Schedules of Reinforcement with Skinner - PDF

Document Sample
Schedules of Reinforcement with Skinner - PDF Powered By Docstoc
					JOURNAL OF THE EXPERIMENTAL ANALYSIS OF BEHAVIOR                       2002, 77, 303–311          NUMBER   3 (MAY)

                                      SCHEDULES OF REINFORCEMENT
                                             WITH SKINNER
                                                     C. B. F ERSTER
                                               THE AMERICAN UNIVERSIT Y




   To tell about the pigeon laboratory at Har-                   only a small amount of behavior was record-
vard during the period I was there, I must                       ed. Experiments, seldom more than an hour
first describe the state of the science at Co-                    long, took place just before the rats were fed.
lumbia while I was a graduate student there.                        I was a third-year graduate student when,
The comparison provides a ‘‘before and af-                       hearing of the chance to work with Skinner,
ter’’ which will help me to communicate what                     I made an appointment to go to Cambridge
happened at Harvard during the years when                        for an interview. I took the midnight train to
B. F. Skinner and I worked on Schedules of Re-                   Boston and wandered around Har vard
inforcement.                                                     Square nervously from six in the morning un-
   The pigeon lab was already operating at                       til what I thought would be a respectable
Harvard in the fall of 1950 when word                            hour to appear at Skinner’s office. The inter-
reached Columbia that Skinner was looking                        view was easy once I got there. We had a coke,
for someone to assist him. Columbia was the                      he showed me some of the equipment in the
obvious place to look because there was so                       lab, and I was scarcely aware of at what point
much activity and excitement there about op-                     I knew that I was to come to work in Febru-
erant conditioning and a functional analysis                     ary. Within two hours I was on my way back
of behavior. Keller and Schoenfeld had just                      to New York and Skinner was back in his of-
completed Principles of Psychology, and the in-                  fice writing.
troductory course at Columbia was in full                           I had finished almost all of my course work
swing. We learned of the impact of a labora-                     at Columbia and was doing exploratory ex-
tory science of behavior on biological sci-                      periments on chaining. A retractable lever
ence, pressing community problems such as                        came into the cage when the rat pulled a
mental illness, education, and rearing chil-                     chain suspended from the ceiling. These
dren for a better life and a basic understand-                   were called exploratory experiments because
ing of human nature. Everyone had condi-                         they preceded the real experiment; because
tioned a rat, read Walden Two, and most were                     only one or two animals were used; because
impatient for a chance to try out a science of
                                                                 the procedures as well as the apparatus were
behavior. Some students fantasied a new In-
                                                                 constantly adjusted during the experiments;
stitute for Operant Behavior with buildings,
                                                                 and because it was impossible to know in ad-
equipment and full-time research. Others
                                                                 vance what was going to happen. Experi-
dreamed of an actual planned community
                                                                 ments in which an animal served as its own
modeled after Walden Two where the products
of laboratory research could be lived and ap-                    control were not quite acceptable at Colum-
plied. I’m sure many of today’s laboratories                     bia as yet.
exceed what were then our wildest expecta-                          Because Skinner wanted me to be in Cam-
tions. For in those days the typical operant                     bridge by the first of February, completing a
experimenter either manually operated                            Ph.D. dissertation before I left posed a large
switches in a darkened room, or programmed                       problem. At Columbia, getting a thesis topic
a half dozen relays cannibalized from vending                    approved was quite an involved process. First
machines. A pressing instrumentation prob-                       there were informal tests with faculty and stu-
lem was a reliable pellet dispenser, but re-                     dent. These consisted of discussions in the
cording problems were not serious because                        corridor with other graduate students and vis-
                                                                 its to several professors’ offices. The pro-
  From: Dews, P. B. (Ed.). (1970). Festschrift for B. F. Skin-
                                                                 posed experiment was received very differ-
ner (pp. 37–46). New York: Irvington. Reprinted with the         ently in different places. First, there were the
kind permission of Irvington Publishers, New York.               kind ear and probing questions of Professor

                                                             303
304                                      C. B. FERSTER

Keller who listened gently until there was no
more time. Later in the day of my ‘‘test’’ with
Keller, I found myself redoing the plan as I
tried to explain answers to his questions. Oth-
ers were not so gentle. A thesis plan also had
to pass muster of a formal departmental
meeting. Since I had not even gotten by the
informal test when I returned from my inter-
view with Skinner, it was clear that the usual
process was much too long and labored to
meet Skinner’s deadline, so chaining was put
aside, for the time being. Instead I formulat-
ed a hypothesis, built equipment, ordered fif-
ty genetically controlled Wistar rats, and test-
ed the hypothesis that a stimulus present
during conditioning would influence the
number of performances the rat would emit
when reinforcement was discontinued.
   The laboratory was in operation when I ar-
                                                     Fig. l. The relay programming and recording equip-
rived in Cambridge. Several graduate stu-          ment for the chained VI FI equipment. There were three
dents were preparing pigeon demonstrations         parallel pairs of bars on the tables into which the relay
for Skinner’s introductory course and there        and other control panels were fastened. The pigeon box
were several pigeon boxes with relay control       rested on the shelf below the table and the recorder sat
                                                   on a bridge across it.
apparatus. The newest behavioral discovery
was aperiodic or random-reinforcement (var-
iable-inter val) schedules which were pro-         mations of the thesis data to make an analysis
grammed by a metal phonograph recording            of covariance possible. Fortunately, the in-
disc covered with plastic. A slow motor turned     crease of the frequency of rewards of the for-
the disc. A wiper, operating on the outside        mer activity could be paced with the early
groove of the disc, like the recording arm of      completion of the latter.
a phonograph, picked up an electric pulse             While I was building the variable-ratio pro-
whenever the covering was scraped away. The        grammer, I spent the rest of the time during
distribution of scratches around the periph-       my first week in the lab exploring all of the
ery of the disc made the variable schedule,        parts in the drawers and cabinets, reorganiz-
and the number of scratches determined the         ing them according to my own habits, and
average interval of reinforcement.                 labeling them to my custom. I found a large
   When I reported to Skinner on my first day,      store of small electrical and mechanical parts:
he showed me parts and plans for a variable-       springs, phosphor bronze, string, glue, bake-
ratio programmer for Elinor Maccoby, then          lite, plexiglas, surplus relays, assortments of
completing her Ph.D. in the Social Relations       capacitors and resistors, cable clamps, lacing
Department. She needed the equipment for           cord, soldering supplies, bits and pieces of
an experimental thesis with pigeons, which         rubber and plastic, a large box of accumulat-
extended the experiments on random-inter-          ed nuts and bolts, small odd pieces of metal,
val to random-ratio schedules as they were         wire and cable, surplus electronic relays and
then called. The programmer, already de-           electrical devices that could be disassembled
signed by Skinner, was to be built from a step-    for parts, cardboard and paper, motors and a
ping switch much along the principle of the        host of the miscellany that seems to come in
motor-driven disc used to arrange a variable-      handy at odd times for unexpected uses. Ger-
interval schedule.                                 brands’ machine shop down the hall had
   My first months in the pigeon lab were a         seemingly endless drawers of bolts and nuts,
strange contrast of days adjusting equipment       cotter pins, hex nuts, brass nuts, steel nuts,
and experimental procedures for one or two         lock washers, Allen nuts, Phillips head,
pigeons, and nights at the calculating ma-         round-head, flat-head, oval-head, and spline-
chine trying log and trigonometric transfor-       head screws, wood, machine and metal
                     SCHEDULES OF REINFORCEMENT WITH SKINNER                                 305

screws, brass and steel washers and lock wash-    paper, and the Leeds-Northrup glass reser-
ers. All were stocked in every length, diame-     voir pen solved the problem of providing a
ter and thread. Further down the hall were        reliable ink line. Twelve recorders were hard-
the psycho-acoustic laboratory shops, direct-     ly completed, however, when the experimen-
ed by Rufus Grason, where resistors, capaci-      tal sessions lengthened because we learned
tors and all of the rest were found in the same   how to sustain high rates of performance with
rows of cabinets and in the same profusion        our pigeons for ten-hour sessions or more.
of varying wattages, resistances and capaci-      Experiments which recorded two or three
tance values, accuracy levels and shapes. The     thousand pecks at the start of our research
pigeon laboratory already had a room dedi-        soon required 100,000 or more pecks to be
cated as a shop with a drill press, two long      recorded during a single experimental ses-
work benches and the usual assortment of          sion. For a long while, I spent much of my
hand tools.                                       time replacing and repairing rotary solenoids
   Instrumentation was easy and natural, and      which lasted only a few hundred thousand
all components for innovative apparatus con-      operations. The discovery of the Automatic
struction were immediately at hand. Herbach       Electric stepping switch mechanism, which
and Rademan sold surplus electrical equip-        stood up to the billions of pecks which were
ment by mail order catalogue, even then, and      recorded on each instrument, freed much
it was the tradition then as it is now to scan    time and energy for other purposes.
the catalogue each month to buy parts and            It was an enormous source of support to
devices that ‘‘might be useful sometime.’’        move into a laboratory which Skinner had al-
One of the first steps for solution of an in-      ready arranged and stocked. A beginner faces
strumentation problem was always to look          so many anxieties and new problems that
through the drawers and cabinets to see what      without this support I doubt that there would
suggested itself.                                 have been enough energy both for producing
   The physical arrangements of the labora-       the physical arrangement of shops, supplies
tory, the supplies, the equipment and the         and equipment that is so critical in order to
shop were important factors in determining        be able to do innovative research, and for ac-
the kind of research that went on. There were     tually carrying out an experiment. The pi-
sufficient parts immediately on hand for con-      geon lab set the pattern for all of my later
struction to begin the moment an experi-          laboratories. For example, I always saved and
ment required new instrumentation. Skinner        carried with me a large box of nuts, screws,
usually built the first model from what was on     hardware, assorted junk and parts and devic-
hand, seldom waiting because parts needed         es that accrued when the bench top was swept
to be ordered. The prototype was usually          and that ‘‘might be useful someday.’’ For al-
makeshift and not quite reliable enough, but      most ten years, I carried around a 244 pole
it served long enough to prove itself. By then    stepping switch (purchased from surplus for
there had been time enough to order proper        a dollar or two) before I finally threw it out.
parts and to build a well-constructed model.         During my first months with Skinner and
   Probably the most serious and pressing in-     the pigeon lab, I learned a great deal about
strumentation problem we faced was the de-        how to run a laboratory, design and interact
sign of a reliable cumulative recorder, and       with experiments, and think through instru-
the construction of enough of them to ser-        mentation and research problems. The teach-
vice the large number of experiments that         ing process was so natural but subtle, that I
ran concurrently. Even more recorders were        had no awareness that I was learning any-
needed because we developed the habit of          thing new or that the research we were car-
using several at once on a single experiment,     rying out was a departure from the existing
as in a multiple schedule, to treat the data      body of knowledge. It was not until months
during recording rather than by numerical         later, around the time we gave our first paper
manipulations later. The first model used a        on schedules (Skinner on mixed and I on
Ledex rotary switch to drive the pen on the       multiple schedules), that I began to con-
performance scale. By this time the paper         sciously sense that our work was extending
drive worked well, using a typewriter platen,     and departing from the current literature.
with its associated mechanism for holding the        I think that part of the reason for the del-
306                                        C. B. FERSTER

icacy and smoothness of the learning process         on the large ratio on a continuing basis. The
was Skinner’s natural style of creating the          idea of a stable state experiment ended the
conditions which allowed learning to take            discussion and began the experiment.
place rather than teaching or telling me                Thereafter our discussion about experi-
things. In retrospect, my personal experience        ments occurred at ‘‘rounds,’’ usually the first
in the spring of 1950 contained many exam-           thing each morning when we toured the lab-
ples of how the laboratory environment con-          oratory to look at the harvest from the day
tained supplementary and collateral variables        and night before. This was when we discov-
which supported my behavior so long as they          ered the apparatus failures, particularly in cu-
were needed, and which faded out as I de-            mulative recorders, which were so frequent
veloped my own ways of providing the same            and discouraging during the early days of the
support. The first task assigned to me in the         pigeon lab. Failures of programming and re-
laboratory, constructing the random-ratio            cording sometimes set an experiment back
(variable-ratio) programmer for Elinor Mac-          the days or even weeks that were necessary to
coby’s experiment, served to move me into            recover the baseline. On these occasions
action at my own pace and with support. The          Skinner always commented on what caused
device had already been designed and the             the failure and we discussed changes that
components were at hand. Although it was a           would reduce the likelihood of failure in the
simple device which I could now complete in          future. Although both of us felt keen disap-
an hour or two, I spent two or three days pok-       pointment in the delay in the experiment,
ing away at it, redoing it several times and at      our remarks always concerned possible re-
the same time getting used to the color of the       medial action rather than the current failure
walls and the other features of my new work-         of the experiment (or perhaps the experi-
ing space. No one checked on the progress            menter). Rounds took thirty minutes to an
of the device during these several days and          hour, depending on the press of other activ-
the most important consequence of finishing           ities, and it was a lively activity with much roll-
it was its installation in the control circuits of   ing and unrolling of cumulative records,
Elinor Maccoby’s pigeon experiment.                  comment on what had happened, ooo’s and
   I began two experiments as soon as I had          aaah’s about a new degree of orderliness and
straightened out the cabinets, swept the floor,       planning of the next procedure. Conversa-
and built the random-ratio programmer. One           tions did not include references to who had
was variable-interval baseline with a time out       pulled the switch, first mentioned the idea for
between reinforcement and the performance            the experiment, built the apparatus, or pre-
that preceded. I don’t remember now why I            dicted the outcome of the experiment. It
did this experiment except perhaps it was the        took almost a year before I stopped predict-
only one I could think of. Fortunately, no one       ing. The pigeon really did know best what it
asked me. I was surprised that the delay be-         was he was likely to do and the conditions
tween the performance and the operation of           under which he would do it. Free of Skinner’s
the food magazine did not decrease the fre-          praise, I was also free of his censure, real or
quency of pecking, so I continued to extend          imagined. Yet I still had the advantage of an
the delay period. No one noticed this exper-         inspiring model I could observe, whose be-
iment for some time. Skinner suggested the           havior prompted me to greater accomplish-
second experiment. He thought we should              ments. I remember how easy it was for me to
do something with ‘‘ratios’’ and we talked           talk with Skinner about experiments and psy-
about how number of pecks could control the          chology in general. I sometimes wondered
bird’s behavior in a ratio schedule and I sug-       how it was that this young man could face the
gested that we reinforce for a long time on          feeling that almost anything he could do
FR 50 and see whether we could see the evi-          Skinner could do better. I think the reason I
dence of the reinforcement after fifty perfor-        could contribute my portion without uneasi-
mances, when reinforcement was discontin-            ness was that I was never evaluated, rewarded
ued. Skinner suggested a random alternation          or punished; nor was my behavior ever mea-
between a small and a large fixed-ratio sched-        sured against his. I found Skinner’s reper-
ule (two-valued ratio) so that the control by        toire an ever-present source of prompts and
the smaller ratio would show up in the effect        supports which I could use whenever I was
                         SCHEDULES OF REINFORCEMENT WITH SKINNER                                         307




 Fig. 2.   The room where the graphs were pasted up and where schedules of reinforcement were written.



able to. It was a very fortunate young man              ed too fast to be digested during morning
from Columbia who had an opportunity to                 rounds, there were planning sessions which
carry out his work with so much intellectual            were always great fun and very exciting. It was
and practical support and with such exciting            during these sessions that I learned the value
chances to ‘‘brainstorm.’’ Nor was it a small           of large sheets of paper which we used to aid
measure of support to be able to watch B. F.            our thought and to chart our progress. Every
Skinner in the laboratory designing new in-             experimental result appeared as an entry
struments, or to be able to turn over a prob-           someplace on paper about ten square feet in
lem to him.                                             size. The theoretical structures and program-
   But I give the reader the wrong impression           matic aspects of our work appeared as the
if I suggest that there was no reinforcement            spatial arrangement of the headings. Later,
for the results of experimentation other than           these headings were to appear as chapter and
the actual behavior generated in the birds.             subchapter titles in Schedules of Reinforcement.
There were many personal, natural conse-                Each entry prompted rearrangements of the
quences of completing a successful experi-              theoretical pattern and suggested new exper-
ment. A successful experiment led to conver-            iments and programs which in turn prompt-
sations about the data, the new devices we              ed further rearrangements of the data. The
could build, the new experiments that had to            interactions between these theoretical exer-
be started and the new ways we could orga-              cises and changes in ongoing experiments in
nize our past experience from the laboratory.           the laboratory were continuous and consti-
   When we discovered a new degree of or-               tuted an important reinforcer. Although al-
derliness or an unexpected but rewarding re-            most all of the entries on the large sheet of
sult on morning rounds, there was always                paper were in Skinner’s hand, I took part sig-
much excitement and talk about where the                nificantly by providing facts and prompts, and
experiment might go next and how to man-                by reacting to the patterns which emerged.
age the equipment for the next experiment               Mostly we arranged and rearranged the find-
that was burning to be done because of the              ings and procedures we had discovered and
new result. When new discoveries accumulat-             worked with, and reacted to the new experi-
308                                      C. B. FERSTER

mental procedures that were suggested by the       called ‘‘great’’ or ‘‘bad’’ or anyone being giv-
arrangements. Skinner did most of the talk-        en credit for doing something especially use-
ing, just as he did most of the writing, but       ful or valuable. Some experiments led to fur-
even when I was silent, I was always intensely     ther planning, new apparatus, exciting
involved because he generally spoke for both       conversations, new theoretical arrangements
of us. It was sometimes like playing a well-       of data and procedures or a rush to tell ev-
tuned organ that could play itself if the right    eryone about them, while others enabled less
key were pressed and a properly reactive lis-      behavior of this kind. I don’t know whether
tener were present. We came to share such          Skinner was conscious of the lack of personal
an extensive repertoire that not everything        praise in interpersonal relations in the labo-
had to be said by each person. When one per-       ratory. I certainly was not. My behavior was
son spoke, the other frequently could have         generated by the natural reinforcement of
said the same thing a few minutes later, or he     the laboratory activity. But some of the grad-
might have been so close to saying it that a       uate students found the absence of personal
small amount of supplementary stimulation          support difficult.
was enough to produce the same perfor-                Recently a distinguished psychologist, who
mance.                                             had come to Harvard when he was a student
   Our interaction as speakers and listeners       to study under Skinner in the pigeon lab, re-
was an apt illustration of the verbal process      minded me of an incident which illustrated
described in Verbal Behavior where Skinner         the personal styles around the laboratory
wrote, in the chapter on supplementary stim-       then. After completing the professional sem-
ulation, about strengthening the behavior of       inar, the main classroom experience in the
the listener. To a degree we were in the same      Har vard curriculum, he appeared before
position as speakers and listeners as the pro-     Skinner saying that he was ready to do re-
verbial prisoners who told jokes by code num-      search in the pigeon laboratory. He asked
bers, indicating stories that they already knew.   what he should start on. The conversation
‘‘When no one laughed when the taps on the         was awkward; the student did not receive the
pipe indicated Joke Number Ten, it was ex-         kind of support and encouragement that he
plained to a visitor that this prisoner didn’t     expected, especially since he had come to
tell jokes very well.’’                            Harvard for the single purpose of working
   One of the unspoken rules of these think-       under Skinner. Finally, in the heat of frustra-
ing, planning and theory sessions was to           tion, he complained, ‘‘Aren’t I even going to
avoid criticism or contradiction. The perfor-      get a pigeon box?’’ This remark galvanized
mances which occurred were delicate and of         Skinner who dashed out of his office into the
such high frequency that criticism or contra-      pigeon laboratory around the corner shout-
diction produced a large and sudden change.        ing, ‘‘Charlie, he needs a pigeon box,’’ and
I learned that there were natural consequenc-      left. I dutifully took one of the unused Sears
es of unproductive or incorrect suggestions        and Roebuck ice chests we used as the shells
or formulations which shaped them and al-          for pigeon experimental spaces, handed it to
tered their frequency. Any thought was fair        him, and left. The student was then left with
game and the worst that could result from an       the problem of assembling all of the com-
error, an inept or an inappropriate sugges-        ponents and constructing the equipment he
tion was that it would be ignored or have no       needed. Although neither Skinner nor I re-
consequences in prompting or aiding other          membered the incident, the anger and dis-
activities. If one or the other of us had strong   appointment could be detected after all these
behavior which was not shared, a record was        years. Yet he went on to complete an experi-
made on the work sheets, apparatus was built,      ment which was an original departure from
or an experiment was started, but there was        the main experimental program of the pi-
no requirement for both to participate or          geon laboratory and which still remains in
speak about it. In some cases an unshared          the literature as a base for much research and
line of work disappeared because the perfor-       thinking. I don’t know whether this particular
mances it led to were not useful. In other cas-    student would have gone on to do the same
es it persisted successfully.                      valuable work had Skinner supported his
   I don’t remember any experiment being           ideas personally, or had I given him equip-
                         SCHEDULES OF REINFORCEMENT WITH SKINNER                                              309




  Fig. 3. The members of the pigeon staff meeting posed for a picture toward the end of one of the meetings. Left
to right are B. F. Skinner, Clair Marshall, W. H. Morse, R. J. Herrnstein, Tom Lohr, Nate Azrin, and James Anliker.
Murray Sidman was visiting. Others attending frequently were Peter Dews, Ogden Lindsley, and Michael Harrison.


ment and supervised his day to day work in                 and usually a prolonged discussion at the
an experiment related to ours. But I think                 end. When the prolonged discussion oc-
many others would have become pale imita-                  curred before all of the data in the experi-
tions of Skinner and Ferster rather than the               ment had been covered, we continued with
original, imaginative, aggressive scientists               the same bird the next time. Later, when stu-
they did become.                                           dents and others had experiments under way,
   The pigeon staff meeting where we re-                   they brought in their data in a similar way.
viewed current experiments with graduate                   The presentations of the pigeon staff meet-
students and others was one of the traditions              ings were seldom a summarized formal re-
of the laboratory. We met, usually weekly, in              port of what had happened in the experi-
the seminar room, reviewing and talking                    ment, but rather an informal scanning of the
about one or two birds. Ogden Lindsley in-                 raw data. I think the feeling of participating
troduced the symbol of the pigeon feather at               in the formulation and identification of the
this time when he made up a sign with a pi-                results contributed to strong interest in the
geon feather that was hung on the bulletin                 meetings.
board on days that there were meetings. Lat-                  Most of the research for schedules of re-
er, he sent a white feather to the charter sub-            inforcement was completed by 1953, when we
scribers of JEAB. The seminar presentation                 began to plan a written report. As more of
consisted of either Skinner or me going                    my time shifted to organizing the data and
through the cumulative records, a day at a                 writing, the laboratory was turned over to
time and a bird at a time, reacting to the                 Morse, Herrnstein, and others at Harvard. By
small details of the results. The substance of             the end of 1954, Skinner and I were writing
the meetings was a very detailed examination               full time. The first problem we faced was how
of the results, even if some participants had              to present the large amount of data we had
to learn to read cumulative records upside                 collected, not only from long experimental
down. There were frequent interruptions                    sessions and protracted experiments, but also
with questions, suggestions, or comments,                  from a large number of separate experi-
310                                     C. B. FERSTER

ments. By the end of our research there were      the afternoon when we were working espe-
about a dozen separate experiments in prog-       cially well or when the data were especially
ress. The problem was to compromise be-           interesting, but our recently acquired data on
tween the need to report enough detail of         fixed-ratio performances convinced us to
our descriptive experiments and the need to       seek a work schedule that kept our perfor-
reduce the bulk of the thousands of feet of       mance at maximum frequency for the period
cumulative curves. Three inventions—the           we were actually writing. The procedure
collapsed record, a razor blade, and a stan-      worked very well. There were no warm-up or
dard cardboard stock thirty inches long—got       inactive periods in the writing room. Natu-
the final report under way. Collapsing the         rally we did not write elsewhere nor did we
record by cutting out blank paper along the       converse about outside matters nor do any-
time axis allowed us to present as much as        thing but work on schedules of reinforce-
fifty to seventy-five thousand pecks in a single    ment so long as we were in the writing room.
figure; the razor blade made it possible to cut    At times the pace of the writing was so in-
the records swiftly and effortlessly; and the     tense, and rewarding, that we began to con-
card stock permitted a storage system that was    trol our outside activities in the fear that they
easily handled. Skinner usually pasted rec-       might compete with or decrease the frequen-
ords on the cards while I cut excerpts from       cy of writing and graph-making. Bridge, chess
the folders. Decisions about what to excerpt      and late social evenings were out.
were made quickly, usually without much dis-         The professional record speaks for the ‘‘be-
cussion because we were both so familiar with     fore and after’’ of the pigeon laboratory.
the records. Skinner took justifiable pride in     There were personal results too, however. B.
his skill and speed with a razor blade. The       F. Skinner has already written his feelings
ultimate test was to cut on several layers of     about our collaborative activities. For my part,
paper, piercing an exact number of layers.        besides the satisfaction of a very rewarding
The figures were pasted up, experiment by          association, I remember most of all how I
experiment, and the categories under which        came away from the pigeon lab with a firmly
the figures were filed turned out to be the         developed attitude toward discovery and un-
chapter and section headings of the book.         known things.
On our best days we could do thirty figures,          There is a fear of the unknown in research
but this was a grueling pace which could not      just as there is a fear of dealing with new peo-
be kept up. Once the figures were completed,       ple. We approach a new problem or a new
the writing turned out to be a relatively rou-    person with a repertoire that comes from our
tine job of describing the main features of a     past experience. When we are successful, the
record and indicating procedures. It became       new person or problem differentially rein-
clear very early in our writing that we could     forces our existing repertoire, and we acquire
not discuss the experiments theoretically or      a new means of dealing with a new environ-
spell out the implications for the casual read-   ment. Unfortunately, the old repertoire often
er.                                               continues without significant influence of the
   We worked slowly at first, but the need to      new contingencies. Such a repertoire is called
finish before my scheduled departure in June       compulsive or neurotic by clinicians. The an-
1955 led us to organize our environment and       alogue, in research, is the experimenter who
to develop several ways of self-management.       is controlled primarily by the social and pro-
All our work was done in a room dedicated         fessional consequences—his colleagues’ ver-
to writing and not used at other times. Inter-    bal behavior—and to a lesser degree by the
ruptions were the first problem, which we          behavior he produces and measures in his ex-
handled by a decision not to take phone calls.    periment. I don’t think we were ever worried
When visitors appeared at the door, we rou-       in the pigeon lab that we would have nothing
tinely stepped in the corridor to speak with      to show for our time or that an experiment
them briefly. The frequency of interruptions       would waste time and money. The pigeon lab
became very low and the writing room came         was a place where an unknown problem be-
to control our behavior. Usually we began be-     came an occasion which led to discovery and
fore nine and stopped by lunch time. There        accomplishment rather than a cause for wor-
was frequently a temptation to continue in        ry. The more a new situation could be seen
                     SCHEDULES OF REINFORCEMENT WITH SKINNER                                       311

as very different from our current experi-        tell the funny story, give affection artfully
ence, the more it signalled an experiment         (manipulate the environment and observe
that would bring results which we valued. Per-    the behavior of the animal) and the world
haps my experience in the pigeon lab with B.      will respond in kind with prestige, money, so-
F. Skinner prompted me to write in 1958: ‘‘A      cial response, love (and recognition for sci-
potential reinforcing environment exists for      entific achievement).’’1
every individual, however, if he will only emit
the required performances on the proper oc-         1 Ferster, C. B. (1958, December). Reinforcement and
casions. One has merely to paint the picture,     punishment in the control of human behavior by social
write the symphony, produce the machine,          agencies. Psychiatric Research Reports, 101–118.