interferometer and photoelastic experiments. Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 1982 1, JAHN: P # " t or. 142Q1Q Wi617 : CIA-RDP96-00788R001100010004-6 145 uu ELGENGO NOISE SOURCE S 3602A15124 LOW- PASS FILTER SAMPLER GATE SAMPLE PULSE GENERATOR Fig. 7. Functional diagram of REG. II s _ ~~ _ I ~M ~. (a) ~~ ~ IIII! VIII (b) :mpera- nitored or is to riificant d drift. me gen- s in the Fig. 8. REG waveforms: (a) Filtered noise. (b) Clipped noise. (c) Sampling pulses. A number of other atomic-scale random system PK experi- ments are under consideration, design, or construction, involv- ing such processes as information storage on a microelectronic chip, the spontaneous decay of phosphorescent surfaces, lam- inar to turbulent transition in a fluid stream, atomic and molec- ular resonators, and resonant acoustical or electrical cavities, but none of these is far enough advanced to merit description here. Rather, we shall concentrate for the remainder of this section on a more detailed presentation of our most serviceable experiment, and the one on which we have the largest data base, the electronic REG.- RANDOM EVENT GENERATOR EXPERIMENTS REG's have been the most widely used and most productive facilities for experimentation with low-level PK. Although a broad variety of such devices exist, most involve four conceptu- ally and functionally separable components: an electronic noise source; a sampling system which examines the noise at pre- scribed intervals and prepares an output pulse train correspond- ing to the samples thus obtained; a system which analyzes the pulse train in accordance with preset instructions and prepares suitable output for a feedback system; and the feedback dis- play itself, which informs the operator -of the results of the 'sampling process. The particular version we have employed utilizes a packaged commercial noise source module based on a solid-state junction and precision preamplifier (Elgenco Model 3602A15124), but modules employing radioactive decay units or glow discharges can be readily substituted. This source produces a random noise spectrum up to several megahertz, which our logic circuit first filters to a flat spectrum from 50 to 20 000 Hz, then amplifies and clips to the flat-topped profile shown in Figs. 7 and 8. This is then sampled by a regular train of gate pulses, yielding a cor- responding random succession of positive and negative output pulses indicative of the sign of the noise at the time of sampling, and these are then counted. Since the average time between Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Approved For Release 2000108/07 : CIA-RDP96-00788RO01100010004-6 INTERNAL NOISE SOURCE External Noise Source . r SAMPLMIG hit RATE GENERATOR Rote Control O +1 Etternot Pulse Generator ( S micro sec.Widlh) Set N PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 SCALE-OF-N COUNTER Reset Signal To Event Counters 7 Analog Output TRIAL COUNTER AND LED DISPLAY Fig. 9. Electrical schematic of REG. zero crossings of the clipped noise waveform is about 30 ?s, sampling rates to about 15 000/s can be tolerated with statis- tical independence. The full functional array is sketched in Fig. 9, and a photo- graph of the boxed units in Fig. 10. By panel setting the sampler may be instructed to take "trials" of 100, 200, or 2000 samples, at a frequency of 1, .10, 100, 1000, or 10 000/s. The Train of N Puts" Print Analog CUMULATIVE Output Gate OFF on _ _ uainur. SCALE-OF-SO START COUNTER CIRCU OL % A Manuel Start or Automatic Repeat 450 Times) Reset to Printer Trial Counter counting system may be set to count only positive pulses, only negative pulses, or to alternate positive and negative counting on successive samples. The alternating positive/negative mode effectively factors out any systematic bias in the noise source, and is the mode employed in all the experiments reported here. The counting results are displayed by LED arrays tracking both the running count of each trial and the concurrent mean rela- Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788 R001100010004-6 1982 JAHN: PSYCHIC PHENOMENA: ENGINEERING PERSPECTIVE 147 TABLE I REG 200-SAMPLE DATA SUMMARY Series Instr. No. Trials Mean Std. Dev. t-score Pr n+/n- REG I B.L. 12 000 100.009 6.994 0.144 0.443 5678/5611 PK+ - 4 550 100.264 7.037 2.528 0.006 2230/2056 PK 3 850 99.509 7.063 -4.313 10-5 1716/1926 A PK 8 400 4.890 5 X 10-7 REG II B.L. 2 500 100.033 6.875 0.239 0.406 1188/1179 PK+ - 1950 100.247 6.849 1.590 0.056 916/919 PK 1800 99.597 6.775 -2.526 0.006 797/902 APK 3750 2.920 0.002 REG III B.L. + 3 500 99.977 7.013 -0.193 0.424 1658/1655 PK - 2 400 100.227 6.821 1.634 0.051 1150/1086 PK 2 600 99.736 7.026 -1.918 0.028 1192/1270 A PK 5 000 2.507 0.006 E REG I B.L. 18 000 100.006 6.981 0.115 0.454 8524/8445 II PK+ - 8 900 100.250 6.938 3.403 3 x 10-4 4296/4061 III PK 8 250 99.600 6.989 -5.203 10-7 3705/4098 A PK 17 150 6.107 5 X 10-10 REG la no B.L. PK+ 2150 100.206 7.091 1.340 0.088 1059/993 PK- 2100 99.945 6.937 -0.365 0.358 954/1019 APK 4250 1.213 0.113 REG Ila B.L. 5 000 100.186 6.974 1.882 0.030 2367/2337 PK+ - 2000 100.117 7.041 0.746 0.228 955/950 PK 1750 99.941 6.898 -0.360 0.359 803/839 A PK 3 750 0.772 0.220 E REG I B.L. 23 000 100.045 6.980 0.978 0.164 10891/10782 la II PK+ 13 050 100.223 6.979 3.644 10-4 6310/6004 IIa III PK- 12100 99.709 6.968 -4.596 2 X 10-6 5462/5956 A PK 25 150 5.828 3 x 10-9 only sting lode .rce, sere. both rela- tive to a preset origin and are permanently recorded on a strip printer. For most of the experiments described below, an AIM- 65 microprocessor interface is also utilized to insert the trial- count data on-line into processing routines supported by a TERAK Model 8510 used as a terminal and PDP 11/45 and VAX 750 employing a UNIX operating system programmed in C language. All of the sampling, counting, and display functions can be simply checked by referring them to an internal or ex- ternal calibrated pulse train generator. The device also has a manual/automatic option, whereby it will either collect its trial samples only when a panel switch or parallel remote switch is pushed, or it will repeat that process for 50 trials automatically once activated by the switch. The operator thus has the option of triggering each trial or of initiating a repetitive flow of 50 such trials with no further intervention. The experiments reported here were performed by a single operator, seated in front of the device with the remote initia- tion switch in hand and the LED count indicators and TERAK terminal display visible. This operator attempted, on instruc- tion or volition, to distort the trial counts either toward higher or lower values. The several options of sampling number, sam- pling frequency, +/- polarity, and manual/automatic sequenc- ing were variously determined by random instruction, operator preference, or experimental practicality, and recorded before the beginning of each trial. Clearly, the full matrix of such pos- sibilities could not be explored, and for our first sequence of experiments only 200-sample trials were used, at 100 or 1000 counts/s, all counted in the +/- alternating mode. The auto- matic/manual and high/low options were more thoroughly tested, in both the volitional and instructed choice modes. Fifty trials of the 200-sample units comprised a test run, and data from these were processed individually and in many concat- enations via a statistics package in the UNIX system developed specifically for this task. Calculated were the mean, standard deviation, range, kurtosis, skew coefficient, z-score, t-score, X2 goodness-of-fit with both 8 and 16 degrees of freedom, and the corresponding one-tailed probabilities against chance of the last four measures. Applied to earlier and ongoing baseline data, this analysis confirmed that in undisturbed operation this REG conforms very well to a Gaussian approximation to the appropriate full binary statistics. The major portion of the results listed below comprised three separate experimental series, extending over fifteen months, labeled REG I, REG II, and REG III, respectively. All other data acquired under slightly less formal conditions of protocol during this period, included for completeness, are grcluped under two other series, labeled REG la and IIa, respectively. Details of these series protocols, calibration tests, and their in- dividual results are available in the reference [931. All told, over 25 000 active PK trials were obtained, corresponding to more than 5 000 000 binary events. Table I summarizes all of the baseline and PK data acquired during these five experimental series. A total of 23 000 base- Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 100 COUNT P ? 05 J t-------- 4000 6000 8000 10000 12000 14000 NUMBER OF TRIALS 6000 5000 4000 3000 2000 1000 0 2000 4000 6000 8000 10000 NUMBER OF PAIRS OF TRIALS Fig. IS. REG cumulative deviations in direction of effort of all 200? sample data. 2000 4000 6000 8000 10000 12000 14000 NUMBER OF TRIALS Fig. 16. REG cumulative average deviations of PK* and PK- 200-sample data. line trials were taken under a variety of conditions before, during, and after the active PK trials. Their overall mean was 100.045, and their standard deviation 6.980, compared with the values of 100.000 and 7.071 predicted by the theoretical Gaussian approximation to the appropriate binary statistical distribution. As shown in Fig. 11, the frequency of count dis- tribution conformed very well with the theoretical curve. The results of the PK trials are also presented in Table I and in Figs. 12 and 13. Briefly, the 13 050 high-instruction trials, denoted PK' , yielded a mean of 100.23 and a standard devia- tion of 6.979; the 12 100 low-instruction trials, denoted PK , . yielded a mean of 99.704 and a standard deviation of 6.968. The one-tailed probability of chance occurrence of the former, computed from t-score, is about 10-4 ; of the latter, about 2 X 10-6. The combined probability of the split, i.e., of this total "direction-of-effort" success, denoted APK, is about 3 X 10-9. ( A number of more elaborate statistical measures have been applied to these data; the results are not qualitatively changed thereby.) As is evident from Figs. 12 and 13, and as is verified by the more detailed statistical tests performed, no significant distor- tion of the frequency-of-count distributions other than the shifting of the means has occurred. In other words, the ob- served effect is to shift the total distributions intact, rather than to distort any of their higher moments significantly. This result clearly has felicitous implications for this class of experi- mentation, since it allows much simpler and faster data collec- tion and analysis than might otherwise have been anticipated. It is illustrative to exhibit the overall data behavior via graphs of the cumulative deviation of the trial score means versus the accumulated data base. Fig. 14 shows such a representation for the total data pool plotting PK, PK-, and baseline data with reference to cumulative 0.05 confidence levels. Fig. 15 uses a similar representation for compounding the PK+ and PK- data in a "direction of effort" cumulative deviation. (Had REG Ia and IIa been excluded from these data, the overall slopes would have been slightly more. severe and uniform.) Alternatively, the cumulative data may be presented in terms of the progressions of the average deviations from the theoreti- cal mean, as shown in Fig. 16, where the stochastic variations Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 BASELINE --------------- P1- ,05 BASELINE tl S! e; 0 h b sl sl tl h t( h, h t1 gi tl S( lil al an 1 N( tiv to istor- a the .e ob- :ather This xperi- ;ollec- ite.d. ;raphs us the tation e data rig. 15 A PK- d REG slopes i terms teoreti- iations JAHN: P9YCAHICpProedNAOfNGIFEERIRZ`i 990919 07 : CIA-RDP96-00788R001100010004-6 149 of the small data bases are seen to damp out to well-defined this operator, who claimed no special talent for this or any terminal values after several thousand trials. other psychic task, reported that any conscious variations in The PK+ and PK- effects also manifest themselves in terms psychological strategy, such as focus of visual attention, or of the number of trial means recorded above and below the intensity of concentration or desire, did not appear to have theoretical value. As displayed in Table I, PK"' efforts were any evident effect on the yield. Similarly, differences in the generally characterized by an excess of trial means above laboratory ambience, such as the lighting level, background 100.00 .and PK- efforts by an excess below. The total con- noise, or peripheral presence of other persons, did not seem catenations in these terms are significant at a level of 0.003 to influence this operator's performance. When queried about for PK' and 3 X 10-5 for PK-. any impressionistic sense of the interaction process, the opera- The ensemble of results acquired in these experiments dis- tor alluded to a "resonance or identification with the system, play certain instructive general features: leading to a loss of self-awareness similar to that experienced .1) The importance of accumulating very large data bases in a game, a movie, or some creative occupation." Clearly this when dealing with such marginal phenoma is emphasized by class of parameter will be the most difficult to specify and cor- the relative scales of the statistical vagaries and the broader relate, and we are far from any definition of its mechanisms. systematic trends in Figs. 14-16. Although the trends are With respect to experimental options on the equipment pa- established early in the data collection sequence, unambigu- rameters, we can make very limited explorations with the ous departures from the accumulated vagaries of chance be- acquired data base. Briefly, binary correlations of the data for havior occur only well into the total 25 000 trial, 5 000 000 the 100/1000 counting rate option, for the volitional /instructed bit, sequence. direction of effort, and for automatic/manual sampling give 2) Over this large a data base, there arises some quantitative little indication of importance of such factors in the overall statistical regularity in the PK process, epitomized by the mean performance. Each category shows clear and significant separa- slopes of the cumulative deviations in Figs. 14 and 15 and by tion of the means for the PK+ and PK- efforts, with little to the terminal values of the average deviations in Fig. 16. Traced choose between the t-scores for the various categories. Thus, back to the elemental binary samples, these values imply direc- at least for the data base at hand, the process seems insensitive ted inversions from chance behavior of about one or one and a to these particular experimental parameters. half bits in every one thousand or, alternatively, of 0.2 or 0.3 We have also attempted correlation in terms of the trial- bits per trial. number sequence. With cognizance of the ubiquitous "de- 3) The differences between the somewhat larger values for cline effect" which is reported over a broad range of psychic the PK- deviation and the lesser values for PK+ are only mar- experimentation, we have prepared an algorithm which cross- ginally significant on this data base, but prevail rather uniformly concatenates from the data base all scores achieved on the throughout all the test series. The suspicion that these reflect first trials of the experimental run, all achieved on the second some subtle bias in the REG itself is not supported by the base- trials, etc., up to the fiftieth, and arrays those fifty means in a line data, which concatenate to a grand mean very slightly graphical form. The results show little systematic profile of above the theoretical value. yield versus trial number. A similar exercise has been per- One of the primary goals of such controlled PK studies at formed to cross-concatenate the data by run-number over the this early phase in the understanding of the phenomena is to various series to search for a decline effect on that larger scale, develop experiments of sufficient yield and replicability that but again no significant correlation is found within this data various parametric correlations may be systematically explored, base. thereby hopefully separating the consequential from the incon- The most extensive parametric exploration attempted to date sequential factors. The experiments outlined above hold some was motivated by the apparently fundamental question implicit promise of serviceability for this purpose, but a great deal of in general conclusion 2) above, i.e., whether the magnitude of data will need be accumulated to establish any such correla- the observed effect correlates with the total number of bits tions. Four classes of parameters could be considered: those processed, or with the number of trials. To explore this aspect, associated with the experimental equipment; those associated the same operator has performed a second ensemble of expert- with the operator's physiological and emotional characteristics; mental series totaling 25 000 trials, all consisting of 2000-sam- those associated with the operator's technique; and various en- ple bits rather than 200. As before, various combinations of virQnmental factors not directly associated with either. So far the automatic/manual and volitional /instructed modes were we have accumulated only small amounts of data from other employed, but to speed data acquisition and reduce the opera- operators, and given the general indication #1 above regarding tor's tedium, only the 1000/s counting rate was used. This, the importance of large data bases, we can make no statement coupled with the more elegant data processing capabilities that about the generality or peculiarity of our principal operator's had evolved over the preceding experiments, allowed this performance. Similarly, we have attempted no systematic vari- sequence to be completed in less than six months. ation of external environmental factors, and although test times, The results of this effort, as presented in Table II and Figs. dates, durations, and laboratory temperature, pressure, and 17 and 18, are curiously ambivalent. As before, there is clear humidity have been routinely recorded, we cannot comment and' significant separation of the means of the PK+ and PK- on the importance of this category of parameter. efforts, and the baseline is well behaved. As could be antici- On the matter of operator technique, it should first be re- pated from the larger standard deviation of the 2000-bit data, emphasized that the sole formal difference between the PK+ the cumulative traces display larger statistical fluctuations and and PK-?trials is the specified intention of the operator to in- require a larger number of trials to settle toward well-defined fluence the device to generate numbers in the assigned direction. terminal values. To the quantitative precision allowed by this No other variation in protocol is permitted, save those subjec- data base, these terminal values appear not to endorse any tive differences in psychological attitude the operator chooses simple bit-level hypothesis in that they fail by a factor of 6 or to invoke. Although no records of such aspects were kept, 7 to achieve the one or one and one-half bits per thousand Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 '15Approved For Release 2000/08/07 : CIA-RD>REDgiGi8or~THE1T9K, VAL. 70, N062, FEBRUARY 1982 TABLE II SUMMARY OF REG 2000-SAMPLE DATA z 0 3750 w 0 0 3750 U Intr. No.. Trials Mean Std. Dev, t-score Pr B.L. 12 500 1000.016 21.879 0.079 0.468 6157/6088 PK+ 12 200 1000.380 21.906 1.914 0.028 6092/5897 PK- 12 800 999.569 22.005 -2.216 0.013 6218/6351 A PK 25 000 - - 2.920 0.002 -JS-J11/ pi-.05 BASELINE ----/-pt 05 - 780 0 2000 4000 6000 8000 10000 12000 14000 NUMBER OF TRIALS Fig. 17. REG cumulative deviations of PK+and PK - 2000-sample data. inversion accomplished in the 200-bit trials. However, the new values are larger on a per-trial basis by a factor of about 1.7, which is not negligible in this context. Again, much more data of this sort will be required to come to grips with this class of correlation. In addition to continuing study of this sample-size parameter, our next generation of experiments employs a number of other operators to explore the variation of yield with operator type and technique, and a number of alternate noise sources, includ- ing pseudo-random sources, in an attempt to localize the effect somewhat and thereby to narrow the range of future experi- ments and models. The results outlined above are by no means the only conse- quential REG data available for contemplation. Of particular interest are a variety of experiments reported by Schmidt, some employing pseudo-random as well as physically random sources, and others using taped source outputs recorded well in advance of their presentation to the operator [87]-[90]. In another approach, May has recently reported an REG study using electronic gear specifically designed to preclude very subtle artifacts that might confound the effects of interest, and includes in his paper a thorough search of the modern REG literature [ 911. In addition, considerable research in the parapsychological community has been performed using REG devices as drivers for various forced-choice video games employed in both the PK and clairvoyance modes [ 1511. Many of these claim sig- nificant yields, but rarely are the data-bases sufficiently large to present quantitative trends, or to allow much parameteric correlation. Regardless of their particular implementation, any potential vulnerability of random electronic noise sources to incidental or intentional distortion by the means under study here would seem to be of some interest to a number of engineering com- munities, given the proliferate application of such devices in various functional and computational capacities. REMOTE PERCEPTION As a second example of contemporary psychic research that has displayed some substantial yield and interlaboratory repli- 0 2000 4000 6000 8000 10000 12000 NUMBER OF TRIALS Fig. 18. REG cumulative average deviations of PK' and PK- 2000- sample data compared to 200-sample data. cability, we select a topic which has come to be called "remote perception" or "remote viewing." The basic concept of this process is far from new; in the early 16th century, Paracelsus stated it unequivocally: Man also possesses a power by which he may see his friends and the circumstances by which they are surrounded, although such persons may be a thousand miles away from him at that time [ 8 ] . In its modern form, the experimental protocol requires a "percipient" to describe, by free-response oral or written narrative or drawing, a remote, unknown target location at which is stationed an "agent," with whom there is no normal sensory mode of communication during the course of the experiment. The targets are usually selected by some prescribed random process from a previously prepared pool of targets, which is unknown to any of the active participants. The quality of the perception is assessed by various impressionistic or analytical judging methods described below. Historically, this experiment has evolved from several genera- tions of free-response clairvoyance and telepathy experiments, which were found to have certain advantages over the more traditional "forced-choice" ESP tasks, such as the Xener card identifications of the early Rhine laboratory [341-[37], in displaying less tendency for percipient stagnation and "decline- effects" over extended testing, and in maintaining some of.the spontaneity of anecdotal clairvoyance experiences. One of the earliest detailed reports of such free-response studies appears as a book by Upton Sinclair entitled Mental Radio, which features an equivocal foreword by Albert Einstein [1521. More modern work of this class was performed at the Mai- monides Medical Center by Ullman and Krippner in the 1960's, and reported in their book Dream Telepathy [64]. From this work emerged the so-called "ganzfeld" or sensory inhibition perception studies of Honorton and many others which pro- pounded the desirability of emotionally stimulating tasks to which the subjects could relate in a personal and spontaneous fashion [66], [67]. The contemporary version of the remote perception protocol was introduced in a sequence of publications by Targ and Put- hoff [94], [1531-[156], which prompted a substantial num- ber of attempted replications [1571-[1741, and considerable Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 JAHNi PSYCHIC Apt Q4rF EEW@9M 08/07: CIA-RDP96-00788RO01100010004-6 151 Fig. 19, Remote perception target: Picnic area, Feathered Pipe Ranch near Helena, MT; 12:00 N MDT, Sept. 5, 1978. Percipient transcript: Princeton, NJ; 8:30 A EDT, Sept. 5, 1978: "Outdoors... open landscape-large areas of trees-pines? Inter- spersed with open fields. Single road. High overcast, cool, breezy. (Agent) in dark jacket talking to someone near road-possibly turn- out area or picnic area. Assembly of stones-possibly pylon or, marker or wall. Large sign somewhere." critical comment. The most extensive of the replications, con- ducted by Dunne and Bisaha in the Chicago area over the period 1976 to 1979, comprised 40 formal trials to which were applied 157 independent transcript judgings, 84 of which assigned first- place rank to the proper targets [ 1611, [ 1621. The type of data which can be acquired in such studies is illustrated in the sequence of Figs. 19-22. In each case is shown a photograph of a particular target, selected by some random process, which was visited by an agent on the date and time indicated. Below each figure is a portion of the cor- responding percipient transcript, with the time and location of the perception effort also noted. The examples shown are drawn from a variety of experimental series conducted under somewhat different protocols, but serve to display some of the characteristics which commonly appear in the more successful efforts: 1) The overall ambience of the scene is accurately perceived. 2) Certain details are accurately identified; others are mis- construed or totally ignored. 3) A feature which is impressive to the agent is not necessarily so to the percipient, and vice versa. 4) The composition of the scene may be distorted by errors in scale, relative positions of key objects, or total right left inversions. - 5) The aesthetic aspects, such as colors, general shapes, degree of activity, noise level, climate, and other ambient features tend to be more accurately perceived than more analytical details such as number, size, or relative positions. 6) The perception is not necessarily centered on the defined target, and may even provide accurate information on ad- jacent areas external to the target, unnoticed by the agent. 7) The fidelity of the perception seems to be independent of the remoteness of the target, up to distances of several thousand miles. 8) The time of the perception effort need not coincide with the time the agent is at the target. Perceptions obtained several hours, or even days, prior to the agent's visit to the target, or even prior to selection of the target, display at least as high a yield as those performed in real time. Fig. 20. Remote perception target: Woodrow Wilson School, Princeton, NJ; 2:15 P EDT, Aug. 28, 1980. Percipient transcript: Princeton, NJ; 12:15 P EDT, Aug. 28, 1980: "Some kind of courtyard, enclosed by buildings on two sides. Paths or walks around periphery, a statue or monument of some kind in the middle surrounded by grass. Could be a fountain; I have the feeling of water. Trees or tall hedges on one side. Fairly quiet, but some people walking around. Not sure of sound, the idea of a fountain sug- gests sound of water but I'm not sure I really hear it or not." The philosophical and practical implications of items 7 and 8 are clearly substantial. If the data are valid, the most parsimo-? nious explications would require access of the percipient's con- sciousness to other portions of the space-time grid than that in which it is currently immersed, or that it can reach by normal processes of communication or memory. These same items also seriously delimit the potential physical mechanisms for such access. Rigorous evaluation of the data from experiments such as these is. confounded by the psychological components of the process, by the impressionistic nature of the information in- volved, and by the inevitable subjective biases of all those par- ticipating in the experiment. Doubtless the earliest and most primitive assessments were informal a posteriori exchanges of impressions about the target between agent and percipient which, although possibly informative and stimulating to them, lacked any quantitative basis and held little scientific credibility. In a somewhat less vulnerable strategy invoked more recently, the percipient, after completing his transcript, visited several possible targets drawn from the pool and attempted to identify the one he perceived, or to rank-order each of them in terms of conformity to his earlier perception. Statistical arguments could then be applied to these ranks to estimate the likelihood that information about the target had been acquired by means other than chance [ 1751- [1771. In an attempt to separate the possible ESP functioning of the percipient during the visitation and ranking process from his original perception effort, the protocol subsequently evolved to invoke independent judges who were provided copies of the various transcripts and taken to the target sites to perform their preferential rankings. Even in this form, the technique has been criticized for possible sensory cuing of the judges [ 1781 and has tended in turn to be replaced by a protocol wherein the judges perform their ranking on the basis of photographs of the targets, usually taken by the agent at the time of the trial. In one such version, the judges, who have not been in- volved in any earlier portion of the experiment, are asked to compare a single percipient transcript with agent-generated descriptions and photographs of a number of alternative targets, Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 Fig. 21. Remote perception target: Rockefeller Chapel, Chicago, IL; 2.15 P CDT, June 10, 1977. Percipient transcript: Mundelein College, Chicago, IL; 1:00 P CDT, June 10, 1977: "I'm seeing a heavy wooden door with a black bolt on it rounded at the top in a dome fashion. I have a feeling of opening the doors and looking in and it's dark inside. My feeling at the moment is that it's a building like a church. And I can see the pews. There is some light but them I feel basically a kind of darkness in there and a quietness. I'm seeing little turrets, very elaborate-looking little turrets, a whole series s offthem ima like across the entire top of the building and there's a straight line and then up to a trianglI.m hgetting ave a demo star nag glass wand wl s -type that statue arch marble, flowing robes. I see the door again and I see some stairs. I think it's very hig building the ornamentation the of it ecture blue. ark d took shape and hey quite 1 k there'sta sect onion the top with t e turrets and then be elow that there are some other kinds of d signs but mo e l nearadesigns. and it looks t tha "I again have a vision of the doors and then maybe a ledged area or an which iareah f this uil air or at prof u es with s mreend of ai desig and there maybe a couple ofthose before you get to the top part There , le filigree turrets s or r something. And within in the building there Sesort oof a continuation of arches, ns It looks like it's ached, but possibly they meet columns or something like Ap~~re~lt i"1~' Pg X011 1O? r -F~DP96-00788RO01100010004-6 and but hem ,tue, arch )oks and igree like , JAHN: pS-Appre EFerEkt#aa?eG209M8 7 : CIA-RDP96-00788RO01100010004-6 153 Fig. 22. Remote perception target: Danube River, Bratislava, Czecho- slovakia; 3:00 P European Standard Time, Aug. 24, 1976. Percipient transcript: Minoqua, WI; 8:30 A CDT, Aug. 23, 1976: "I have the feeling that the agent is somewhere near water. I seem to have the sensation of a very large expanse of water. There might be boats. Several vertical lines, sort of like poles. They're narrow, not heavy. Maybe lamp posts or flag poles. Some kind of circular shape. It's round on its side, like a disc, it's like a round thing flat on the ground, but it seems to have height as well. Maybe with poles. Could possibly come to a point on top.. Seeing vertical lines again. Seems to be a strong impression, these vertical lines. Predominant colors seem to be blue and green. Water again., Some very quick impression of a fence, a low fence. The steps seem to go up to some kind of fence. It's a dark fence and it's along like a walk sort of at the top of the steps. The steps sort of lead up to like a path or walkway. Like a boardwalk. And there's a fence along it. There's people walking along it, and there's vertical lines along that walkway." including the proper target, and to rate or rank-order them by some prescribed criteria. Again, statistical assessment of the significance of the rankings follows. The bulk of the remote perception data reported in the litera- ture has been evaluated by some form of these independent judging processes, and displays sufficiently high yield to encour- age further refinement of the protocol and analysis before attempting categorical judgment on the validity and viability of the phenomena. ' Beyond minor tightening of the target selection, agent maneuver, and perception acquisition and re- cording phases of the experiment, the major potential improve- ments would still seem to pertain to the judging process, which remains potentially vulnerable to subtle cues in the transcripts, to vagaries in the judges' capability, to their subjective biases toward the individual experiments and to the topic as a whole, and to possible psychic input of their own [ 1611, 1621. In an effort to improve the judging process further, our labora- tory has explored. the applications of various information theo- retic methods to the quantification of the data, and an analytic technique has been devised which is based on a limited binary alphabet of target/perception descriptors [1791. While less sensitive to the Gestalt impressionism and symbolic represen- tations which a human judge might capture, this method does provide a rudimentary framework for evaluation of signal-to- noise ratio in the information transfer, and an assessment standardization less dependent on subjective interpretation. In essence, the strategy is to replace impressionistic assessment of the quality of a perception by the identification of specified elements of information therein, after which a mechanical scoring and ranking procedure takes over. In the hope of con- veying a bit more substantive flavor of the data acquired in remote perception experiments and the processing thereof, permit us to describe this analysis in a little detail. The heart of the method is the establishment of a code, or alphabet, of simple descriptive queries which may be addressed to all targets and all perceptions, responses to which serve to distinguish them and to permit quantification of the informa- tion acquired in the perception process. In one version, these "descriptors," thirty in number, are posed in binary form and range over a spectrum from quite factual discriminations, e.g., whether the scene is indoors or outdoors, whether trees are present, or whether there are automobiles, to much more im- pressionistic aspects, such as whether the ambience is noisy or quiet, confined or expansive, hectic or tranquil. The particular ensemble of descriptors has evolved in part through personal experience and intuition, and in part through trial-and-error application to various pilot data. The goal has been a balanced alphabet whose elements are a) relatively unambiguous; b) com- monly perceived by a broad selection of percipients; c) individu- ally instructive in defining the scene; d) complementary to one another; and e) sufficient in number to permit reasonable syn- thesis of the scene, but not so numerous to burden the data collection or computation excessively. Given this descriptor alphabet, each target in the pool is then represented in terms of 30 binary bits, corresponding to the ap- propriate YES/NO responses to the queries. This encoding is normally performed by the agent at the time of target visita- tion, although reference may be made to the target selector's judgment or to photographs of the target for verification. Each perception is similarly rendered into a corresponding sequence of binary digits, but only after the percipient has been allowed to form a free-response impression of the target. Vari- ous scoring recipes are then invoked for quantitative comparison of the perceptions with the targets, using for computation the UNIX operating system of a PDP 11 /45 or VAX 750. The simplest recipe merely counts the number of correct responses to the 30 descriptors, i.e., the positive correlations between the target and descriptor matrices. This does not nor- mally provide a particularly accurate index of the quality of the individual perceptions, since the a priori probabilities of the various descriptors are widely different. For example, a given pool may have more outdoor than indoor targets, and hence a correct identification of an indoor context should be given higher credit than identification of an outdoor context. To facilitate such weighting, a step is included in the computational program to provide the a priori probabilities of all descriptors in the prevailing target pool, on the basis of which more elabo- rate scoring recipes may be invoked. Since the various targets have substantially different charac- teristics and hence different capacities for achieved scores, a variety of normalization procedures also have been developed, using as denominators the total number of descriptors, the per- fect score, i.e., the score that would be achieved for a given target if all descriptors were identified correctly, and various "chance" scores for the target, defined by some random or Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 PROCEEDINGS OF THE IEEE, VOL. 70, NO. 2, FEBRUARY 1982 TABLE III PRECOGNITIVE REMOTE PERCEPTION 24 X 24 "CHICAGO" SERIES PROPER TARGET RANKS Human Jud es E Avg. Avg./24 g Mean Rankb 1 3/4 7 2 5.9 0.25 2.7/8 = 0.34 2 1/2 2 1 1.9 0.08 1.0/8 = 0.13 3 1 1 1 1.0 0.04 1.5/8 = 0.19 4 2/8 7/2 2/2 4/2 2/2 4.5 0.19 2.7/8 = 0.34 5 7/3 9 11 4 8 8.0 0.33 1.7/8 = 0.21 6 9/5 12/2 7/2 16/2 7/2 11.0 0.46 3.5/8 = 0.44 7 13/2 11 14 10 13 12.3 0.51 2.3/8 = 0.29 8 20/3 22 20 14 19 19.2 0.80 1.818 = 0.23 9 4/2 4 8 1 5 4.5 0.19 2.6/7 = 0.37 10 10/7 13 13 9 5 10.6 0.44 1.4/7 = 0.20 11 9/4 9 11 6 12 9.7 0.40 3.6/7 = 0.51 12 1/3 2 2 2 5 2.6 0.11 1.8/7 = 0.26 13 1/3 1 3 1 3 2.0 0.08 2.2/7 = 0.31 14 1 2 2 1 1 1.4 0.06 1.4/7 = 0.20 15 1/2 2 2 1 1 1.5 0.06 1.0/7 = 0.14 16 2 2 2 1 2 1.8 0.08 1.0/10 = 0.10 17 1/4 9 1 9 1 4.5 0.19 1.0/5 = 0.20 18 2/2 10 5 1 2 4.1 0.17 1.0/5 = 0.20 19 14/3 14 14 19 17 15.8 0.66 unjudged 20 7/6 .11 11 8 10 9.9 0.41 5.0/6 = 0.83 21 1/2 2 3 4 1 2.3 0.10 2.0/6 = 0.33 22 5/6 7 12 2 7 7.1 0.30 3.0/6 = 0.50 23 16/4 23 11 15 9 15.1 0.63 3.0/6 = 0.50 24 3 3 4 3 4 3.4 0.14 2.0/6 = 0.33 5.79 6.67 0.28 0.31 a Computed rank/number of ties for that rank. bAssigned rank/number of possible ranks. arbitrary process of descriptor response. A "selective" scoring/ normalization process has also been applied which effectively allows the percipient to reject any descriptor on which he feels unqualified to comment, and thence to be scored only on the reduced descriptor set. The statistical significances of these various normalized per- ception scores are assessed by a collective ranking process reminiscent of the traditional human judging techniques, but having the advantages that the ranking proceeds on a much more standardized and analytical basis, and that many more alternative targets can be ranked by the machine than by a human judge. Specifically, the program scores each transcript not only against its proper target, but against every other tar- get in the pool, and then ranks these targets in order of descend- ing score and specifies the rank of the match with the proper target. This process is repeated for every scoring method, and the results displayed in corresponding matrix arrays. Table III displays typical results' of these analytical ranking procedures as applied to a group of 24 perceptions of 24 targets in the Chicago area. Tabulated are the ranks of the proper tar- gets compared with all other targets for each of the perception efforts, as computed by five of the scoring methods we have found to be most instructive, namely, A) number of correct descriptors/total number of descriptors; B) weighted full de- scriptor score/perfect score; C) weighted lull descriptor score/ number of descriptors; D) weighted selective descriptor score/ perfect score; and E) weighted selective descriptor score/chance score. Also included in the table are the mean ranks assigned by independent human judges subjectively comparing these per- ceptions with a much smaller number of alternative targets. Although the bases of comparison are quite different, it appears that in the majority of these cases the analytical and impressionistic evaluations concur at least roughly in their estimate of the quality of the perceptions, particularly for those which consistently obtain low rank assignments. If the analyt- ical computation is carried through using as target pool only those alternative targets available to the human judges, the agreement in mean ranks is found to be somewhat closer, per- haps fortuitously so, given the categorically different bases of assessment implicit in the two methods. To this analytically scored and ranked data it is possible to apply a variety of statistical assessments of widely ranging sophistication and complexity. Consistent with the rather broad mesh of the descriptor code and the elementary scoring recipes invoked in this version of the concept, we confine our- selves to correspondingly simple statistical measures which provide at least semi-quantitative indication of the yield beyond chance. Specifically, we address only the distribution of proper target ranks achieved in the series of perceptions, such as sum- marized in columns 2-6 of Table III. Using the common z- score method for a discrete distribution, the probability of achieving the mean rank of any of these columns by chance may be directly computed. Table IV displays the results of such calculation for the same 24 X 24 "Chicago" series. Note that, whereas all of the methods suggest significant departures of the computed mean ranks from chance, there is relatively little disparity among them, indicating that the specific method of scoring and normalization is not a sensitive element in the overall evaluation of the perception series. The departure of the shape of the proper target rank distribu- tion from chance is also displayed in Table IV in terms of the number of perceptions achieving first-place ranks, the number rank its a T1 num den( in w pool cipie Nori Be illusi metl Eurc utilh The perci local roug tion resul previ consi To some score Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 Y 1982 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100010004-6 JAHN: P Y HIC PHENOMENA: ENGINEERING PERSPECTIVE TABLE IV PRECOGNITIVE REMOTE PERCEPTION. 24 X 24 "CHICAGO" SERIES SCORE SUMMARY Method' Mean Rank No. 1st (2nd) Ranks No. Ranks Below Mean Z Pz X2(4) PX Chance expectation 12.5 1.0 (1.0) 12 A 6.73 4.4 (4.0) 19 -4.08 2 x 10-5 18.5