The Hyman-Utts statistical debate (1995)

esearch AN EVALUATION OF REMOTE VIEWING: RESEARCH AND APPLICATIONS MICHAEL D. MUMFORD, PHD ANDREW M. ROSE, PHD DAVID A. GOSLIN, PHD PREPARED BY THE AMERICAN INSTITUTES FOR RESEARCH SEPTEMBER 29, 1995 3333 K Street, NW Washington, DC 20007 (202) 342-5000 AN EVALUATION OF REMOTE VIEWING: RESEARCH AND APPLICATIONS MICHAEL D. MUMFORD, PHD ANDREW M. ROSE, PHD DAVID A. GOSLIN, PHD PREPARED BY THE AMERICAN INSTITUTES FOR RESEARCH SEPTEMBER 29, 1995 Table of Contents Executive Summary ............................. E-1 Chapter One: Background and History .............. 1-1 Chapter Two: Evaluation Plan ..................... 2-1 Chapter Three: Research Reviews .................. 3-1 Review 1 (Dr. Jessica Utts) ........................ 3-2 Review 2 (Ray Hyman) ............................ 3-43 The Reply (Dr. Jessica Utts) ..................... . . 3-79 Points of Agreement and Disagreement ........... . . . 3-83 Conclusions from the Expert Reviews .............. . . 3-86 Chapter Four: Evaluating the Utility of Remote Viewing in Intelligence Operations ...................... 4-1 Chapter Five: Conclusions ........................ 5-1 Appendix A - Reviewer Vitae Appendix B - Bibliography Appendix C - Interviewer Material Appendix D - Star Gate Operational Tasking Evaluation Executive Summary Executive Summary Studies of paranormal phenomena have nearly always been associated with controversy. Despite the controversy concerning their nature and existence, many individuals and organizations continue to be avidly interested in these phenomena. The intelligence community is no exception: beginning in the 1970s, it has conducted a program intended to investigate the application of one paranormal phenomenon - remote viewing, or the ability to describe locations one has not visited. Conceptually, remote viewing would seem to have tremendous potential utility for the intelligence community. Accordingly, a three-component program involving basic research, operations, and foreign assessment has been in place for some time. Prior to transferring this program to a new sponsoring organization within the intelligence community, a thorough program review was initiated. The part of the program review conducted by the American Institutes for Research (AIR), a nonprofit, private research organization, consisted of two main components. The first component was a review of the research program. The second component was a review of the operational application of the remote viewing phenomenon in intelligence gathering. Evaluation of the foreign assessment component of the program was not within the scope of the present effort. Research Evaluation To evaluate the research program, a "blue-ribbon" panel was assembled. The panel included two noted experts in the area of parapsychology: Dr . Jessica Utts, a Professor of Statistics at the University of California/Davis, and Dr. Raymond Hyman, a Professor of Psychology at the University of Oregon. In addition to their extensive credentials, they were selected to represent both sides of the paranormal controversy: Dr. Utts has published articles that view paranormal interpretations positively, while Dr. Hyman was selected to represent a more skeptical position. Both, however, are viewed as fair and open-minded scientists. In American Institutes for Research E-1 '"'" Executive Summary addition to these experts, this panel included two Senior Scientists from AIR; both have recognized methodological expertise, and both had no prior background in parapsychological research. They were included in the review panel to provide an unbiased methodological perspective. In addition, Dr. Lincoln Moses, an Emeritus Professor at Stanford University, provided statistical advice, while Dr. David A. Goslin, President of AIR, served as coordinator of the research effort. Panel members were asked to review all laboratory experiments and meta-analytic reviews conducted as part of the research program; this consisted of approximately 80 separate publications, many of which are summary reports of multiple experiments. In the course of this review, special attention was given to those studies that (a) provided the strongest evidence for the remote viewing phenomenon, and (b) represented new experiments controlling for methodological artifacts identified in earlier reviews. Separate written reviews were prepared by Dr. Utts and Dr. Hyman. They exchanged reviews with other panel members who then tried to reach a consensus. In the typical remote viewing experiment in the laboratory, a remote viewer is asked to visualize a place, location, or object being viewed by a "beacon" or sender. A judge then examines the viewer's report and determines if this report matches the target or, alternatively, a set of decoys. In most recent laboratory experiments reviewed for the present. evaluation, National Geographic photographs provided the target pool. If the viewer's reports match the target, as opposed to the decoys, a hit is said to have occurred. Alternatively, accuracy of a set of remote viewing reports is assessed by rank-ordering the similarity of each remote viewing report to each photograph in the target set (usually five photographs). A better-than- chance score is presumed to represent the occurrence of the paranormal phenomenon of remote viewing, since the remote viewers had not seen the photographs they had described (or did not know which photographs had been randomly selected for a particular remote viewing trial). In evaluating the various laboratory studies conducted to date, the reviewers reached the following conclusions: E-2 American nstitutes or Research Executive Summary ? A statistically significant laboratory effort has been demonstrated in the sense that hits occur more often than chance. ? It is unclear whether the observed effects can unambiguously be attributed to the paranormal ability of the remote viewers as opposed to characteristics of the judges or of the target or some other characteristic of the methods used. Use of the same remote viewers, the same judge, and the same target photographs makes it impossible to identify their independent effects. ? Evidence has not been provided that clearly demonstrates that the causes of hits are due to the operation of paranormal phenomena; the laboratory experiments have not identified the origins or nature of the remote viewing phenomenon, if, indeed, it exists at all. Operational Evaluation The second component of the program involved the use of remote viewing in gathering intelligence information. Here, representatives of various intelligence groups - "end users" of intelligence information - presented targets to remote viewers, who were asked to describe the target. Typically, the remote viewers described the results of their experiences in written reports, which were forwarded to the end users for evaluation and, if warranted, action. To assess the operational value of remote viewing in intelligence gathering, a multifaceted evaluation strategy was employed. First, the relevant research literature was reviewed to identify whether the conditions applying during intelligence gathering would reasonably permit application of the remote viewing paradigm. Second, members of three groups involved in the program were interviewed: (1) end users of the information; (2) the remote viewers providing the reports, and (3) the program manager. Third, feedback information obtained from end user judgments of the accuracy and value of the remote viewing reports was assessed. American Institutes for Research E-3 This multifaceted evaluation effort led to the following conclusions: The conditions under which the remote viewing phenomenon is observed in laboratory settings do not apply in intelligence gathering situations. For example, viewers cannot be provided with feedback and targets may not display the characteristics needed to produce hits. ? The end users indicated that, although some accuracy was observed with regard to broad background characteristics, the remote viewing reports failed to produce the concrete, specific information valued in intelligence gathering. ? The information provided was inconsistent, inaccurate with regard to specifics, and required substantial subjective interpretation. ? In no case had the information provided ever been used to guide intelligence operations. Thus, remote viewing failed to produce actionable intelligence. Conclusions The foregoing observations provide a compelling argument against continuation of the program within the intelligence community. Even though a statistically significant effect has been observed in the laboratory, it remains unclear whether the existence of a paranormal phenomenon, remote viewing, has been demonstrated. The laboratory studies do not provide evidence regarding the origins or nature of the phenomenon, assuming it exists, nor do they address the important methodological issue of inter judge reliability. Further, even if it could be demonstrated unequivocally that a paranormal phenomenon occurs under the conditions present in the laboratory paradigm, these conditions have limited applicability and utility for intelligence gathering operations. For example, the nature of the remote viewing targets are vastly dissimilar, as are the specific tasks required of the remote viewers. Most importantly, the information provided by remote viewing is vague and ambiguous, making it difficult, if not impossible, for the technique to yield information of E-4 American Institutes or Research Executive Summary sufficient quality and accuracy for actionable intelligence. Thus, we conclude that continued use of remote viewing in intelligence gathering operations is not warranted. American Institutes for Research "" Executive Summary American Institutes for Research Chapter One: Background and History Background and History In their continuing quest to improve effectiveness, many organizations have sought techniques that might be used to enhance performance. For the most part, the candidate techniques come from rather traditional lines of inquiry stressing interventions such as selection, training, and performance appraisal. However, some other, more controversial performance enhancement techniques have also been suggested. These techniques range from implicit learning and mental rehearsal to the enhancement of paranormal abilities. In the mid-1980s, at the request of the Army Research Institute, the National Research Council of the National Academy of Sciences established a blue-ribbon panel charged with evaluating the evidence bearing on the effectiveness of a wide variety of techniques for enhancing human performance. This review was conducted under the overall direction of David A. Goslin, then Executive Director of the Commission on Behavioral and Social Sciences and Education (CBASSE), and now President of the American Institutes for Research (AIR). The review panel's report, Enhancing Human Performance: Issues, Theories, and Techniques, was published by the National Academy Press in 1988 and summarized by Swets and Bjork (1990). They noted that although the panel found some support for certain alternative performance enhancement techniques - for example, guided imagery - little or no support was found for the usefulness of many other techniques, such as learning during sleep and remote viewing. Although the findings of the National Research Council (NRC) were predominantly negative with regard to a range of paranormal phenomena, work on remote viewing has continued under the auspices of various government programs. Since 1986, perhaps 50 to 100 additional studies of remote viewing have been conducted. At least some of these studies represent significant attempts to address the methodological problems noted in the review conducted by the NRC panel. At the request of Congress, the Central Intelligence Agency (CIA) is considering assuming responsibility for this the remote viewing program. As part of its decision-making American Institutes for Research 1-1 P" Chapter One: Background and History process, the CIA was asked to evaluate the research conducted since the NRC report. This evaluation was intended to determine: (a) whether this research has any long-term practical value for the intelligence community, and (b) if it does, what changes should be made in methods and approach to enhance the value of remote viewing research. To achieve these goals, the CIA contracted with the American Institutes for Research to supervise and conduct: the evaluation. This report contains the results of our evaluation. Before presenting our results, we begin by presenting a brief overview of the remote viewing phenomenon and a short history of the applied program that involves remote viewing. Remote Viewing Although parapsychological research has a long history, studies of "remote viewing" - also referred to as a form of "anomalous cognition" -- as a unique manifestation of psychic functioning began in the 1970s. In its simplest form, a typical remote viewing study during this early period of investigation consisted of the following: A person, referred to as a "beacon" or "sender," travels to a series of remote sites. The remote viewer, a person who putatively has the parapsychological ability, is asked to describe the locations of the beacon. Typically, these location descriptions include drawings and a verbal description of the location. Subsequently, a judge evaluates this description by rank ordering the set of locations against the descriptions. If the judge finds that the viewer's description most closely matched the actual location of the sender, a hit is said to have occurred. If hits occur more often than chance, or if the assigned ranks are more accurate than a random assignment, one might argue that a psychic phenomenon has been observed: the viewer has described a location not visited during the session. This phenomenon has been studied by various investigators throughout the intervening period, using several variants of this basic paradigm. If certain people (or all people to a greater or lesser extent, as has been proposed by some investigators) possess the ability to see and describe target locations they have not visited, this ability might prove of great value to the intelligence community. As an adjunct: method to gathering intelligence, people who possess this ability could be asked to describe various intelligence targets. This information, especially if considered credible and reliable, American nstitutes or Research Chapter One: Background and History could supplement and enhance more time-consuming and perhaps dangerous methods for collecting data. Although certain (perhaps unwarranted) assumptions, such as the availability of a sender, are implicit in this argument, the possibility of gathering intelligence through this mechanism has provided the major impetus for government interest in remote viewing. Remote viewing was and continues to be a controversial phenomenon. Early research on remote viewing was plagued by a number of statistical and methodological flaws.' One statistical flaw found in early studies of remote viewing, for example, was due to failure to control for the elimination of locations already judged. For example, if there were five targets in the set, judges might lower their rankings for a viewing already judged as a "hit" or ranked first. In other words, all targets did not have an equal probability of being assigned all ranks. Another commonly noted methodological flaw was that cues in the remote viewing paradigm, such as the time needed to drive to various locations, may have allowed viewers to produce hits without using any parapsychological ability. More recent research has attempted to control for many of these problems. New paradigms have been developed where, for example, viewers - in double-blind conditions - are asked to visualize pictures drawn from a target pool consisting of National Geographic photographs. In addition to this experimental work, an applied program of intelligence operations actually using remote viewers has been developed. In the following section, we describe the history of the government's remote viewing program. Program History "Star Gate" is a Defense Intelligence Agency (DIA) program which involved the use of paranormal phenomena, primarily "remote viewing," for intelligence collection. During Star Gate's history, DIA pursued three basic program objectives: "Operations," using remote viewing to collect intelligence against foreign targets; "Research and Development," using laboratory studies to find new ways to improve remote viewing for use in the intelligence 1Many of these problems are described in the National Research Council Report. American Institutes for Research 1-3 Chapter One: Background and History world; and "Foreign Assessment," the analysis of foreign activities to develop or exploit the paranormal for any uses which might affect our national security. Prior to the advent of Star Gate in the early 1990s, the DIA, the Central Intelligence Agency (CIA), and other government organizations conducted various other programs pursuing some or all of these objectives. CIA's program began in 1972, but was discontinued in 1977. DIA's direct involvement began about 1985 and has continued up to the time of this review. During the last twenty years, all government programs involving parapsychology have been viewed as highly controversial and high-risk, and have been subjected to various reviews. In 1995, the CIA declassified its past parapsychology program efforts in order to facilitate a new, external review. In addition, CIA worked with DIA to continue declassification of Star Gate program documents, a process which had already begun at DIA. All relevant CIA and DIA program documents were collected and inventoried. La June of 1995, CIA's Office of Research and Development (ORD) contracted with AIR for this external review, based on our long-standing expertise in carrying out studies relating to behavioral science issues and our neutrality with respect to the subject matter. Evaluation Objectives The CIA asked AIR to address a number of key objectives during the technical review of Star Gate. These included: ? a comprehensive evaluation of the research and development in this area, with a focus on the validity of the technical approach(es) according to acceptable scientific standards ? an evaluation of the overall program utility or usefulness to the government (The CIA believes that the controversial nature of past parapsychology programs within the intelligence community, and the scientific controversy clouding general 1-4 American Instituaes for Research Chapter One: Background and History acceptance of the validity of paranormal phenomena, demand that these two issues of utility and scientific validity be addressed separately.) ? consideration of whether any changes in the operational or research and development activities of the program might bring about improved results if the results were not already optimum ? development of recommendations for the CIA as to appropriate strategies for program activity in the future We were directed to base our findings on the data and information provided as a result of DIA and CIA program efforts, since it was neither possible nor intended that we review the entire field of parapsychological research and its applications. Also, we would not review or evaluate the "Foreign Assessment" component of the program. In the next chapter, we present our methodology for conducting the evaluation. A major component of the evaluation was to commission two nationally-regarded experts to review the program's relevant research studies; their findings are presented in Chapter 3, along with our analysis of areas of agreement and disagreement. In Chapter 4, we present our findings concerning the operational component of the program. Finally, in Chapter 5 we present our conclusions and recommendations. American Institutes or Research Chapter One: Background and History 1-6 American Institu tes for Research Chapter Two: Evaluation Plan Now Chapter Two: Evaluation Plan Evaluation Plan The broad goal of the present effort was to provide a thorough and objective evaluation of the remote viewing program. Because of the multiple components of the program, a multifaceted evaluation plan was devised. As mentioned previously, only the research and intelligence gathering components of the program were considered here. In this section, we describe the general approach used in evaluating these two components of the program, beginning with the research program. Remote Viewing Research The Research Program. The government-sponsored research program had three broad objectives. The first and primary objective was to provide scientifically compelling evidence for the existence of the remote viewing phenomenon. It could be argued that if unambiguous evidence for the existence of the phenomenon cannot be provided, then there is little reason to be concerned with its potential applications. The second objective of the research program was to identify causal mechanisms that might account for or explain the observed (or inferred) phenomenon. This objective of the program is of some importance; an understanding of the origins of a phenomenon provides a basis for developing potential applications. Further, it provides more compelling evidence for the existence of the phenomenon (Cook & Campbell, 1979; James, Muliak, & Brett, 1982). Thus, in conducting a thorough review, an attempt must be made to assess the success of the program in developing an adequate explanation of the phenomenon. The third objective of the research program was to identify techniques or procedures that might enhance the utility of the information provided by remote viewings. For example, how might more specific information be obtained from viewers and what conditions set American institutes for Research Chapter Two: Evaluation Plan - - boundaries on the accuracy of viewings? Research along those lines is of interest primarily because it provides the background necessary for operational applications of the phenomenon. The NRC provided a thorough review of the unclassified remote viewing research through 1986. In this review (summarized in Swets & I3jork, 1990), the nature of the research methods led the reviewers to question whether there was indeed any effect that could clearly be attributed to the operation of paranormal phenomena. Since then, the Principal Investigator, Dr. Edwin May, under formerly classified government contracts, has conducted a number of other studies not previously reviewed. These studies were expressly intended to address many of the criticisms raised in the initial NRC report. Because these studies might provide new evidence for the existence of the remote viewing phenomenon, its causal mechanisms, and its boundary conditions, a new review seemed called for. The Review Panel. With these issues in mind, a blue-ribbon review panel was commissioned, with the intent of ensuring a balanced and objective appraisal of the research. Two of the reviewers were scientists noted for their interest, expertise, and experience in parapsychological research. The first of these two expert reviewers, Dr. Jessica. Utts, a Professor of Statistics at the University of California-Davis, is a nationally recognized scholar who has made major contributions to the development and application of new statistical methods and techniques. Among many other positions and awards, Dr. Utts is an Associate Editor of the Journal of the American Statistical Association (Theory and Methods) and the Statistical Editor of the Journal of the American Society for Psychical Research. She has published several articles on the application of statistical methods to parapsychological research and has direct experience with the remote viewing research program. The second expert reviewer, Dr. Raymond Hyman, is a Professor of Psychology at the University of Oregon. Dr. Hyman has published over 200 articles in professional journals on perception, pattern recognition, creativity, problem solving, and critiques of the paranormal. He served on the original NRC Committee on Techniques for the Enhancement of Human Performance. Dr. Hyman serves as a resource to the media on topics related to the paranormal, and has testified as an expert witness in court cases involving paranormal claims. He is recognized as one of the most important and fair-minded skeptics working in this area. Curriculum Vitae for Dr. Utts and Dr. Hyman are included in Appendix A. 2-2 American Institutes for Research Chapter Two: Evaluation Plan In addition to these two experts, four other scientists were involved in the work of the review panel. Two senior behavioral scientists and experts in research methods at the American Institutes for Research, Dr. Michael Mumford and Dr. Andrew Rose, served both as members of and staff to the panel. Dr. Mumford holds a Ph.D. in Industrial/Organizational Psychology from the University of Georgia. He is a Fellow of the American Psychological Association's Division 5, Measurement, Evaluation, and Statistics. Dr. Rose is a cognitive psychologist with a Ph.D. from the University of Michigan. He has over 22 years of experience in designing and conducting basic and applied behavioral science research. Dr. Rose is Chief Scientist of the Washington Office of AIR. They were to bring to the panel a methodological perspective unbiased by prior work in the area of parapsychology. The third participant was Dr. Lincoln Moses, an Emeritus Professor of Statistics at Stanford University, who participated in the review as a resource with regard to various statistical issues. Finally, Dr. David A. Goslin, President of AIR, participated as both a reviewer and coordinator for the review panel. Research Content. Prior to convening the first meeting of the review panel, the CIA transferred to AIR all reports and documents relevant to the review. We organized and copied these documents. In addition, the Principal Investigator for the program, Dr. Edwin May, was asked to provide two other pieces of information for the panel. First, he was asked to list those studies which he believes provide the strongest evidence bearing on the nature and significance of the remote viewing phenomenon. Second, he was asked to identify all unique studies conducted since the initial NRC report that provide evidence bearing on the nature and significance of the phenomenon. Additionally, he was asked to participate in an interview with members of the review panel following its first meeting to clarify any ambiguities about these studies. The complete list of documents, including notations of the "strongest evidence" set and the "unique" set, is included in Appendix B.' Review Procedures. Remote viewing, like virtually all other parapsychological phenomena, represents one of the most controversial research areas in the social sciences 1One document pertaining to the program remained classified during the period of this review. One of the review panel (Dr. Mumford) examined this document and provided an unclassified synopsis to the review panel. American Institutes for Research 2-3 Chapter Two: Evaluation Plan (e.g., Bem & Honorton, 1994; Hyman, 1994). Therefore, any adequate review of the research program must take this controversy into account in such a way that the review procedures are likely to result in a fair and unbiased assessment of the research. To ensure a fair and comprehensive review, Drs.. Utts and Hyman agreed to examine all program documents. In the course of this review it was agreed that all members of the review panel would carefully consider: ? those studies recommended by the Principal Investigator as providing compelling evidence for the phenomenon, and ? those empirical studies conducted since the NRC review that might provide new evidence about the existence and nature of the phenomenon. The members of the review panel convened at the Palo Alto office of AIR to structure exactly how the review process would be carried out. To ensure that different perspectives on paranormal phenomena would be adequately represented, Drs. Utts and Hyman were asked to prepare independent reports based on their review. In this review, they were to cover four general topics: ? Was there a statistically significant effect? ? Could the observed effect, if any, be attributed to a paranormal phenomenon? ? What mechanisms, if any, might plausibly be used to account for any significant effects and what boundary conditions influence these effects? ? What would the findings obtained in these studies indicate about the characteristics and potential applications of information obtained through the remote viewing process? After they had each completed their reports, they presented the reports to other members of the panel. After studying these reports, all members of the review panel (except AmericanTnstitutes for Research Chapter Two: Evaluation Plan Dr. Moses) participated in a series of conference calls. The primary purpose of these exchanges was to identify the conclusions on which the experts agreed and disagreed. Next, in areas where they disagreed, Drs. Utts and Hyman were asked to discuss the nature of the disagreements, determine why they disagreed, and if possible, attempt to resolve the disagreements. Both the initial reports and the dialogue associated with discussion of any disagreements were made a part of the written record. In fact, Dr. Hyman's opinions on areas of agreement and disagreement are included in his report; in addition to her initial report, Dr. Utts prepared a reply to Dr. Hyman's opinions of agreement and disagreement. This reply, in addition to their original reports, are included in Chapter 3 below. If disagreements could not be resolved through this dialogue, then the other members of the review panel were to consider the remaining issues from a general methodological perspective. Subsequently, they were to provide an addendum to the dialogue indicating which of the two positions being presented seemed to be on firmer ground both substantively and methodologically. This addendum concludes Chapter 3 below. Intelligence Gathering: The Operational Program The Program. In addition to the research component, the program included two operational components. One of those components was "foreign assessment," or analysis of the paranormal research being conducted by other countries. This issue, however, is beyond the scope of the present review. The other component involved the use of remote viewing as a technique for gathering intelligence information. In the early 1970s, the CIA experimented with applications of remote viewing in intelligence gathering. Later in the decade, they abandoned the program. However, other government agencies, including the Department of Defense, used remote viewers to obtain intelligence information. The viewers were tasked with providing answers to questions posed by various intelligence agencies. These operations continued until the Spring of 1995, when the program was suspended. American Institutes for Research 2-5 Chapter Two: Evaluation Plan Although procedures varied somewhat during the history of the program, viewers typically were presented with a request for information about a target of interest to a particular agency. Multiple viewings were then obtained for the target. The results of the viewings then were summarized in a three- or four-page report and sent to the agency that had posed the original question. Starting in 1994, members of the agencies receiving the viewing reports were formally. asked to evaluate their accuracy and value. Any comprehensive evaluation of the remote viewing program must consider how viewings were used by the intelligence community. One might demonstrate the existence of a statistically significant paranormal phenomenon in experiments conducted in the laboratory; however, the phenomenon could prove to be of limited operational value either because it does not occur consistently outside the laboratory setting or because the kind of information provided is of limited value to the intelligence community. General Evaluation Procedures. No one piece of evidence provides unequivocal support for the usefulness of a program. Instead, a more accurate and comprehensive picture can be obtained by considering multiple sources of evidence (Messick, 1989). Three basic sources of information were used in evaluation of the intelligence gathering component: ? prior research studies ? interviews with program participants ? analyses of user feedback Prior Research Studies. As noted above, one aspect of the laboratory research program was to identify those conditions that set bounds on the accuracy and success of the remote viewing process. Thus, one way to analytically evaluate potential applications in intelligence gathering is to enumerate the conditions under which viewers were assigned tasks and then examine the characteristics of the remote viewing paradigm as studied through experimentation in the laboratory. The conditions under which operational tasks occur - that is, the requirements imposed by intelligence gathering -- could then provide an assessment of the applicability of the remote viewing process. American Institutes or Research Chapter Two: Evaluation Plan Interviews. As part of the Star Gate program, the services of remote viewers were used to support operational activities in the intelligence community. This operational history provides an additional basis for evaluating the Star Gate program; ultimately, if the program is to be of any real value, it must be capable of serving the needs of the intelligence community. By examining how the remote viewing services have been used, it becomes possible to draw some initial, tentative conclusions about the potential value of the Star Gate program. Below, we describe how information bearing on intelligence applications of the remote viewing phenomenon was gathered. Later, in Chapter 4, we describe the results of this information-gathering activity and draw some conclusions from the information we obtained. Although a variety of techniques might be used to accrue retrospective information (questionnaires, interviews, diaries, etc.), the project team decided that structured interviews examining issues relevant to the various participants would provide the most appropriate strategy. Accordingly, structured interviews were developed for three participant groups in intelligence operations: ? end-users: representatives from agencies requesting information from remote viewers ? the Program Manager ? the remote viewers Another key issue to be considered in an interview procedure is the nature of the people to be interviewed. Although end-users, program managers, and viewers represent the major participants, many different individuals have been involved in intelligence applications of remote viewing over the course of the last twenty years. Nevertheless, it was decided to interview only those persons who were involved in the program at the time of its suspension in the Spring of 1995. This decision was based on the need for accurate, current information that had not been distorted by time and could be corroborated by existing documentation and follow-up interviews. American Institutes for Research Chapter Two: Evaluation Plan Information about operational applications was gathered in a series of interviews conducted during July and August of 1995. We interviewed seven representatives of end-user groups, three remote viewers, and the incumbent Program Manager. With regard to the data collection procedures that we employed, a number of points should be borne in mind. First, members of the groups we interviewed could only speak to recent operations. Although it would have been desirable to interview people involved in earlier operations, for example during the 1970s, the problems associated with the passage of time, including forgetting and the difficulties involved in verifying information, effectively precluded this approach. Accordingly, the interviews focused on current operations. Second, it should be noted that the end-user representatives represented a :range of current concerns in the intelligence community. The relevant user groups were involved in operations ranging from counterintelligence and drug interdiction to search and rescue operations. This diversity permitted operational merits to be assessed for a number of different contexts. The interviews were conducted by one of the two panel members from AIR. A retired intelligence officer took notes during the interviews. A representative of the CIA attended interviews as necessary to describe the reasons the interviews were being conducted and to address any security concerns. Each interview was conducted using a standard protocol. Different protocols were developed for members of the three groups because they had somewhat different perspectives on current operations. Appendix C presents the instructions given to the interviewer. This Appendix also lists the interview questions presented to users, viewers, and the program manager. User interviews were conducted in the offices of the client organization; interviews with the program manager and the viewers were conducted at the Washington Office of AIR. The interviews were one to two hours long. A total of 12 to 16 questions were asked in the interviews. We developed the questions presented in each interview as follows: Initially, the literature on remote viewing and available information bearing on operations within the intelligence community were reviewed by AIR scientists. This review was used to formulate 2-8 American Institutes or Research I Chapter Two: Evaluation Plan an initial set of interview questions. Subsequently, these candidate questions were presented to a panel of three psychologists at AIR. In addition, review panel members were asked to review these candidate questions to insure they were not leading and covered the issues that were relevant to the particular group under consideration. With regard to operational users, four types of questions were asked. These four types of questions examined the background and nature of the tasks presented to the remote viewers, the nature and accuracy of the information resulting from the viewings, operational use of this information, and the utility of the resulting information. The remote viewers were asked a somewhat different set of questions. The four types of questions presented to them examined recruitment, selection, and development; the procedures used to generate viewings; the conditions that influenced the nature and success of viewings; and the organizational factors that influenced program operations. The Program Manager was not asked about the viewing process. Instead, questions presented to the program manager primarily focused on broader organizational issues. The four types of managerial questions focused on the manager's background, client recruitment, factors influencing successes and failures, and needs for effective program management. The interview questions presented in each protocol were asked in order, as specified in Appendix C. Typically these interviews began by asking for objective background information. Questions examining broader evaluative issues were asked at the end of the interview. The AIR scientist conducting the interviews produced reports for each individual interview. They also are contained in Appendix C. Analyses of User Feedback. In addition to the qualitative data provided by the interviews, some quantitative information was available. For all of the operational tasks conducted during 1994, representatives from the requesting agencies were asked to provide two summary judgments: one with respect to the accuracy of the remote viewing, and the second of the actual or potential value of the information provided. These data - the accuracy and value evaluations obtained for viewings as program feedback from the users - were analyzed and summarized in a report prepared prior to the current evaluation. A copy American Institutes for Research 2-9 Chapter Two: Evaluation Plan , of this report is provided in Appendix D. Although these judgments have been routinely collected for only a relatively short period of time, they provided an important additional source of evaluative information. This information was of some value as a supp'.lement to interviews in part because it was collected prior to the start of the current review, and in part because it reflects user assessments of the resulting information. We present the findings flowing from this multifaceted evaluation of the operational component of the program in Chapter 4 of this report. In that section, we first present the findings emerging from prior research and the interviews and then consider the results obtained from the more quantitative evaluations. Prior to turning to this evaluation of operations, however, we first present the findings from review of the basic research, examining evidence for the existence and nature of the remote viewing phenomenon. American Institutes for Research Chapter Three: Research Reviews Review 1 (Dr. Jessica Utts) Review 2 (Dr. Ray Hyman) The Reply (Dr. Jessica Utts) Points of Agreement and Disagreement Conclusions from the Expert Reviews American Institutes for Research ~'" Research Reviews In this section, we present the conclusions drawn by the two experts after reviewing the research studies bearing on remote viewing. We begin by presenting the review of Dr. Jessica Utts. Subsequently, a rejoinder is provided by Dr. Raymond Hyman. Finally, Dr. Utts presents a reply to Dr. Hyman. The major points of agreement and disagreement are noted in the final section, along with our conclusions. In conducting their reviews, both Dr. Hyman and Dr. Utts focused on the remote viewing research. However, additional material is provided as indicated by the need to clarify certain points being made. Furthermore, both reviewers provided unusually comprehensive reviews considering not only classified program research, but also a number of earlier studies having direct bearing on the nature and significance of the phenomenon. American Institutes for Research Review 1 An Assessment of the Evidence for Psychic Functioning Dr. Jessica Utts Division of Statistics, University of California, Davis September 1, 1995 ABSTRACT Research on psychic functioning, conducted over a two decade period, is examined to determine whether or not the phenomenon has been scientifically established. A secondary question is whether or not it is useful for government purposes. The primary work examined in this report was government sponsored research conducted at Stanford Research Institute, later known as SRI International, and at Science Applications International Corporation, known as SAIC. Using the standards applied to any other area of science, it is concluded that psychic functioning has been well established. The statistical results of the studies examined are far beyond what is expected by chance. Arguments that these results could be due to methodological flaws in the experiments are soundly refuted. Effects of similar magnitude to those found in government-sponsored research at SRI and SAIC have been replicated at a number of laboratories across the world. Such consistency cannot be readily explained by claims of flaws or fraud. The magnitude of psychic functioning exhibited appears to be in the range between what social scientists call a small and medium effect. That means that it is reliable enough to be replicated in properly conducted experiments, with sufficient trials to achieve the long-run statistical results needed for replicability. A number of other patterns have been found, suggestive of how to conduct more productive experiments and applied psychic functioning. For instance, it doesn't appear that a sender is 3-2 American institutes for Research Chapter Three: Research Reviews needed. Precognition, in which the answer is known to no one until a future time, appears to work quite well. Recent experiments suggest that if there is a psychic sense then it works much like our other five senses, by detecting change. Given that physicists are currently grappling with an understanding of time, it may be that a psychic sense exists that scans the future for major change, much as our eyes scan the environment for visual change or our ears allow us to respond to sudden changes in sound. It is recommended that future experiments focus on understanding how this phenomenon works, and on how to make it as useful as possible. There is little benefit to continuing experiments designed to offer proof, since there is little more to be offered to anyone who does not accept the current collection of data. 1. INTRODUCTION The purpose of this report is to examine a body of evidence collected over the past few decades in an attempt to determine whether or not psychic functioning is possible. Secondary questions include whether or not such functioning can be used productively for government purposes, and whether or not the research to date provides any explanation for how it works. There is no reason to treat this area differently from any other area of science that relies on statistical methods. Any discussion based on belief should be limited to questions that are not data-driven, such as whether or not there are any methodological problems that could substantially alter the results. It is too often the case that people on both sides of the question debate the existence of psychic functioning on the basis of their personal belief systems rather than on an examination of the scientific data. One objective of this report is to provide a brief overview of recent data as well as the scientific tools necessary for a careful reader to reach his or her own conclusions based on that data. The tools consist of a rudimentary overview of how statistical evidence is typically evaluated, and a listing of methodological concerns particular to experiments of this type. American Institutes for Research .40 Government-sponsored research in psychic functioning dates back to the early 1970s when a program was initiated at what was then the Stanford Research Institute, now called SRI International. That program was in existence until 1989. The following year, government sponsorship moved to a program at Science Applications International Corporation (SAIL) under the direction of Dr. Edwin May, who had been employed in the SRI program since the mid 1970s and had been Project Director from 1986 until the close of the program. This report will focus most closely on the most recent work, done by SAIC. Section 2 describes the basic statistical and methodological issues required to understand this work; Section 3 discusses the program at SRI; Section 4 covers the SAIC work (with some of the details in an Appendix); Section 5 is concerned with external validation by exploring related results from other laboratories; Section 6 includes a discussion of the usefulness of this capability for government purposes and Section 7 provides conclusions and recommendations. 2. SCIENCE NOTES 2.1 DEFINITIONS AND RESEARCH PROCEDURES There are two basic types of functioning that are generally considered under the broad heading of psychic or paranormal abilities. These are classically known as extrasensory perception (ESP), in which one acquires information through unexplainable means and psychokinesis, in which one physically manipulates the environment through unknown means. The SAIC laboratory uses more neutral terminology for these abilities; they refer to ESP as anomalous cognition (AC) and to psychokinesis as anomalous perturbation (AP). The vast majority of work at both SRI and SAIC investigated anomalous cognition rather than anomalous perturbation, although there was some work done on the latter. Anomalous cognition is further divided into categories based on the apparent source of the information. If it appears to come from another person, the ability is called telepathy, if it appears to come in real time but not from another person it is called clairvoyance and if the information could have only been obtained by knowledge of the future, it is called precognition. 3-4 American Institutes for Research Chapter Three: Research Reviews It is possible to identify apparent precognition by asking someone to describe something for which the correct answer isn't known until later in time. It is more difficult to rule out precognition in experiments attempting to test telepathy or clairvoyance, since it is almost impossible to be sure that subjects in such experiments never see the correct answer at some point in the future. These distinctions are important in the quest to identify an explanation for anomalous cognition, but do not bear on the existence issue. The vast majority of anomalous cognition experiments at both SRI and SAIC used a technique known as remote viewing. In these experiments, a viewer attempts to draw or describe (or both) a target location, photograph, object or short video segment. All known channels for receiving the information are blocked. Sometimes the viewer is assisted by a monitor who asks the viewer questions; of course in such cases the monitor is blind to the answer as well. Sometimes a sender is looking at the target during the session, but sometimes there is no sender. In most cases the viewer eventually receives feedback in which he or she learns the correct answer, thus making it difficult to rule out precognition as the explanation for positive results, whether or not there was a sender. Most anomalous cognition experiments at SRI and SAIC were of the free-response type, in which viewers were simply asked to describe the target. In contrast, a forced-choice experiment is one in which there are a small number of known choices from which the viewer must choose. The latter may be easier to evaluate statistically but they have been traditionally less successful than free-response experiments. Some of the work done at SAIC addresses potential explanations for why that might be the case. 2.2 STATISTICAL ISSUES AND DEFINITIONS Few human capabilities are perfectly replicable on demand. For example, even the best hitters in the major baseball leagues cannot hit on demand. Nor can we predict when someone will hit or when they will score a home run. In fact, we cannot even predict whether or not a home run will occur in a particular game. That does not mean that home runs don't exist. American-institutes for Research 3-5 Review 1: Dr. Jessica Utts Scientific evidence in the statistical realm is based on replication of the same average performance or relationship over the long run. We would not expect a fair coin to result in five heads and five tails over each set of ten tosses, but we can expect the proportion of heads and tails to settle down to about one half over a very long series of tosses. Similarly, a good baseball hitter will not hit the ball exactly the same proportion of times in each game but should be relatively consistent over the long run. The same should be true of psychic functioning. Even if there truly is an effect, it may never be replicable on demand in the short run even if we understand how it works. However, over the long run in well controlled laboratory experiments we should see a consistent level of functioning, above that expected by chance. The anticipated level of.functioning may vary based on the individual players and the conditions, just as it does in baseball, but given players of similar ability tested under similar conditions the results should be replicable over the long run. In this report we will show that replicability in that sense has been achieved. 2.2.1 P-VALUES AND COMPARISON WITH CHANCE. In any area of science, evidence based on statistics comes from comparing what actually happened to what should have happened by chance. For instance, without any special interventions about 51 percent of births in the United States result in boys. Suppose someone claimed to have a method that enabled one to increase the chances of having a baby of the desired sex. We could study their method by comparing how often births resulted in a boy when that was the intended outcome. If that percentage was higher than the chance percentage of 51 percent over the long run, then the claim would have been supported by statistical evidence. Statisticians have developed numerical methods for comparing results to what is expected by chance. Upon observing the results of an experiment, the p-value is the answer to the following question: If chance alone is responsible for the results, how likely would we be to observe results this strong or stronger? If the answer to that question, i.e. the p-value is very small, then most researchers are willing to rule out chance as an explanation. In fact it is commonly accepted practice to say that if the p-value is 5 percent (0.05) or less, then we can rule out chance as an explanation. In such cases, the results are said to be statistically significant. Obviously the smaller the p-value, the more convincingly chance can be ruled out. American Institutes for Research 3- Chapter Three: Research Reviews Notice that when chance alone is at work, we erroneously find a statistically significant result about 5 percent of the time. For this reason and others, most reasonable scientists require replication of non-chance results before they are convinced that chance can be ruled out. 2.2.2 REPLICATION AND EFFECT SIZES: In the past few decades scientists have realized that true replication of experimental results should focus on the magnitude of the effect, or the effect size rather than on replication of the p-value. This is because the latter is heavily dependent on the size of the study. In a very large study, it will take only a small magnitude effect to convincingly rule out chance. In a very small study, it would take a huge effect to convincingly rule out chance. In our hypothetical sex-determination experiment, suppose 70 out of 100 births designed to be boys actually resulted in boys, for a rate of 70 percent instead of the 51 percent expected by chance. The experiment would have a p-value of 0.0001, quite convincingly ruling out chance. Now suppose someone attempted to replicate the experiment with only ten births and found 7 boys, i.e also 70 percent. The smaller experiment would have a p-value of 0.19, and would not be statistically significant. If we were simply to focus on that issue, the result would appear to be a failure to replicate the original result, even though it achieved exactly the same 70 percent boys! In only ten births it would require 90 percent of them to be boys before chance could be ruled out. Yet the 70 percent rate is a more exact replication of the result than the 90 percent. Therefore, while p-values should be used to assess the overall evidence for a phenomenon, they should not be used to define whether or not a replication of an experimental result was "successful." Instead, a successful replication should be one that achieves an effect that is within expected statistical variability of the original result, or that achieves an even stronger effect for explainable reasons. A number of different effect size measures are in use in the social sciences, but in this report we will focus on the one used most often in remote viewing at SRI and SAIC. Because the definition is somewhat technical it is given in Appendix 1. An intuitive explanation will be given in the next subsection. Here, we note that an effect size of 0 is consistent with chance, and social scientists have, by convention, declared an effect size of 0.2 as small, 0.5 as American Institutes or esearc medium and 0.8 as large. A medium effect size is supposed to be visible to the naked e;,e of a careful observer, while a large effect size is supposed to be evident to any observer. 2.2.3 RANDOMNESS AND RANK-ORDER JUDGING. At the heart of any statistical method is a definition of what should happen "randomly" or "by chance." Without a random mechanism, there can be no statistical evaluation. There is nothing random about the responses generated in anomalous cognition experiments; in other words, there is no way to define what they would look like "by chance." Therefore, the random mechanism in these experiments must be in the choice of the target. In that way, we can compare the response to the target and answer the question: "If chance alone is at work, what is the probability that a target would be chosen that matches this response as well as or better than does the actual target?" In order to accomplish this purpose, a properly conducted experiment uses a set of targets defined in advance. The target for each remote viewing is then selected randomly, in such a way that the probability of getting each possible target is known. The SAIC remote viewing experiments and all but the early ones at SRI used a statistical evaluation method known as rank-order judging. After the completion of a remote viewing, a judge who is blind to the true target (called a blind judge) is shown the response and five potential targets, one of which is the correct answer and the other four of which are "decoys." Before the experiment is conducted, each of those five choices must have had an equal chance of being selected as the actual target. The judge is asked to assign a rank to each of the possible targets, where a rank of one means it matches the response most closely, and a rank of five means it matches the least. The rank of the correct target is the numerical score for that remote viewing. By chance alone the actual target would receive each of the five ranks with equal likelihood, since despite what the response said, the target matching it best would have the same chance of selection as the one matching it second best, and so on. The average rank by chance would be three. Evidence for anomalous cognition occurs when the average rank over a series of 3-8 American Institutes for Research Chapter Three: Research Reviews trials is significantly lower than three. (Notice that a rank of one is the best possible score for each viewing.) This scoring method is conservative in the sense that it gives no extra credit for an excellent match. A response that describes the target almost perfectly will achieve the same rank of one as a response that contains only enough information to pick the target as the best choice out of the five possible choices. One advantage of this method is that it is still valid even if the viewer knows the set of possible targets. The probability of a first place match by chance would still be only one in five. This is important because the later SRI and many of the SAIC experiments used the same large set of National Geographic photographs as targets. Therefore, the experienced viewers would eventually become familiar with the range of possibilities since they were usually shown the answer at the end of each remote viewing session. For technical reasons explained in Appendix 1, the effect size for a series of remote viewings using rank-order judging with five choices is (3.0 - average rank)/12. Therefore, small, medium and large effect sizes (0.2, 0.5 and 0.8) correspond to average ranks of 2.72, 2.29, and 1.87, respectively. Notice that the largest effect size possible using this method is 1.4, which would result if every remote viewing achieved a first place ranking. 2.3 METHODOLOGICAL ISSUES One of the challenges in designing a good experiment in any area of science is to close the loopholes that would allow explanations other than the intended one to account for the results. There are a number of places in remote viewing experiment where information could be conveyed by normal means if proper precautions are not taken. The early SRI experiments suffered from some of those problems, but the later SRI experiments and the SAIC work were done with reasonable methodological rigor, with some exceptions noted in the detailed descriptions of the SAIC experiments in Appendix 2. American Institutes or Research Review 1: Dr. Jessica Utts The following list of methodological issues shows the variety of concerns that must be addressed. It should be obvious that a well designed experiment requires careful thought and planning: ? No one who has knowledge of the specific target should have any contact with the viewer until after the response has been safely secured. ? No one who has knowledge of the specific target or even of whether or not the session was successful should have any contact with the judge until after that task has been completed. No one who has knowledge of the specific target should have access to the. response until after the judging has been completed. ? Targets and decoys used in judging should be selected using a well tested randomization device. ? Duplicate sets of targets photographs should be used, one during the experiment and one during the judging, so that no cues (like fingerprints) can be inserted onto the target that would help the judge recognize it. ? The criterion for stopping an experiment should be defined in advance so that it is not called to a halt when the results just happen to be favorable. Generally, that means specifying the number of trials in advance, but some statistical procedures require or allow other stopping rules. The important point is that the rule be defined in advance in such a way that there is no ambiguity about when to stop. Reasons, if any, for excluding data must be defined in advance and followed consistently, and should not be dependent on the data. For example, a rule specifying that a trial could be aborted if the viewer felt ill would be legitimate, but only if the trial was aborted before anyone involved in that decision knew the correct target. 3-10 American nsiitutes for Research Chapter Three: Research Reviews Statistical analyses to be used must be planned in advance of collecting the data so that a method most favorable to the data isn't selected post hoc. If multiple methods of analysis are used, the corresponding conclusions must recognize that fact. 2.4 PRIMA FACIE EVIDENCE According to Webster's Dictionary, in law, prima facie evidence is "evidence having such a degree of probability that it must prevail unless the contrary be proved." There are a few examples of applied, non-laboratory remote viewings provided to the review team that would seem to meet that criterion for evidence. These are examples in which the sponsor or another government client asked for a single remote viewing of a site, known to the requester in real time or in the future, and the viewer provided details far beyond what could be taken as a reasonable guess. Two such examples are given by May (1995) in which it appears that the results were so striking that they far exceed the phenomenon as observed in the laboratory. Using a post hoc analysis, Dr. May concluded that in one of the cases the remote viewer was able to describe a microwave generator with 80 percent accuracy, and that of what he said almost 70 percent of it was reliable. Laboratory remote viewings rarely show that level of correspondence. Notice that standard statistical methods cannot be used in these cases because there is no standard for probabilistic comparison. But evidence gained from applied remote viewing cannot be dismissed as inconsequential just because we cannot assign specific probabilities to the results. It is most important to ascertain whether or not the information was achievable in other standard ways. In Section 3 an example is given in which a remote viewer allegedly gave codewords from a secret facility that he should not have even known existed. Suppose the sponsors could be absolutely certain that the viewer could not have known about those codewords through normal means. Then even if we can't assign an exact probability to the fact that he guessed them correctly, we can agree that it would be very small. That would seem to constitute prima facie evidence unless an alternative explanation could be found. Similarly, the viewer who described the microwave generator allegedly knew only that the target was a technical site in the United States. Yet, he drew and described the microwave generator, including its function, its approximate size, how it was housed and that it had "a beam divergence angle of 30 degrees" (May, 1995, p. 15). American Institutes for Research Anecdotal reports of psychic functioning suffer from a similar problem in terms of their usefulness as proof. They have the additional difficulty that the "response" isn't even well- defined in advance, unlike in applied remote viewing where the viewer provides a fixed set of information on request. For instance, if a few people each night happen to dream of plane crashes, then some will obviously do so on the night before a major plane crash. Those individuals may interpret the coincidental timing as meaningful. This is undoubtedly the reason many people think the reality of psychic functioning is a matter of belief rather than science, since they are more familiar with the provocative anecdotes than with the laboratory evidence. 3. THE SRI ERA 3.1 EARLY OPERATIONAL SUCCESSES AND EVALUATION According to Puthoff and Targ (1975) the scientific research endeavor at SRI may never have been supported had it not been for three apparent operational successes in the early days of the program. These are detailed by Puthoff and Targ (1975), although the level of the matches is not clearly delineated. One of the apparent successes concerned the "West Virginia Site" in which two remote viewers purportedly identified an underground secret facility. One of them apparently named codewords and personnel in this facility accurately enough that it set off a security investigation to determine how that information could have been leaked. Based only on the coordinates of the site, the viewer first described the above ground terrain, then proceeded to describe details of the hidden underground site. The same viewer then claimed that he could describe a similar Communist Bloc site and proceeded to do so for a site in the Urals. According to Puthoff and Targ "the two reports for the West Virginia Site, and the report for the Urals Site were verified by personnel in the sponsor organization as being substantially correct (p. 8)." American Institutes for Research Chapter Three: Research Reviews The third reported operational success concerned an accurate description of a large crane and other information at a site in Semipalatinsk, USSR. Again the viewer was provided with only the geographic coordinates of the site and was asked to describe what was there. Although some of the information in these examples was verified to be highly accurate, the evaluation of operational work remains difficult, in part because there is no chance baseline for comparison (as there is in controlled experiments) and in part because of differing expectations of different evaluators. For example, a government official who reviewed the Semipalatinsk work concluded that there was no way the remote viewer could have drawn the large gantry crane unless "he actually saw it through remote viewing, or he was informed of what to draw by someone knowledgeable of [the site]." Yet that same analyst concluded that "the remote viewing of [the site] by subject Si proved to be unsuccessful" because "the only positive evidence of the rail-mounted gantry crane was far outweighed by the large amount of negative evidence noted in the body of this analysis." In other words, the analyst had the expectation that in order to be "successful" a remote viewing should contain accurate information only. Another problem with evaluating this operational work is that there is no way to know with certainty that the subject did not speak with someone who had knowledge of the site, however unlikely that possibility may appear. Finally, we do not know to what degree the results in the reports were selectively chosen because they were correct. These problems can all be avoided with well designed controlled experiments. 3.2 THE EARLY SCIENTIFIC EFFORT AT SRI During 1974 and early 1975 a number of controlled experiments were conducted to see if various types of target material could be successfully described with remote viewing. The results reported by Puthoff and Targ (1975) indicated success with a wide range of material, from "technical" targets like a Xerox machine to natural settings, like a swimming pool. But these and some of the subsequent experiments were criticized on statistical and methodological grounds; we briefly describe one of the experiments and criticisms of it to show the kinds of problems that existed in the early scientific effort. American Institutes for Research 3-13 The largest series during the 1973 to 1975 time period involved remote viewing of natural sites. Sites were randomly selected for each trial from a set of 100 possibilities. They were selected "without replacement," meaning that sites were not reused once they had been selected. The series included eight viewers, including two supplied by the sponsor. Many of the descriptions showed a high degree of subjective correspondence, and the overall statistical results were quite striking for most of the viewers. Critics attacked these experiments on a number of issues, including the selection. of sites without replacement and the statistical scoring method used. The results were scored by having a blind judge attempt to match the target material with the transcripts of the responses. A large fraction of the matches were successful. But critics noted that some successful matching could be attained just from cues contained in the transcripts of the material, like when a subject mentioned in one session what the target had been in the previous session. Because sites were selected without replacement, knowing what the answer was on one day would exclude that target site from being the answer on any other day. There was no way to determine the extent to which these problems influence the results. The criticisms of these and subsequent experiments, while perhaps unwelcome at the time, have resulted in substantially improved methodology in these experiments. 3.3 AN OVERALL ANALYSIS OF THE SRI EXPERIMENTS: 1973-1988 In 1988 an analysis was made of all of the experiments conducted at SRI from 1973 until that time (May et al, 1988). The analysis was based on all 154 experiments conducted during that era, consisting of over 26,000 individual trials. Of those, almost 20,000 were of the forced choice type and just over a thousand were laboratory remote viewings. There were a total of 227 subjects in all experiments. The statistical results were so overwhelming that results that extreme or more so would occur only about once in every 1020 such instances if chance alone is the explanation (i.e., the p- value was less than 10.20). Obviously some explanation other than chance must be found. Psychic functioning may not be the only possibility, especially since some of the earlier work contained methodological problems. However, the fact that the same level of functioning continued to hold in the later experiments, which did not contain those flaws, ]ends support to American Institutes or Research Chapter Three: Research Reviews the idea that the methodological problems cannot account for the results. In fact, there was a talented group of subjects (labeled GI in that report) for whom the effects were stronger than for the group at large. According to Dr. May, the majority of experiments with that group were conducted later in the program, when the methodology had been substantially improved. In addition to the statistical results, a number of other questions and patterns were examined. A summary of the results revealed the following: 1. "Free response" remote viewing, in which subjects describe a target, was much more successful than "forced choice" experiments, in which subjects were asked to choose from a small set of possibilities. 2. There was a group of six selected individuals whose performance far exceeded that of unselected subjects. The fact that these same selected individuals consistently performed better than others under a variety of protocols provides a type of replicability that helps substantiate the validity of the results. If methodological problems were responsible for the results, they should not have affected this group differently from others. 3. Mass-screening efforts found that about one percent of those who volunteered to be tested were consistently successful at remote viewing. This indicates that remote viewing is an ability that differs across individuals, much like athletic ability or musical talent. (Results of mass screenings were not included in the formal analysis because the conditions were not well controlled, but the subsequent data from subjects found during mass-screening were included.) 4. Neither practice nor a variety of training techniques consistently worked to improve remote viewing ability. It appears that it is easier to find than to train good remote viewers. 5. It is not clear whether or not feedback (showing the subject the right answer) is necessary, but it does appear to provide a psychological boost that may increase performance. American Institutes for Research Review 1: Dr. Jessica Utts 6. Distance between the target and the subject does not seem to impact the quality of the remote viewing. 7. Electromagnetic shielding does not appear to inhibit performance. 8. There is compelling evidence that precognition, in which the target is selected after the subject has given the description, is also successful. 9. There is no evidence to support anomalous perturbation (psychokinesis), i.e. physical interaction with the environment by psychic means. 3.4 CONSISTENCY WITH OTHER LABORATORIES IN THE SAME ERA One of the hallmarks of a real phenomenon is that its magnitude is replicable by various researchers working under similar conditions. The results of the overall SRI analysis are consistent with results of similar experiments in other laboratories. For instance, an overview of forced choice precognition experiments (Honorton and Ferrari, 1989) found an average "effect size" per experimenter of 0.033, whereas all forced choice experiments at SRI resulted in a similar effect size of .052. The comparison is not ideal since the SRI forced choice experiments were not necessarily precognitive and they used different types of target material than the standard card-guessing experiments. Methodologically sound remote viewing has not been undertaken at other laboratories, but a similar regime called the ganzfeld (described in more detail in Section 5) has shown to be similarly successful. The largest collection of ganzfeld experiments was conducted from 1983 to 1989 at the Psychophysical Research Laboratories in Princeton, NJ. Those experiments were also reported by separating r.ovices from experienced subjects. The overall effect size for novice remote viewing at SRI was 0.164, while the effect size for novices in the ganzfeld at PRL was a very similar 0.17. For experienced remote viewers at SRI the overall effect size was 0.385; for experienced viewers in the ganzfeld experiments it was 0.35. These consistent results across laboratories help refute the idea that the successful experiments at any one lab are the result of fraud, sloppy protocols or some methodological problem and also provide an indication of what can be expected in future experiments. 3-16 American institutes for Research Chapter Three: Research Reviews 4. THE SAIL ERA 4.1 AN OVERVIEW The review team decided to focus more intensively on the experiments conducted at Science Applications International Corporation (SAIC), because they provide a manageable yet varied set to examine in detail. They were guided by a Scientific Oversight Committee consisting of experts in a variety of disciplines, including a winner of the Nobel Prize in Physics, internationally known professors of statistics, psychology, neuroscience and astronomy and a medical doctor who is a retired U.S. Army Major General. Further, we have access to the details for the full set of SAIC experiments, unlike for the set conducted at SRI. Whatever details may be missing from the written reports are obtainable from the principal investigator, Dr. Edwin May, to whom we have been given unlimited access. In a memorandum dated July 25, 1995, Dr. Edwin May listed the set of experiments conducted by SAIC. There were ten experiments, all designed to answer questions about psychic functioning raised by the work at SRI and other laboratories, rather than just to provide additional proof of its existence. Some of the experiments were of a similar fonnat to the remote viewing experiments conducted at SRI and we can examine those to see whether or not they replicated the SRI results. We will also examine what new knowledge can be gained from the results of the SAIC work. 4.2 THE TEN EXPERIMENTS Of the ten experiments done at SAIC, six of them involved remote viewing and four did not. Rather than list the details in the body of this report, Appendix 2 gives a brief description of the experiments. What follows is a discussion of the methodology and results for the experiments as a whole. Because of the fundamental differences between remote viewing and the other types of experiments, we discuss them separately. In the memorandum of 25 July 1995, Dr. May provided the review team with details of the ten experiments, including a short title, number of trials, effect size and overall p-value for American Institutes or esearc Review 1: Dr. Jessica Utts each one. His list was in time sequence. It is reproduced in Table 1, using his numbering system, with the experiments categorized by type, then sequentially within type. The effect size estimates are based on a limited number of trials, so they are augmented with an interval to show the probable range of the true effect (e.g., .124+.071 indicates a range from .053 to .195). Remember that an effect size of 0 represents chance, while a positive effect size indicates positive results. TABLE 1: SAIC EXPERIMENTS LISTED BY DR. EDWIN MAY Expr Title Trials Effect Size p-value Remote Viewing Experiments 1 Target dependencies 200 .124?.071 0.040 4 AC with binary coding 40 -.067?.158 0.664 5 AC lucid dreams, base 24 .088?.204 0.333 6 AC lucid dreams, pilot 21 .368?.218 0.046 9 ERD AC Behavior 70 .303?.120 0.006 10 Entropy II 90 .550?.105 9.1X10'$ Other Experiments 2 AC of binary targets 300 .123?.058 0.017 3 MEG Replication 12,000s MCE MCE 7 Remote observation 48 .361?.144 0.006 8 ERD EEG investigation 7,000s MCE MCE American Tn-Tt tutes for Research Chapter Three: Research Reviews 4.3 ASSESSING THE REMOTE VIEWING EXPERIMENTS BY HOMOGENEOUS SETS OF SESSIONS While Table I provides an overall assessment of the results of each experiment, it does so at the expense of information about variability among viewers and types of targets. In terms of understanding the phenomenon, it is important to break the results down into units that are as homogeneous as possible in terms of procedure, individual viewer and type of target. This is also important in order to assess the impact of any potential methodological problems. For example, in one pilot experiment (E6, AC in Lucid Dreams) viewers were permitted to take the targets home with them in sealed envelopes. Table 2 presents the effect size results at the most homogeneous level possible based on the information provided. For descriptions of the experiments, refer to Appendix 2. Overall effect sizes for each viewer and total effect sizes for each experiment are weighted according to the number of trials, so each trial receives equal weight. American Institutes for Research 3-19 Review 1: Dr. Jessica Utts TABLE 2: INDIVIDUAL EFFECT SIZES Experiment Experiment Remote Viewers Viewer 009 131 372 389 518 Unknown Total /Other Static Targets (National Geographic) El: Static .424 -.071 .424 .177 .283 n.a. .247 E9 .432 n.a. .354 177 n.a. n.a. .303 E10: Static .566 n.a. .801 -.071 .778 n.a. .550 E5 (Note 1) n.a. n.a. n.a. n.a. n.a. 088 .088 E6 (Note 2) n.a. n.a. n.a. n.a. n.a. .370 .370 E4 (Note 3) -.112 n.a. 0 n.a. .112 -.559 -.067 Dynamic Targets (Video Film Clips) El: Dynamic 0 .354 -.283 0 -.071 n.a._ .000 E10: Dynamic Overall .919 352 n.a. .141 .754 .340 0 ]_.090 .424 .271 n.a. n.a_ .550 ~ Notes: 1. Experiment 5 did not include any expert viewers. 2. Experiment 6 included 4 expert viewers but separate results were not provided. 3. Experiment 4 used a specially designed target set and only 4 choices in judging. 4.4 CONSISTENCY AND REPLICABILITY OF THE REMOTE VIEWING RESULTS One of the most important hallmarks of science is replicability. A phenomenon with statistical variability, whether it is scoring home runs in baseball, curing a disease with chemotherapy, or observing psychic functioning, should exhibit about the same level of es ?o r Research 3-20 American Instrtcit-es- Chapter Three: Research Reviews success in the long run over repeated experiments of a similar nature. The remote viewing experiments are no exception. Remember that such events should not replicate with any degree of precision in the short run because of statistical variability, just as we would not expect to always get five heads and five tails if we flip a coin ten times, or see the same batting averages in every game. The analysis of SRI experiments conducted in 1988 singled out the laboratory remote viewing sessions performed by six "expert" remote viewers, numbers 002, 009, 131, 372, 414 and 504. These six individuals contributed 196 sessions. The resulting effect size was 0.385 (May et al, 1988, p. 13). The SRI analysis does not include information individually by viewer, nor does it include information about how many of the 196 sessions used static versus dynamic targets. One report provided to the review team (May, Lantz and Piantineda) included an additional experiment conducted after the 1988 review was performed, in which Viewer 009 participated with 40 sessions. The effect size for Viewer 009 for those sessions was .363. None of the other five SRI experts were participants. The same subject identifying numbers were used at SAIC, so we can compare the performance for these individuals at SRI and SAIC. Of the six, three were specifically mentioned as participating in the SAIC remote viewing experiments. As can be seen in Table 2, viewers 009, 131 and 372 all participated in Experiment 1 and viewers 009 and 372 participated in Experiments 4, 9 and 10 as well. The overall effect sizes for two of the three, viewers 009 and 372, were very close to the SRI effect size of 0.385 for these subjects, at .35 and .34, respectively, and the .35 effect size for Viewer 009 was very similar to his .363 effect size in the report by May, Lantz and Piantineda (1994). Therefore, we see a repeated and, more importantly, hopefully a repeatable level of functioning above chance for these individuals. An effect of this size should be reliable enough to be sustained in any properly conducted experiment with enough trials to obtain the long run statistical replicability required to rule out chance. It is also important to notice that viewers 009 and 372 did well on the same experiments and poorly on the same experiments. In fact the correlation between their effect sizes across experiments is .901, which is very close to a perfect correlation of 1.0. This kind of American Institutes for Research 3-21 Review 1: Dr. Jessica Utts consistency warrants investigation to determine whether it is the nature of the experiments, a statistical fluke or some methodological problems that led these two individuals to perform so closely to one another. If methodological problems are responsible, then they must be subtle indeed because the methodology was similar for many of the experiments, yet the results were not. For instance, procedures for the sessions with static and dynamic targets in Experiment I were almost identical to each other, yet the dynamic targets did not produce evidence of psychic functioning (p-value = .50) and the static targets did (p-value = .0073). Therefore, a methodological problem would have had to differentially affect results for the two types of targets, even though the assignment of target type was random across sessions. 4.5 METHODOLOGICAL ISSUES IN THE REMOTE VIEWING EXPERIMENTS AT SAIC As noted in Section 2.3, there are a number of methodological considerations needed to perform a careful remote viewing experiment. Information necessary to determine how well each of these was addressed is generally available in the reports, but in some instances I consulted Dr. May for additional information. As an example of how the methodological issues in Section 2.3 were addressed, an explanation will be provided for Experiment 1. In this experiment the viewers all worked from their homes (in New York, Kansas, California, and Virginia). Dr. Nevin Lantz, who resided in Pennsylvania, was the principal investigator. After each session, viewers faxed their response to Dr. Lantz and mailed the original to SAIC. Upon receipt of the fax, Dr. Lantz mailed the correct answer to the viewer. The viewers were supposed to mail their original responses to SAIC immediately, after faxing them to Dr. Lantz. According to Dr. May, the faxed versions were later compared with the originals to make sure the originals were sent without any changes. Here are how the other methodological issues in Section 2.3 were handled: ? No one who has knowledge of the specific target should have any contact with the viewer until after the response has been safely secured. No one involved with the experiment had any contact with the viewers, since they were not in the vicinity of either SAIC or Dr. Lantz's home in Pennsylvania. 3-22 American nstitutes for Research Chapter Three: Research Reviews No one who has knowledge of the specific target or even of whether or not the session was successful should have any contact with the judge until after that task has been completed. Dr. Lantz and the individual viewers were the only ones who knew the correct answers, but according to Dr. May, they did not have any contact with the judge during the period of this experiment. ? No one who has knowledge of the specific target should have access to the response until after the judging has been completed. Again, since only the viewers and Dr. Lantz knew the correct target, and since the responses were mailed to SAIC by the viewers before they received the answers, this condition appears to have been met. Targets and decoys used in judging should be selected using a well tested randomization device. This has been standard practice at both SRI and SAIC. ? Duplicate sets of targets photographs should be used, one during the experiment and one during the judging, so that no cues (like fingerprints) can be inserted onto the target that would help the judge recognize it. This was done; Dr. Lantz maintained the set used during the experiment while the set used for judging was kept at SAIC in California. ? The criterion for stopping an experiment should be defined in advance so that it is not called to a halt when the results just happen to be favorable. Generally, that means specifying the number of trials in advance, but some statistical procedures require other stopping rules. The important point is that the rule be defined in advance in such a way that there is no ambiguity about when to stop. American Institutes fo-r-Researc In advance it was decided that each viewer would contribute 40 trials, t(-,n under each of four conditions (all combinations of sender/no sender and static/dynamic). All sessions were completed. Reasons, if any, for excluding data must be defined in advance and followed consistently, and should not be dependent on the data. For example, a rule specifying that a trial could be aborted if the viewer felt ill would be legitimate, but only if the trial was aborted before anyone involved in that decision knew the correct target. No such reasons were given, nor was there any mention of any sessions being aborted or discarded. ? Statistical analyses to be used must be planned in advance of collecting the data so that a method most favorable to the data isn't selected post hoc. If multiple methods of analysis are used, the corresponding conclusions must recognize that fact. The standard rank-order judging had been planned, with results reported separately for each of the four conditions in the experiment for each viewer. Thus, 20 effect sizes were reported, four for each of the five viewers. 4.6 WAS ANYTHING LEARNED AT SAIC? 4.6.1 TARGET SELECTION. In addition to the question of whether or not psychic functioning is possible, the experiments at SAIC were designed to explore a number of hypotheses. Experiments 1 and 10 were both designed to see if there is a relationship between the "change in visual entropy" in the targets and the remote viewing performance. Each of the five senses with which we. are familiar is a change detector. Our vision is most readily drawn to something that is moving, and in fact if our eyes are kept completely still, we cease to see at all. Similarly, we hear because of moving air, and our attention is drawn to sudden changes in sound levels. Other senses behave similarly. Thus, it is reasonable that if there really is a "psychic sense" then it would follow that same pattern. 3-24 American institutes for Research Chapter Three: Research Reviews Experiments 1 and 10 were designed to test whether or not remote viewing performance would be related to a particular type of change in the target material, namely the "change in visual entropy." A target with a high degree of change would be one in which the colors changed considerably throughout the target. A detailed explanation can be found in the SAIC reports of this experiment, or in the article "Shannon Entropy: A Possible Intrinsic Target Property" by May, Spottiswoode and James, in the Journal of Parapsychology, December 1994. It was indeed found that there was a correlation between the change in entropy in the target and the remote viewing quality. This result was initially shown in Experiment I and replicated in Experiment 10. A simulation study matching randomly chosen targets to responses showed that this was unlikely to be an artifact of target complexity or other features. It is worth speculating on what this might mean for determining how psychic functioning works. Physicists are currently grappling with the concept of time, and cannot rule out precognition as being consistent with current understanding. Perhaps it is the case that we do have a psychic sense, much like our other senses, and that it works by scanning the future for possibilities of major change much as our eyes scan the environment for visual change and our ears are responsive to auditory change. That idea is consistent with anecdotal reports of precognition, which are generally concerned with events involving major life change. Laboratory remote viewing may in part work by someone directing the viewer to focus on a particular point in the future, that in which he or she receives the feedback from the experiment. It may also be the case that this same sense can scan the environment in actual time and detect change as well. Another hypothesis put forth at SAIC was that laboratory remote viewing experiments are most likely to be successful if the pool of potential targets is neither too narrow nor too wide in terms of the number of possible elements in the target. They called this feature the "target- pool bandwidth" and described it as the number of "differentiable cognitive elements." They reasoned that if the possible target set was too small, the viewer would see the entire set and be unable to distinguish that information from the psychic information. If the set was too broad, the viewer would not have any means for editing an extensive imagination. American Institutes fo-r-Research 3-25 Review 1: Dr. Jessica Utts Combining these two results would indicate that a good target set would contain targets with high change in visual entropy, but that the set would contain a moderately-sized set of possibilities. The set of 100 National Geographic photographs used in the later days at SRI and at SAIC may have inadvertently displayed just those properties. 4.6.2 REMOTE STARING. Experiment 7, described in Appendix 2, provided results very different from the standard remote viewing work. That experiment was designed to test claims made in the Former Soviet Union and by some researchers in the United. States, that individuals could influence the physiology of another individual from a remote location. The study was actually two separate replications of the same experiment, and both replications were successful from a traditional statistical perspective. In other words, it appeared that the physiology of one individual was activated when he or she was being watched by someone in a distant room. If these results are indeed sound, then they may substantiate the folklore indicating that people know when they are being observed from behind. 4.6.3 ENHANCED BINARY COMPUTER GUESSING. Experiment 2 was also very different from the standard remote viewing experiments, although it was still designed to test anomalous cognition. Three subjects attempted to use a statistical enhancement technique to increase the ability to guess forced choice targets with two choices. This clever computer experiment showed that for one subject, guessing was indeed enhanced from a. raw rate of just above chance (51.6% instead of 50%) to an enhanced rate of 76 percent. The method was extremely inefficient, and it is difficult to imagine practical uses for this ability, if indeed it exists. 5. EXTERNAL VALIDATION: REPLICATIONS OF OTHER EXPERIMENTS 5.1 CONCEPTUAL SIMILARITY: GANZFELD EXPERIMENTS While remote viewing has been the primary activity at SRI and SAIC, other researchers have used a similar technique to test for anomalous cognition, called the ganzfeld. As noted in the SAIC Final Report of 29 Sept. 1994, the ganzfeld experiments differ from. remote viewing in three fundamental ways. First, a "mild altered state is used," second, senders are [usually] 3-26 American Institutes or Research Chapter Three: Research Reviews used, so that telepathy is the primary mode, and third, the receivers (viewers) do their own judging just after the session, rather than having an independent judge. The ganzfeld experiments conducted at Psychophysical Research Laboratories (PRL) were already mentioned in Section 3.4. Since the time those results were reported, other laboratories have also been conducting ganzfeld experiments. At the 1995 Annual Meeting of the Parapsychological Association, three replications were reported, all published in the peer- reviewed Proceedings of the conference. The ganzfeld experiments differ in the preferred method of analysis as well. Rather than using the sum of the ranks across sessions, a simple count is made of how many first place matches resulted from a series. Four rather than five choices are given, so by chance there should be about 25% of the sessions resulting in first place matches. 5.2 GANZFELD RESULTS FROM FOUR LABORATORIES In publishing the ganzfeld results from PRL, Bern and Honorton (1994) excluded one of the studies from the general analysis for methodological reasons, and found that the remaining studies showed 106 hits out of 329 sessions, for a hit rate of 32.2 percent when 25 percent was expected by chance. The corresponding p-value was .002. As mentioned earlier, the hallmark of science is replication. This result has now been replicated by three additional laboratories. Bierman (1995) reported four series of experiments conducted at the University of Amsterdam. Overall, there were 124 sessions and 46 hits, for a hit rate of 37 percent. The hit rates for the four individual experiments were 34.3 percent, 37.5 percent, 40 percent and 36.1 percent, so the results are consistent across his four experiments. Morris, Dalton, Delanoy and Watt (1995) reported results of 97 sessions conducted at the University of Edinburgh in which there were 32 successes, for a hit rate of 33 percent. They conducted approximately equal numbers of sessions under each of three conditions. In one condition there was a known sender, and in the other two conditions it was randomly determined at the last minute (and unknown to the receiver) that there would either be a American Institutes for Research 3-27 Review 1: Dr. Jessica Utts sender or not. Hit rates were 34 percent when there was a known sender and when there was no sender, and 28 percent when there was a sender but the receiver did not know whether or not there would be. They did discover post hoc that one experimenter was more successful than the other two at achieving successful sessions, but the result was not beyond what would be expected by chance as a post hoc observation. Broughton and Alexander (1995) reported results from 100 sessions at the Institute for Parapsychology in North Carolina. They, too, found a similar hit rate, with 33 hits out of 100 sessions, or 33 percent hits. Results from the original ganzfeld work and these three replications are summarized in Table 3, along with the SRI and SAIC remote viewing results. The effect sizes for the ganzfeld replications are based on Cohen's h, which is similar in type to the effect size used for the remote viewing data. Both effect sizes measure the number of standard deviations the results fall above chance, using the standard deviation for a single session. TABLE 3: REMOTE VIEWING AND GANZFELD REPLICATIONS Laboratory Sessions Hit Rate Effect Size All Remote Viewing at SRI 770 N/A .209 All Remote Viewing at SAIC 455 N/A .230 PRL, Princeton, NJ 329 32 percent 167 University of Amsterdam, Netherlands 124 37 percent .261 University of Edinburgh, Scotland 97 33 percent .177 Institute for Parapsychology, NC 100 33 percent .177 3-28 American Institutes for Research Chapter Three: Research Reviews 5.3 CONCLUSIONS ABOUT EXTERNAL REPLICATION The results shown in Table 3 show that remote viewing has been conceptually replicated across a number of laboratories, by various experimenters and in different cultures. This is a robust effect that, were it not in such an unusual domain, would no longer be questioned by science as a real phenomenon. It is unlikely that methodological problems could account for the remarkable consistency of results shown in Table 3. 6. IS REMOTE VIEWING USEFUL? Even if we were all to agree that anomalous cognition is possible, there remains the question of whether or not it would have any practical use for government purposes. The answer to that question is beyond the scope of this report, but some speculations can be made about how to increase the usefulness. First, it appears that anomalous cognition is to some extent possible in the general population. None of the ganzfeld experiments used exclusively selected subjects. However, it also appears that certain individuals possess more talent than others, and that it is easier to find those individuals than to train people. It also appears to be the case that certain individuals are better at some tasks than others. For instance, Viewer 372 at SAIC appears to have a facility with describing technical sites. Second, if remote viewing is to be useful, the end users must be trained in what it can do and what it cannot. Given our current level of understanding, it is rarely 100 percent accurate, and there is no reliable way to learn what is accurate and what is not. The same is probably true of most sources of intelligence data. Third, what is useful for one purpose may not be useful for another. For instance, suppose a remote viewer could describe the setting in which a hostage is being held. That information may not be any use at all to those unfamiliar with the territory, but could be useful to those familiar with it. American Institutes 7. CONCLUSIONS AND RECOMMENDATIONS It is clear to this author that anomalous cognition is possible and has been demonstrated. This conclusion is not based on belief, but rather on commonly accepted scientific criteria. The phenomenon has been replicated in a number of forms across laboratories and cultures. The various experiments in which it has been observed have been different enough that if some subtle methodological problems can explain the results, then there would have to be a different explanation for each type of experiment, yet the impact would have to be similar across experiments and laboratories. If fraud were responsible, similarly, it would require an equivalent amount of fraud on the part of a large number of experimenters or an even larger number of subjects. What is not so clear is that we have progressed very far in understanding the mechanism for anomalous cognition. Senders do not appear to be necessary at all; feedback of the correct answer may or may not be necessary. Distance in time and space do not seem to be an impediment. Beyond those conclusions, we know very little. I believe that it would be wasteful of valuable resources to continue to look for proof. No one who has examined all of the data across laboratories, taken as a collective whole, has been able to suggest methodological or statistical problems to explain the ever-increasing and consistent results to date. Resources should be directed to the pertinent questions about how this ability works. I am confident that the questions are no more elusive than any other questions in science dealing with small to medium sized effects, and that if appropriate resources are targeted to appropriate questions, we can have answers within the next decade. American institutes for Research Chapter Three: Research Reviews 8. REFERENCES Bern, Daryl J. and Charles Honorton (1994). "Does psi exist? Replicable evidence for an anomalous process of information transfer," Psychological Bulletin, 115, 4-18. Bierman, Dick J. (1995). "The Amsterdam Ganzfeld Series III & IV: Target clip emotionality, effect sizes and openness," Proceedings of the 38th Annual Parapsychological Association Convention, 27-37. Broughton, Richard and Cheryl Alexander (1995). "Autoganzfeld II: The first 100 sessions," Proceedings of the 38th Annual Parapsychological Association Convention, 53-61. May, Edwin C. (1995). "AC Technical trials: Inspiration for the target entropy concept," May 26, 1995, SAIC Technical Report. May, Edwin C., Nevin D. Lantz and Tom Piantineda (1994). "Feedback considerations in anomalous cognition experiments," Technical Report, Nov. 29, 1994. May, Edwin C., J.M. Utts, V.V. Trask, W.W. Luke, T.J. Frivold and B.S. Humphrey (1988). "Review of the psychoenergetic research conducted at SRI International (1973-1988)," SRI International Technical Report, March 1989. Morris, Robert L., Kathy Dalton, Deborah Delanoy and Caroline Watt (1995). "Comparison of the sender/no sender condition in the ganzfeld, " Proceedings of the 38th Annual Parapsychological Association Convention, 244-259. Puthoff, Harold E. and Russell Targ (1975). "Perceptual augmentation techniques: Part two - research report," Stanford Research Institute Final Report, Dec. 1, 1975. American Institutes for Research APPENDIX 1 EFFECT SIZE MEASURE USED WITH RANK ORDER JUDGING In general, effect sizes measure the number of standard deviation the true population value of interest falls from the value that would be true if chance alone were at work. The standard deviation used is for one subject, trial, etc., rather than being the standard error of the sample statistic used in the hypothesis test. In rank-order judging, let R be the rank for a single trial. If the number of possible choices is N, then we find: E(R)=(N+1)/2 Var(R) = Nx-1 Therefore, when N = 5, we find E(R) = 3 and Var(R) _= 2. The effect size is therefore: Effect Size = (3.0 - Average Rank) 12 3-32 American Institutes or Research Chapter Three: Research Reviews APPENDIX 2 A BRIEF DESCRIPTION OF THE SAIC EXPERIMENTS EXPERIMENTS INVOLVING REMOTE VIEWING There were six experiments involving remote viewing, done for a variety of purposes. EXPERIMENT 1: TARGET AND SENDER DEPENDENCIES: PURPOSE: This experiment was designed to test whether or not a sender is necessary for successful remote viewing and whether or not dynamic targets, consisting of short video clips, would result in more successful remote viewing than the standard National Geographic photographs used in most of the SRI experiments. METHOD: Five experienced remote viewers participated, three of whom (#s 009, 131 and 372) were included in the experienced group at SRI; their identification numbers were carried over to the SAIC experiments. Each viewer worked from his or her home and faxed the results of the sessions to the principal investigator, Nevin Lantz, located in Pennsylvania. Whether the target was static or dynamic and whether or not there was a sender was randomly determined and unknown to the viewer. Upon receiving the fax of the response, Dr. Lantz mailed the correct answer to the viewer. The original response was sent to SAIC in California, where the results were judged by an analyst blind to the correct target. Standard rank-order judging was used. Since it is not explicitly stated, I asked Dr. May what measures were taken to make sure the viewer actually mailed the original response to SAIC before receiving the correct answer in the mail. He said that the original faxed responses were compared with the responses received by SAIC to make sure they were the same, and they all were. RESULTS: Each viewer contributed ten trials under each of the four possible conditions (sender/no sender and static/dynamic target), for a total of 40 trials per viewer. There was a American Institutes for Research 3-33 moderate difference (effect size = .121, p = .08) between the static and dynamic targets, with the traditional National Geographic photographs faring better than the dynamic video clips. There was no noticeable difference based on whether or not a sender was involved, supporting the same conclusion reached in the overall analysis of the SRI work. Combined over all conditicas and all viewers, the effect size was 0. 124 (p = .04); for the static targets alone it was .248 (exact p = .0073) while for the dynamic targets it was 0.00 (p = .50). DISCUSSION: The SAIC staff speculated that the dynamic targets were not successful because the possibilities were too broad. They chose a new set of dynamic targets to be more similar to the static targets and performed another experiment the following year to compare the static targets with the more similar set of dynamic ones. That experiment is described below (Experiment 10.) EXPERIMENT 4: ENHANCING DETECTION OF AC WITH BINARY CODING: PURPOSE: This experiment was designed to see if remote viewing could be used to develop a message-sending capability by focusing on the presence or absence of five specific features of a target. The target set was constructed in packets of four, with possible combinations of the absence (0) or presence (1) of each of the five features chosen to correspond to the numbers 00000, 01110, 10101, and 11011. This is standard practice in information theory when trying to send a two digit number (00, 01, 10 or 11); the remaining three bits are used for "error corrections." Different sets of five features were used for each of ten target packs. METHOD: Five viewers each contributed eight trials, but the same eight targets were used for all five viewers. There was no sender used, and viewers were told that each target would be in a fixed location for one week. They were to spend 15 minutes trying to draw the target, then fax their responses to SAIC in California. The results were blind judged and the binary features were coded by both the viewers and an independent analyst. RESULTS: The results were unsuccessful in showing any evidence of psychic functioning. Neither standard rank-order judging nor analysis based on the binary guesses showed any promise that this method works to send messages. 3-34 American Institutes for Research Chapter Three: Research Reviews EXPERIMENT 5: AC IN LUCID DREAMS (BASELINE): PURPOSE: Despite its name, this experiment did not involve lucid dreaming. Instead, it was used to test three novice remote viewers who were to participate in an experiment involving remote viewing while dreaming. This baseline experiment was designed to see if these individuals would be successful at standard laboratory remote viewing. METHOD: For this baseline experiment, each of the three viewers contributed eight trials using a standard protocol common in the SRI era. For each trial, a target was randomly chosen from the set of 100 National Geographic targets used at SRI and SAIC. The target was placed on a table (so no sender was used) while the viewer, in another room, was asked to provide a description. The response was later blind judged by comparing it to the target and four decoys, and providing a rank-ordering of the five choices. RESULTS: Of the three novice viewers, one obtained a promising effect size of .265, although the result was not statistically significant due to the small number of trials (8). Individual results were not provided for the other two viewers, but the overall effect size was reported as 0.088 for the three viewers. EXPERIMENT 6: AC IN LUCID DREAMS (PILOT): PURPOSE: A lucid dream is a dream in which one becomes aware that he or she is dreaming, and can control subsequent events in the dream. This ability has apparently been successfully trained by Dr. Stephen LaBerge of the Lucidity Institute. He was the Principal Investigator for this experiment. The experiment was designed to see if remote viewing could be successfully employed while the viewer was having a lucid dream. METHOD: Seven remote viewers were used; four were experienced SAIC remote viewers and three were experienced lucid dreamers from the Lucidity Institute. The latter three were the novice viewers used in Experiment 5. The experienced SAIC remote viewers were given training in lucid dreaming. The number of trials contributed by each viewer could not be fixed in advance because of the difficulty of attaining the lucid dream state. A total of 21 American Institutes for Research 3-35 trials were conducted, with the seven viewers contributing anywhere from one to seven trials each. The report did not mention whether or not the stopping criterion was fixed in advance, but according to Dr. May, the experiment was designed to proceed for a fixed time period and to include all sessions attained during that time period. Unlike with standard well controlled protocols, the viewers were allowed to take the target material home with them. The targets, selected from the standard National Geographic pool, were sealed in opaque envelopes with covert threads to detect possible tampering (there were no indications of such tampering). Viewers were instructed to place the targets at bedside and to attempt a lucid dream in which the envelope was opened and the target viewed. Drawings and descriptions were then to be produced upon awakening. RESULTS: Th.: results were blind judged using the standard sum of ranks. Since the majority of viewers contributed only one or two trials, analysis by individual viewer would be meaningless. For the 21 trials combined, the effect size was 0.368 (p = .046). Information was not provided to differentiate the novice remote viewers from the experienced ones. EXPERIMENT 9: ERD AC BEHAVIOR: PURPOSE: The remote viewing in this experiment was conducted in conjunction with measurement of brain waves using an EEG. The purpose of the experiment was to see whether or not EEG activity would change when the target the person was attempting to describe was briefly displayed on a computer monitor in a distant room. Details of the EEG portion will be explained as Experiment 8. Here, we summarize the remote viewing part of the study. METHOD: Three experienced remote viewers (#s 009, 372, and 389) participated. Because of the pilot nature of the experiment, the number of trials differed for each viewer based on availability, with viewers 009, 372, and 389 contributing 18, 24, and 28 trials, respectively. Although it is not good protocol to allow an unspecified number of trials, it does not appear that this problem can explain the results of this experiment. 3-36 American Institutes for Research Chapter Three: Research Reviews RESULTS: Responses were blind judged using standard rank-order analysis. The effect sizes for the viewers 009, 372 and 389 were 0.432 (p = .033), 0.354 (p = .042) and 0.177 (p = .175), respectively. The overall effect size was 0.303 (p = 0.006). EXPERIMENT 10: ENTROPY II: PURPOSE: This experiment was designed as an improved version of Experiment 1. After the unsuccessful showing for the dynamic targets in Experiment 1, the SAIC team speculated that the "target pool bandwidth" defined as the number of "cognitively differentiable elements" in the target pool might be an important factor. If the possible target material was extremely broad, viewers might have trouble filtering out extraneous noise. If the set of possibilities was too small, as in forced choice experiments, the viewer would see all choices at once and would have trouble filtering out that knowledge. An intermediate range of possibilities, too large to be considered all at once, was predicted to be ideal. The standard National Geographic pool seemed to fit that range. For this experiment, a pool of dynamic targets was created with a similar "bandwidth." In both Experiments (1 and 10) the researchers predicted that remote viewing success would correlate with the change in visual entropy of the target, as explained in Section 4.6.1. METHOD: Four of the five viewers from Experiment I were used (#s 009, 372, 389 and 518). They each contributed equal numbers of sessions with static and dynamic targets, with the viewers blind to which trials had which type. Senders were not used, and all sessions were conducted at SAIC in California, unlike Experiment I in which the viewers worked at home. Viewer #372 contributed 15 of each type while the others each contributed 10 of each type. Standard rank-order judging was used. RESULTS: Table 4 shows the results for this experiment. Unlike in Experiment 1, the static and dynamic targets produced identical effect sizes, with both types producing very successful results. The combined effect size for all trials is .55, resulting in a z-score of 5.22. American Institutes for Research 3-37 TABLE 4: RESULTS FOR EXPERIMENT 10 Static Targets Dynamic Targets Viewer Rank ES p Rank ES p 009 2.20 .565 .037 1.70 .919 1.8X10-3 372 1.87 .801 9.7X10 1.93 .754 1.8X10-3 389 3.10 -.071 .589 3.0 .000 .500 518 1.90 .778 7.2X103 2.4 .424 .091 Total 2.22 .550 1.1X10'5 2.22 .550 1.1X10'5 THE OTHER EXPERIMENTS AT SAIC There were four additional experiments at SAIC, not involving remote viewing. Two of them (experiments 3 and 8) involved trying to measure brain activity related to psychic functioning and will be described briefly. Experiment 3 used a magnetoenchephalograph (MEG) to attempt to detect anomalous signals in the brain when a remote stimulus was present. Due to the background noise in the brain measurements and the expected strength of the signal, the experimenters realized too late that they would not be able to detect a signal even if it existed. Experiment 8 utilized an EEG to try to detect the interruption of alpha waves when a remote viewing target was briefly displayed on a computer monitor in another room. The area of the brain tested was that corresponding to visual stimuli. No significant change in alpha was seen. The remaining two experiments were replications of previous work measuring psychic functioning in areas other than remote viewing. They will be described in detail. 3-38 American instit5tes for Research Chapter Three: Research Reviews EXPERIMENT 2: AC OF BINARY TARGETS: PURPOSE: This experiment attempted to replicate and enhance random number generator experiments conducted at SRI. In these types of experiments, a computer randomly selects one of two choices to be the target, denoted as 0 or 1. The internal workings of the computer then rapidly oscillate between 0 and 1 and the subject pushes a mouse button when he or she thinks the internal choice matches the target choice. This process is repeated over many trials. The computer tabulates the results and the experiment is a success if the subject guesses the correct answer more often than would be expected by chance. The purpose is to see if humans can correctly guess computer-selected binary targets, and, hopefully, by extension, correctly solve binary choice problems in real situations. If that were to be the case, then real problems could be posed as binary ones (e.g., is the lost child still in this city or not) to narrow down possibilities. METHOD: This SAIC experiment was designed to enhance the accuracy of binary guessing by using a statistical technique called sequential analysis. Rather than just one guess for each decision, the subject continues to guess until the computer ascertains that a decision has been reached. The computer keeps track of the number of times zero and one have each been guessed and announces a decision when one of the choices has clearly won out over the other, or when it is clear that it is essentially an ongoing tie. In the latter case, no decision is recorded. Three subjects participated (#s 007, 083 and 531) in this experiment. Subject #531 had been successful in similar experiments at SRI. RESULTS: Using this method for enhancing the accuracy of the guesses, subject #531, who had been successful in previous similar experiments, was able to achieve 76 correct answers out of 100 tries. This remarkable level of scoring for this type of experiment resulted in an effect size of .520 and a z-score of 5.20. The other two subjects did not differ from chance results, with 44 and 49 correct decisions out of 100 or 101. (One subject accidentally contributed an additional trial.) Although the result for subject 531 is remarkably successful, it does not represent a very efficient method of obtaining the decision. To reach the 100 decisions required a total of 21,337 guesses, i.e., over 200 guesses for each decision. Of the individual guesses, only 51.6 American Institutes for Research 3-39 percent were correct, for an effect size of .032, similar to other forced choice experiments. Due to the large number of guesses, the corresponding z, score was 4.65. Combined over all three subjects, 56 percent of the 301 decisions were correct and the effect size was 0.123. The combined results were still statistically significant, with p = 0. 017, as sho'am in Table 1. EXPERIMENT 7: REMOTE OBSERVATION: PURPOSE: It is often reported anecdotally that people know when they are being watched. Two experiments were conducted at SAIC to determine whether or not these anecdotes could be supported by a change in physiology when someone is being observed from a distance. The experimental design was essentially the same for the two experiments. This work was a conceptual replication of results reported by researchers in the Former Soviet Union (FSU), the United States, and Scotland. The experiments in the FSU were interpreted to mean that the physiology of the recipient was being manipulated by the sender, an effect that, if rea,l could have frightening consequences. METHOD: The "observee" was seated in a room with a video camera focused on him or her, and with galvanic skin response measurements being recorded. In a distant room the "observer" attempted to influence the physiology of the observee at randomly spaced time intervals. During those time intervals, an image of the observee appeared on a computer monitor watched by the observer. During "control" periods, the video camera remained focused on the observee but the computer monitor did not display his or her image to the observer. There were 16 "influence" periods randomly interspersed with 16 "control" periods, each of 30 seconds, with blank periods of 0 to 5 seconds inserted to rule out patterns in physiology. RESULTS: To determine whether or not the galvanic skin response of the observees was activated while they were being watched, the response during the control periods was compared with the response during the "influence" periods for each subject. The results were then averaged across subjects. In both experiments, there was greater activity during the periods of being watched than there was during the control periods. The results were statistically significant in each case (p = .036 and .014) and the effect sizes were similar, at 0.39 and 0.49. As preplanned, the results were combined, yielding an effect size of .39 (p = 3-40 American Institutes for Research Chapter Three: Research Reviews .005). As an interesting post hoc observation, it was noted that the effect was substantially stronger when the observer and observee were of opposite sexes than when they were of the same sex. DISCUSSION: This experiment differs from the others conducted at SAIC since it involves interaction between two people rather than one person ascertaining information about the environment or the future. It raises substantially more questions than it answers, since the mechanism for the shift in physiology is unknown. Possibilities range from the idea that the observee was able to know when the computer in the distant room was displaying his or her image, not unlike remote viewing, to the possibility that the observer actually did influence the physiology of the observee. Further experimentation as well as a review of similar past experiments may be able to shed light on this important question. American Institutes or esearc Review 1: Dr. Jessica Utts 3-42 American Institrltes for Research Chapter Three: Research Reviews Review 2 Evaluation of Program on "Anomalous Mental Phenomena" Ray Hyman University of Oregon Department of Psychology Eugene, Oregon September 11, 1995 INTRODUCTION Professor Jessica Utts and I were given the task of evaluating the program on "Anomalous Mental Phenomena" carried out at SRI International (formerly the Stanford Research Institute) from 1973 through 1989 and continued at SAIC (Science Applications International Corporation) from 1992 through 1994. We were asked to evaluate this research in terms of its scientific value. We were also asked to comment on its potential utility for intelligence applications. The investigators use the term "Anomalous Mental Phenomena" to refer to what the parapsychologists label as psi. Psi includes both extrasensory perception (called Anomalous Cognition by the present investigators) and psychokinesis (called Anomalous Perturbation by the present investigators). The experimenters claim that their results support the existence of Anomalous Cognition - especially clairvoyance (information transmission from a target without the intervention of a human sender) and precognition. They found no evidence for the existence of Anomalous Perturbation. Our evaluation will focus on the 10 experiments conducted at SAIC. These are the most recent in the program as well as the only ones for which we have adequate documentation. American Institutes for-R-es-e-a-rch Review 2: Dr. Ray Hyman The earlier SRI research on remote viewing suffered from methodological inadequacies. Another reason for concentrating upon this more recent set of experiments is the limited time frame allotted for this evaluation. I will not ignore entirely the earlier SRI research. I will also consider some of the contemporary research in parapsychology at other laboratories. This is because a proper scientific evaluation of any research program has to place it in the context of the broader scientific community. In addition, some of this contemporary research was subcontracted by the SAIC investigators. Professor Utts has provided an historical overview of the SRI and SAIC programs as well as descriptions of the experiments under consideration. I will not duplicate what she has written on these topics. Instead, I will focus on her conclusions that: Using the standards applied to any other area of science, it is concluded that psychic functioning has been well established [Utts, Sept. 1995, p 1] Arguments that these results could be due to methodological flaws in the experiments are soundly refuted effects of similar magnitude to those found in government- sponsored research at SRI and SAIC have been replicated at a number C~r la