Murray Sachs, at EPL 1964-1966

 

My Life at EPL and Beyond

     Several years ago I was invited to write a history of research on the physiology of hearing for the Acoustical Society.  The invitation brought back a vivid memory of a summer day some twenty-five years earlier.  I had been playing ball with my young sons and their friends when one of the kids picked up my Stan Musial mitt and asked innocently “who is Stan Musial?” The question shocked me. Who wouldn’t know about my Hall of Fame hero who spent his career in my home town as a St. Louis Cardinal?  As I write this I am also reminded of an event from my days as a student in EPL.  It was October 15, 1964 and for some reason I was in the lab instead of watching as another childhood hero, the Cardinals’ greatest pitcher Bob Gibson, beat the hated New York Yankees in the final game of the World Series.  What could possibly have kept me from seeing my Cardinals win game 7? The answer was easy—Nelson!

     More recently a Hopkins BME graduate student had the temerity to ask me, “who is Jerzy Rose?”  That question struck even deeper. I had only idolized Stan the Man from afar, but I had known Jerzy “up close and personal”.  Jerzy was without question one of the great auditory researchers of our time.  The crowning blow came when a postdoc told me he had thought that Nelson Kiang was my first student,  although nearly the inverse was true – I was Nelson’s second student. In light of this evidence I could only conclude that the next scientific generation has almost no appreciation of the rich history of our field.  I accepted the invitation, in part to try to improve this situation.  For the same reason I applaud Nelson’s putting together this history of the EPL and I am pleased to contribute some of my memories to it.

     Since I was one of the earliest EPL students, I thought I would talk a little about how I got there in the first place.  In the summer of 1959 just before my sophomore year at MIT I met Merle Diener who was to become my wife after a courtship of nine years.  Very early in that long courtship, Merle introduced me to her brother-in-law David Shainberg, a New York psychoanalyst who had a strong interest in relating the work of neurophysiologists to his clinical practice.  To paraphrase a very long discussion, David encouraged me to apply my electrical engineering education to the study of the brain –‘after all the brain is just a bunch of electrical circuits’. I took David’s advice and now fifty years later I look back in awe at how far we have progressed from the naïve view of the brain as electrical circuits to our view of the working of enormously complex genetically controlled systems of atoms and molecules.

     My career in auditory physiology got a fortuitous boost when I entered graduate school at MIT in 1962.  For reasons which I do not recall, I was assigned Moise Goldstein as my academic advisor.  Moise not only became my masters thesis advisor but also a life-long friend and mentor. My masters thesis was carried out in the Communications Biophysics Laboratory in the basement of Building 20. My first experience with neurophysiology was not soon to be forgotten when at Moise’s suggestion I was an observer in a decerebrate cat experiment.  Suffice it to say that my role ended abruptly when I was showered with blood as the result of a ‘slip of the scalpel’ by the post-doc in charge. Fortunately Moise and I had the good sense to focus my masters thesis on recordings from the auditory nerve of a smaller and more manageable animal, the green tree frog.  The thesis related to the coding of species-specific vocalizations, an area pioneered by Moise, Bob Capranica and Larry Frishkopf.

     Three incidents from my days in CBL remain engraved in my memory.  In 1963 I returned from spring vacation to learn to my dismay that Moise was leaving MIT to go to a place I had barely heard of – Johns Hopkins University.  Little did I know that Moise’s move presaged my subsequent 40+ year career at Hopkins.  Nor could I have known how his move brought me closer ties with other mentors in CBL, including Bob Capranica, Tom Weiss, Larry Frishkopf (then at Bell Labs), Bill Siebert, Walter Rosenblith, Bill Peake and Nelson.  The second event was my answer to the question “where were you when Kennedy was shot?”  I was in the sound chamber in the basement of Bldg. 20 when the news first hit and I ran breathlessly to my office in Bldg. 6 where I got the horrible news that the President had died.

     The last memory of my time in CBL was my thesis presentation at the group’s weekly seminar.  These seminars had a fierce reputation for harsh critiques, from which students were not immune. My preparation was not made more relaxed by Walter’s comment to me “not to worry, we just like to push speakers against the wall and see how they bounce back”.  Tom Weiss tried to calm me with the notion that the seminars “helped build resilience”.  It did not help that the speaker in the week preceding my turn had walked out of the meeting when the questioning got too heated (and he was a visiting full professor!).  Moise arrived from Baltimore just in time for the inquisition and we were off and running.  My first “mistake” was to start off by saying that my thesis was supervised by Moise Goldstein.  Like a shot Walter replied “That’s Professor Goldstein”, but in typical Moise fashion he responded equally quickly “In Baltimore they call me Mugsy!”  The laughter got me to the first slide, at which point Walter asked a difficult probing question and I simply had to admit that I did not know the answer.  Moise came to my rescue once more by saying that the data were not yet in.  From that point on the seminar was just a big happy family affair, apparently because the group appreciated someone admitting to not knowing the answer to a tough question.

     By the time I had finished my masters thesis I knew enough about EPL that I wanted to join the group to do my doctoral thesis with Nelson.  Much of my exposure had been through my friendship with Russ Pfeiffer, the first student to do a PhD in the lab.  Russ set up a memorable meeting at EPL with Nelson, Peter Gray and me, memorable not only because of the content but because of the time – 4:00 AM!  Nelson is of course well known for his very short sleeping times.  To quote from a 1989 Boston Globe feature article:

Scientists call them "short sleepers," people who need less than six hours a night of sack time to feel happy and refreshed. They're the kids who used to read under the sheets at night by flashlight. Some of them grew up to become leaders in politics, science and industry …. True, some of them are also just regular folks who simply log more couch potato time and complain that the TV programs aren't very good at 4 in the morning. But many short sleepers seem to be unusually active, even hyperactive, folks who would get more done than the rest of us even if they slept normally. Given the two to four extra hours a day they get, they can inspire downright awe as well as envy. How do you keep up with someone like Nelson Kiang, a four-hour-a-night man who is currently a professor in the department of brain and cognitive sciences at MIT, professor of physiology in the department of otolaryngology at the Harvard Medical School, a neurophysiologist in the neurology service at Massachusetts General Hospital and director of the Eaton-Peabody Laboratory at the Massachusetts Eye and Ear Infirmary?

"My father used to say I'd burn myself out. But since I made it to 60, he now says I've lived twice as long as most people," said Kiang, who is so revved up he finds it impossible to sit through a movie in a theater. He prefers to watch TV so he can eat, read and talk with his wife at the same time”. 

     So, I always count my time at EPL starting at 4:00 AM on some summer morning in 1964.  It is difficult for me to realize that the end of my resident status in EPL ended only two years later on September 1, 1966, the day I began my two years active duty at the Navy Underwater Sound Lab in New London.  The two years at EPL were among the most intense times in my life, both professionally and personally.  As a very early student I can claim a number of “firsts” in my program.  My first single unit experiment was also the first at EPL to use the LINC computer for on-line data processing.  After spending more than a year at CBL working out the problems associated with isolating and holding single units in the auditory nerve, I was ecstatic when sometime after midnight of my first EPL experiment I actually got data from many single units.  In fact I was so excited that I called Nelson in the middle of the night and must have literally yelled into the phone “Nelson, I’ve got units!”

     That first experiment marked the start of an extremely productive year where data just seemed to flow in as if by magic.  I thoroughly loved the experimental part of the work, not to mention the happy social aspects of EPL.  These two aspects of graduate school life came together late one night when Shelley Norton (later Shelley Guinan for a brief time and one of the technicians Nelson had imported from the three sisters schools) appeared in the lab with a friend she had wanted me to meet.  I think this woman (who will remain nameless ) was best described as a classical blond bomb shell.  She spent the night watching my experiment, making extremely acute comments and adding to my already strong sense that science should and could be fun.  But alas, good things sometimes come to an end and this one did when she set her Beacon Hill apartment on fire.

     Another first attached to my stay at EPL came during what was clearly the most difficult time of my years as a student at MIT.  At the time I had completed gathering enough data on two-tone inhibition to make what I thought would be a terrific PhD thesis.  I had, I thought, cleared the last major obstacle on my route to a PhD.  Walter Rosenblith had wanted to expose the scientific world to the graduate students in CBL and I was the first to go on display.  At the time the scientific world of neurophysiology revolved around the future Harvard Nobel laureates David Hubel and Torsten Weisel.   So Walter invited Wiesel as an examiner at the second oral examine then required by the Electrical Engineering department.  This exam concentrated on the student’s thesis activities.  I recall over-hearing Wiesel say to Nelson that he wanted to be sure not to embarrass this engineerng student by asking an unfair biological question.  I do not remember what he asked and in fact the only question I do remember came from Ron Howard, a famous operations research mathematician whose home I was renting while he was on sabbatical.  His question, which I can quote almost verbatim was “Suppose you had a warehouse full of cats and 10,000 technicians, what scientific question would you answer?”  What I do not remember is how long the exam committee let me babble through completely nonsensical attempts at an answer before sending me off to think about it for a few months.  It could not have taken too long for me to prove that I had never looked hard enough to see the forest for the trees in the mass of data I had collected.

     At that point I felt like an absolute failure with no hope of digging myself out of a deep hole.  Nelson insisted that we walk back to the lab where he would explain the world of science to me.  Of course the following “session” took place in his tiny office where I could have my back to the wall while he explained to me that I was smarter than 99% of humanity, but that meant nothing here in the big leagues of MIT.  Nonetheless he would lead me out of the desert.  We had gathered enough data to describe the phenomenon of two tone inhibition, all I had to do was to figure out from those data what were the mechanisms underlying the phenomenon.  Furthermore, he knew the answer and would somehow lead me to the promised-land.  Suffice it to say that at least two rejected thesis proposals later I had made no progress on the mechanism question.  Although we have made a good deal of progress on the mechanism in the past 40+ years and can say now with some confidence, thanks to Bill Rhode and his collaborators, that mechanical nonlinearities are at the root of the “inhibition”, we are little closer to understanding the mechanisms underlying the mechanical nonlinearities.  I use quotation marks around inhibition because it did not take long for the field to conclude that what we were studying was nothing like classical inhibition and we were forced to follow the lead of the Wisconsin group in calling the phenomenon two-tone “suppression”.  But that’s getting a little ahead of the story.

     I was still locked into an intellectual tussle with Nelson and sitting on two non-starters for a thesis proposal when Walter suggested that I take the question to Bill Peake. I cannot give Bill enough credit for turning things around in one very brief conversation where he said to me: “If you want to call yourself an engineer your thesis better have a real modeling component”.  I think back to that remark as the tuning point in my career and in a very real sense it has had a direct impact on biomedical engineering at Johns Hopkins.  More on that relationship later.  So, I finally had an answer to Ron Howard’s question about what I would do with the warehouse full of cats: I would use my ten thousand technicians to get enough data from the cats to build a quantitative model relating discharge rate in auditory-nerve fibers to the parameters of two-tone stimuli.  And  back I went to the thesis committeefor what I thought would be a slam-dunk oral exam. Wrong!  This time I had to face probably the most formidable challenge.  Bill Siebert was one of the deepest thinking modelers of the auditory nerve basis of auditory discrimination in the world.  He also carried the reputation as one of the most critical thesis advisers at MIT, as evidenced by the fact that Russ Pfeiffer was the first student ever to complete a doctorate with Bill as a primary adviser.  It came as no surprise that Bill’s questioning virtually dominated this second exam.  Probably out of self-defense I do not remember the details of the three hours I spent defending my viewpoint, but I think it went something like:

Siebert: “Why don’t you model the effect of inhibition on auditory discrimination?”

Sachs: “First things first.  I need to be able to describe the effects of inhibition on the auditory nerve first and from my nearly three years experience I think that will produce a good thesis.”

Siebert: “OK, you think you know what is best and I see it differently. Do it your way and we’ll see what happens.”

     One thing that I do remember vividly about that exam was the role Nelson played.  Throughout he was seated almost directly in front of me enjoying one of his favorite pastimes—doodling.  I could not help seeing the sketch he was working on and although I have lost track of it somewhere in the past 43 years, my memory is good enough that I have been able to reproduce it with the help of Phyllis Taylor my long-time Hopkins technician:


                                       


The implication was clear.  I must say, however, that Nelson never really left me to drown and the friendship that was born at 4:00 a.m. that summer morning is still strong today.

     What followed from this exam was an intense effort over several months to create the model that I had promised the two Bills and Nelson.  One of the great privileges that came to me was the chance to spend many intense hours in Bill Siebert’s office talking about my progress.  For some reason those meetings always seemed to start innocently enough an hour or two before lunch and never ended until I was ready to pass out from hunger.  Bill’s mind ran so fast that it would take me two to three weeks to decipher what he had said and then prepare for the next lunch-less encounter.  My close ties with Siebert were helped by my roommate Peter Gray who was himself a student of Bill’s.  Peter’s thesis involved detailed analysis of auditory-nerve spike trains, which required massive amounts of data from single units and  he and I worked together to accumulate those data.  Not only was Peter my roommate but became an assistant professor in time to serve on my thesis committee.

     The final tale in my quest for a PhD began the night before my thesis defense, which was another “first” inspired by Walter Rosenblith’s desire to introduce the CBL graduates to a broad audience.  My defense was to be advertised and open to the public.  So late in the afternoon before the defense I sat nervously in my building 20 office reviewing for the umpteenth time the calculations underlying the model I had produced for predicting auditory-nerve rate from the parameters of a two-tone stimulus.  To my horror I could not reproduce the model fits to the nerve data either quantitatively or qualitatively, not even close.  I was so shaken by this apparent disaster that I was unable even to drive home to dinner.  The resolution was simple as it turned out and is a good example of the fact that we were still in the dark ages of data processing.  I had, believe it or not, done all of my model computations using the “latest” in K and E slide rule technology.  On the other hand, the checking of the computations that afternoon was done on a borrowed slide rule and it only took me several hours to understand that I was then making a computational error with the borrowed slide rule.

     Having survived this latest horror show I still had to look forward with considerable trepidation to the defense itself. The program was to be in two acts—the public presentation followed by the private “grilling” by the thesis committee.  The two acts weighed on my psyche in two different ways.  Although I was now the world’s leading expert on all the details of the thesis, there was no doubt in my mind that someone in the audience for the public presentation would stumble on a fatal flaw that had passed me by.  And if that did not happen in Act 1, Act 2 would expose the limitations of my efforts in some way.

     The public audience was almost exclusively people whom I did not recognize, and turned out were predominately from the Department of Psychology.  From the beginning I was unable to answer a single question from the audience, not because I did not know the answers, but before I could open my mouth to answer Bill Siebert waved me off and gave the answer himself, as if to say “That question was too trivial to bother with”.  Act 1 concluded with the usual applause, which although appreciated did nothing to relieve my anxiety over Act 2.  I remember clearly the thesis committee gathered around the podium and hearing Bill Peake uttering what I was sure would be the fatal invitation:

     “Should we go to my office for the private grilling”.

But then as if from nowhere came Siebert’s final comment:

     “I’m satisfied” and since no one ever questioned Bill Siebert’s intellectual judgment there would be no Act 2!  I guess Bill had learned enough in all of those lunch-free meetings.

     This thesis defense for all intents and purposes marked the end of my direct involvement with EPL.  On the last day of August in 1966 I turned in the written document and required copies.  It is with a real sense of nostalgia that I pick up my own copy now and then and can hardly believe that it is printed in the classical mimeo-blue.  And then I recall that the 300+ page document was typed in one weekend (apparently without error by a secretary on Cape Cod).

      I will avoid the great temptation to spin the countless number of sea stories that resulted from my tour of active navy duty, except to say that the high point of my Navy career had to have been the three week submarine research trip from New London to the Azores!  After serving as a Lieutenant in the Navy at the Underwater Sound Lab in New London for two years, I got a post-doctoral fellowship at Cambridge University.  I thank Nelson for his support in this endeavor as he made the original contact with Sir William Rushton in the Physiological Laboratory at Cambridge. 

     My time in Cambridge was certainly a high point in my career and my personal life—it was the first year of marriage for Merle and me and we are now in the 41st year.  There was even a significant tie to EPL.  As soon as I had arrived in Cambridge I began to work on two papers based on my thesis. The first presented all of the data that had been gathered before Bill Peake encouraged me to pursue a modeling effort and was co-authored with Nelson and published in the Journal of the Acoustical Society in 1968. The second paper presented the model and was published by JASA in 1969 and taught me an important lesson about care in publishing.  Reviewing the galley proofs and finding an error in the paper’s title., I carefully marked the error and the publisher correctly made the change I requested.  Unfortunately, in making the change the publisher added a new excruciatingly more embarrassing error—Auditory-Nerve fibers became Auditory-Noise Fibers, leaving the title as Stimulus-Response Relation for Auditory-Noise Fibers: Two-Tone Stimuli.  The error has never been corrected.  As I eventually began to mature, the error began to amuse me and I recognized the two groups of people in the world—those who were my friends and made the correction on their own and those who did not.

     Charlie Liberman’s invitation to take part in EPL50 included the following charge: to talk “about what the lab was like, and how it prepared you (or not) for the real world”.  Because professionally speaking my “real world” for the past 40 years was centered on Hopkins Biomedical Engineering I will focus to some degree on how my EPL experience affected my Hopkins experience.  That experience to a degree can be organized in two closely related aspects – my leadership role in the University and my scientific achievements. I have discussed the development of my scientific life with the numerous vignettes above and want to make clear that the lessons I learned at EPL and MIT have formed the core of my approach to science ever since.  Perhaps Jerzy Rose put it most succinctly when he introduced me to an audience at the University of Wisconsin many years ago.  His words were something like “Murray knows what it means to really solve a problem.”  His words were so important to me that I promised myself that I would always strive to pass on to my students the problem solving skills, which after all were learned at MIT and EPL.  Let me try to express as briefly as possible the principles that I believe now also characterize the problem-solving philosophy of the students we have trained in hearing research in BME at Hopkins.

1.  A problem is not really solved until a quantitative model has been shown to accurately describe the experimental data which were gathered to solve the problem.  This principle was driven home to me by Bill Peake in his exhortation that I create a two-tone suppression model and by my several modeling mentors including Bill Siebert, Tom Weiss and Russ Pfeiffer.

2.  A model in the absence of confirming experiments is an easy target for accusations of irrelevance to the “real world”.  (Some might call this “mental masturbation.”)  A close corollary says that a good modeler must also be a good experimenter, or at least a very close collaborator of a good experimenter.  There are clearly no more talented experimenters in our field than Nelson and the group he has established around him and many of whom I had the privilege of working with as a student.

3.  In basic science, perhaps more than any other professional endeavor, perseverance is not only important it is necessary.  When I was at the Navy Underwater Sound Lab my supervisor simply could not understand how I could be so committed to working on problems that might take a decade to solve.  My formative experience in scientific perseverance occurred early in my training when I simply watched Nelson and his colleagues spend several years polishing the famed “monograph”.

     So, did I succeed in passing these principles to the next generation of hearing researchers at Johns Hopkins?  The answer to this question is complicated because one of the joys of working here is that we are able to recruit an incredible cadre of graduate students, who I am sure would have done very well without me and undoubtedly better than I would have done without them.  First, of course, there was my first student and colleague and closest personal friend for almost 40 years, Eric Young.  It is no exaggeration to say that Eric is recognized as one of the most talented experimenters and theoreticians in auditory research in the world and has been honored with the ARO’s Award of Merit.  Paul Abbas survived the trials of nonlinearities in auditory-nerve fiber rate functions to become a professor of Otolarygology at the University of Iowa,.  Mike Miller in essence did two PhD theses on questions of the temporal responses of auditory nerve fibers.  One was a purely experimental study following which he literally disappeared from the lab to study stochastic processes and became a devotee of everything Siebert.  Mike is now a full professor of BME here and director of the Hopkins Center for Imaging Sciences.  Rai Winslow followed a similar path from intense experimental studies of the role of efferent pathways to the cochlea to extremely innovative computational biological studies of the heart, for which he received the Smithsonian Information Technology Innovators Award. He is currently the Founding Director of the Hopkins Institute for Computational Medicine.  Finally, Xiaoqin Wang may be the best mathematicians who has trained with me and is now recognized as the leader in the study of cortical auditory processing.

     In parallel with my scientific efforts I had the very good fortune of serving this University in a number of leadership roles.  Interestingly, until I began to put this piece together I considered that I had only one mentor as a leader, namely Richard Johns who was the first Director of the Department of Biomedical Engineering and a giant in the field.  I am currently preparing a history of this department and will have much more to say about Dick in that piece, including his role in my becoming the second Director of the Department.  On reading Nelson’s Overview to this history I realized that almost by osmosis I had gained an appreciation for the role of the scientist in leadership in the programs of the university beyond the lab bench.  Indeed, I really had two role models in program leadership at MIT – Nelson and Walter Rosenblith.  Walter founded the Communications Biophysics Group and went on to be a force in the growth of biomedical engineering at MIT and ultimately becoming the Provost of the University




Present position:

University Distinguished Service Professor

Department of Biomedical Engineering

Johns Hopkins University