Memory Search By A Memorist

Charles P. Thompson, Thaddeus M. Cowan, Jerome Frieman

• Publisher: Psychology Press
• Year: 2013
• Print ISBN: 9781138876354
• eBook ISBN: 9781134774418
• Topic: Psychology
• Subtopic: Cognitive Psychology & Cognition
• Index: Psychology

This book describes the first comprehensive experimental research program on an individual who exhibits exceptional memory. Rajan Mahadevan, the subject of these investigations, won a place in the Guinness Book of Records for reciting pi to 31,811 decimal places, can learn matrices up to size 20 X 20, and can produce memory spans above 60 for digits. Utilizing the methodology and theories of modern cognitive psychology, the authors systematically investigated Rajan's memory skills. A wide range of experiments and tests were conducted with Rajan and four control subjects. These include memory span tests for digits and letters, memory for various kinds of non-numeric information, tests of working memory, learning and retention of numeric matrices, memory and visual searches of the digits of pi, and lexical decision tasks with the digits of pi.

The authors describe how they came to understand the way Rajan stored and now retrieves the decimal digits of pi, how he learns and retrieves matrices, and how he encodes and retrieves digits in a memory span task. Although his strategy for memorizing and retrieving digits is unique in the literature on people with extraordinary memory, the authors show how their investigations of Rajan contribute to our understanding of memory.

Chapter1

Rajan the Person

Rajan Srinivasan Mahadevan developed an extraordinary skill at learning and reciting long strings of digits. On July 5, 1981, he demonstrated this skill by reciting the first 31,811 digits of pi to earn a place in the Guinness Book of World Records. That record stood until March 9, 1987 when another memorist (Hideaki Tomoyori) recited 40,000 digits of pi.

In addition to his memory skills, Rajan is an interesting and rather eccentric fellow. We had the opportunity to interact with Rajan virtually every day for 4 years. He is absolutely delightful in many ways and somewhat exasperating in others.

In this book, our main purpose is to document his exceptional performance in the laboratory with data we collected over a 3-year period. We think the reader will agree that his laboratory performance is much more impressive than his recitation of pi.

Before we begin that task, we describe Rajan the person. This first chapter portrays many of the characteristics that define his rich and unique personality. We begin with a brief autobiography provided by Rajan.

Family and Early Life

Rajan’s Autobiography

I was born in Madras, India, in 1957. Two years later, my family moved to Man-galore as a consequence of my father securing a professorship at the medical school there.

The earliest evidence of my unusual memory was furnished at age 5 when my parents hosted a party for about 50 guests. Being somewhat restless by nature, I embarked on a small tour of the nearest parking lot and submitted the license plate of every guest to close scrutiny. Several minutes later, I returned to my living room and reproduced the license plate number of every guest in the order the vehicles were parked.

Although our guests were impressed with my license plate number mastery, my memory skills never merited any serious consideration from my teachers or my colleagues throughout my early schooling. Their indifference could be attributed at least partially to the fact that I hailed from a predominantly academic family, with many members earning a reputation for strong memory skills.

Throughout my schooling, I used to entertain people by reproducing railway timetables and cricket scores. I recall a particular incident that occurred when I was in Grade 4. Our instructor cited several reasons to demonstrate the safety of railroad travel in India. Somewhat piqued, I got up and gave the class several details of railroad accidents in India over the previous decade (I was able to relive that experience when I watched the movie, Rain Man, in which Dustin Hoffman reproduced, from memory, the data on airplane accidents to Tom Cruise).

In the fall of 1975, I enrolled at the engineering school at Manipal. It was there that people showed considerable curiosity at my numerical memory. Several people urged me to set a world record for memorizing numbers. I wrote to the editors of the Guinness Book of World Records and they advised me to focus on pi. Pi is a well-known mathematical constant and was chosen as the yardstick to measure numerical memory.

I discontinued my engineering education in 1977 (primarily due to a waning of my interest in the field), after which I procured a computer listing of the digits of pi. I had memorized the first 10,000 digits by the time I was invited to demonstrate my memory skills at an international conference on yoga and meditation held in Chicago in 1980.

At the conclusion of the conference, I visited friends at Minnesota. They suggested that a psychologist test my memory, and they contacted a cognitive psychologist, Bill Fox, at the University of Minnesota. Fox put me through an extensive series of tests over a 3-month period. Those tests revealed quite clearly that my memory was predominantly numerical with a vast variation in performance depending on the type of material and the task demands.

On my return to India in early 1981, I enrolled as a psychology major at the School of Social Work in Mangalore. Later that year, I memorized an additional 25,000 digits of pi. On July 5th, 1981, I reproduced 31,811 digits of pi flawlessly and entered the Guinness Book of World Records.

I completed my Bachelor’s degree in psychology in 1984 and received my Master’s degree in clinical psychology from the University of Mysore in 1986. My memory was tested periodically by my instructors during the 5-year period from 1981-1986. The results were generally consistent with Fox’s findings in 1980.

Although my skill is predominantly with numbers, I consider myself as having a good memory for verbal material as well. I can attest to this fact by demonstrating fluency and reasonable accuracy in a dozen languages.

I enrolled as a graduate student in physiological psychology at Kansas State University in the fall of 1987. At Kansas State, a research project was funded in early 1989 to probe the characteristics of my memory. The results of that research constitute the major part of this book.

Rajan and Ramanujan

An interesting addition to Rajan’s autobiography is the origin of his given name, Srinivasan. He is named after a distant cousin who perhaps was India’s most famous mathematician, Srinivasa Ramanujan. Ramanujan was a man for whom it was claimed that every number was a personal friend. A widely circulated story (Newman, 1956) tells how Ramanujan’s mentor, G. H. Hardy, remarked in passing that the license number of a certain taxi, 1729, was a particularly dull number. Ramanujan quickly pointed out that it was an interesting number because it was the smallest number expressible as a sum of two cubes in two different ways. In a poem about Ramanujan, Holden (1985) stated that:

This modest, mousy little boy from India
could reel off pi’s digits to any
decimal place his classmates dared him to.
No mean feat. But for Ramanujan it
was a breeze. Pi was merely one of his
first cousins, in fact a favorite.
And his cousins were innumerable.

It is doubtful that Ramanujan’s playmates ever challenged him with decimal places as remote as those held by one of his mortal cousins, Rajan.

Origin of This Project

Entering Kansas State University

Our part of the story begins in late August of 1987 when Rajan introduced himself to each of us. He was then 30 years old and had just begun graduate work in our department. Rajan’s initial meeting with Charles Thompson demonstrated a central theme in his personality and also led directly to our research project:

Rajan walked into my office, introduced himself, and said that he was a new graduate student in psychology. I welcomed him to the department and said that I hoped he would have an enjoyable and profitable stay with us. I then asked him what areas of psychology interested him. He replied that he was interested in memory. I said something like, “That’s great! My specialty is memory. Is there any particular topic in memory that interests you?” “Yes,” he replied, “my memory.” My thoughts upon hearing that reply were not completely charitable. Nonetheless, I asked what there was about his memory that made it of particular interest to him. Rajan went on to describe some of his memory feats with particular focus on his recitation of almost 32,000 digits of pi. The performance he described was very impressive. Rajan mentioned he had chatted with two of my colleagues (Thad Cowan and Jerry Frieman) and they were both intrigued with his memory skills.

It seemed clear to me that it was worth investing a bit of time to establish whether he had the skills he described. Shortly after my meeting with Rajan, Thad Cowan, Jerry and Jeanne Frieman, and I discussed Rajan at Max Miller’s traditional pre-Labor Day party at beautiful Lake Wabaunsee. I asked whether they would be interested in evaluating Rajan’s skilled performance. This book testifies to their response.

Rajan’s meeting with Thompson was typical of other initial meetings that we observed or had described to us. In virtually every case, Rajan made the description or demonstration of his memory skill the central feature of the conversation. That characterization seems unkind until one considers all the people one knows who are adept at turning almost all conversations to topics in which they are interested and have expertise. The defining differences between those people and Rajan are that their topics do not usually depend on their personal performance, and their expertise is not quite so startling.

More importantly, it appeared to us that Rajan found focusing the conversation on his memory skill an easy way to get through an introduction with grace and charm. Once he mentions his expertise to someone who does not know him, the questions tumble out like oranges from an upset basket. Because the questions are all fairly predictable, Rajan can answer them with the fluency and skill of a great story teller.

As we observed Rajan’s interactions with others, it became clear to us that his memory skill was the central defining feature of his view of himself. At first, that perspective seemed somewhat eccentric to us; however, his behavior became very understandable when we were made aware of the immense amount of attention that Rajan has received as a direct result of his memory performances.

Initial Tests

Our individual meetings with Rajan convinced us that it was worth the effort to determine what his performance looked like under controlled laboratory conditions. We chose memory span and recitation of the decimal digits of pi as our initial tests.

Memory span is a simple test in which a string of digits is presented at a rate of one digit per second. At the end of the presentation, the subject is asked to recite the string in the order presented. The length of the string is increased by one if the recitation is correct and is decreased by one if the subject makes an error. After 10 or 20 trials, performance stabilizes, with subjects usually able to recite strings of a given length but failing when the strings are one digit longer. The memory span for adults is remarkably stable with a typical span of seven. Deviations of more than two digits from that level are extremely rare. We quickly were able to establish that Rajan’s digit span was about 40 digits. We obviously were dealing with an individual who was extremely skilled at recalling digit strings.

Rajan’s performance in reciting pi was equally impressive. We made a videotape of him reciting the first 8,000 digits of pi. Both his accuracy and his speed were impressive: He made no errors while reciting the digits at a rate slightly greater than five digits per second. His recitation rate was so rapid that it took some practice to follow the tape in order to verify the accuracy of his performance.

Rajan’s performance convinced us that we should study his memory. We needed financial support to do the extensive studies that would be necessary. Thus, we quickly submitted a grant proposal to the National Institutes of Mental Health to meet an October 1 deadline. (As one of its many functions, NIMH sponsors basic research in memory.) We paid for our haste, however, as the reviewing committee wisely asked for additional information demonstrating his expertise prior to funding our grant.

As part of our response, we submitted our videotape of Rajan reciting the first 8,000 digits of pi. We heard, both directly and indirectly, that the committee (and others) enjoyed viewing that tape. It was difficult not to be interested in his performance. For example, the program officer in charge of processing our grant application called one day—allegedly to get some piece of information. It soon became clear, however, that she had called to chat about “that interesting young man on whom you propose to do some research.”

Ethical Problems and Solutions

Our decision to use Rajan as a research subject posed several ethical problems. Most importantly, we thought that it would be inappropriate for us to interfere in any way with Rajan’s progress as a graduate student. We solved that problem in two ways: First, we undertook no additional research with Rajan until the grant was funded. Second, we specified in the grant that Rajan would be funded as a graduate research assistant on the project under rather special conditions. Typically, graduate research assistants can use some of their work on the sponsored research (which pays their assistantship) to fulfill the research requirements for their degree. However, that strategy would not work for Rajan because work on oneself is wholly inappropriate for a research degree. To provide Rajan the same opportunity for research relevant to his degree as other research assistants, we specified in our grant proposal that, of the 20 hours per week normally expected of a graduate research assistant, Rajan would spend 10 hours as a subject and would be free to spend the other 10 hours on his own research.

In all other respects, Rajan benefitted from the project in the same way as any other graduate research assistant. He was coauthor on a major paper resulting from the research (Thompson, Cowan, Frieman, Mahadevan, Vogl, & Frie-man, 1991) and coauthor on five paper or poster presentations.

The second problem was that Rajan had not decided whether he would pursue research in memory or in physiological psychology. If he chose memory, Charles Thompson was an obvious candidate to either serve on, or head, his committee. However, we viewed that as an unacceptable conflict of interest because Thompson had also taken the principal investigator role in the grant proposal. Therefore, we agreed that he would play no role in the research Rajan conducted for his degree.

The third problem that concerned us was the possibility that Rajan would be besieged with requests to perform in classes and elsewhere. We resolved to shield Rajan from those requests as much as possible. We soon found that our concern was misplaced. Rajan loved to perform and sought out opportunities to do so. Thus, most of our large introductory courses and most of our cognitive psychology courses had the opportunity to see Rajan perform during the period he was a graduate student at Kansas State University. Rajan also spent many hours at the Espresso Royale Caffe (a local coffee shop) giving performances to individual patrons.

Interviews

Grant Leads to Interviews

Although we were able to begin our research early in 1989, the official starting date of the grant was May 1, 1989. Shortly before that date, the publicity office at Kansas State University put out a routine bulletin describing the grant and its purpose. To our surprise, the Associated Press sent the bulletin on its wire service, and stories of the grant began to appear around the country. Those stories, in turn, led to requests for personal interviews with Rajan.

One of the earliest requests was from T. R. Reid of the Washington Post. He came to Kansas, interviewed Rajan, and published an entertaining story on Rajan in the Washington Post (Sunday, June 18, 1989). His was a very informative and influential story, which led to dozens of interviews. The Washington Post story was reprinted in a large number of newspapers. In January 1990, it appeared in the Reader’s Digest and later in the Spanish language version of that magazine. The Washington Post story apparently was responsible for later reports by Nancy Shulins of the Associated Press and Nancy Kruh of the Dallas Morning News. The latter story was distributed by Knight-Ridder News Service.

Rajan reported to us that he had given 100 interviews as of March of 1990. Those included interviews with the Kansas City Star, Miami Herald, Los Angeles Times, Denver Post, Calcutta Telegraph, and Times of India. He was in...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright
• Dedication
• Contents
• Preface
• 1. Rajan the Person
• 2. Memory Span and Memory Load Tasks
• 3. Non-Numeric Performance
• 4. Number Matrices
• 5. Memory Search
• 6. Searching for Locations in a Visual Display
• 7. Priming Digit Strings
• 8. Comparison with Other Memorists
• 9. Overview and Interpretation of Rajan's Performance
• References
• Author Index
• Subject Index