Observing Ben Wyckoff: from Basic Research to Programmed Instruction and Social Issues Rogelio Escobar and Kennon A

The Behavior Analyst 2011, 34, 149–170 No. 2 (Fall)

Observing Ben Wyckoff: From Basic Research to Programmed Instruction and Social Issues Rogelio Escobar and Kennon A. Lattal West Virginia University

L. Benjamin Wyckoff’s seminal contributions to both psychological theory and application are the subject of this review. Wyckoff started his academic career as a graduate student at Indiana University, where he developed the observing-response procedure under the guidance of B. F. Skinner and C. J. Burke. At the University of Wisconsin–Madison, Wyckoff refined his mathematical theory of secondary reinforcement. This theory was the impetus for his creation of an electronic simulation of a rat running a T maze, one of the first ‘‘computer models’’ of learning. Wyckoff next went to Emory University, leaving there to help create two of the most successful companies dedicated to the advancement of programmed instruction and teaching machines: Teaching Machines, Inc. and the Human Development Institute. Wyckoff’s involvement in these companies epitomizes the application of basic behavior-analytic principles in the development of technology to improve education and human relationships. The emergent picture of Wyckoff is that of a man who, through his research, professional work in educational applications of behavioral principles, and active involvement in the civil rights movement of the 1960s, was strongly committed to applying behavioral science to positively influence human behavior change. Key words: Benjamin Wyckoff, observing responses, conditioned reinforcement, teaching machines, programmed instruction, human relationships, diversity issues, client-centered therapy

The first author is now at the National Lewis Benjamin Wyckoff, Jr., Ben Autonomous University of Mexico and was to friends and colleagues, was born in supported by a postdoctoral fellowship award- 1922 in Niagara Falls, New York. He ed by the PROFIP of DGAPA-UNAM. This paper was possible thanks to the recollection died at age 84 on June 6, 2007, at St. of many contributors to whom the authors are Simons Island, Georgia. He is per- indebted: Estelle Wyckoff and Andrew Weisk- haps best remembered among behav- off (Benjamin Wyckoff’s family); John Cotton ior analysts as the creator of the and Richard Atkinson (Indiana University); observing-response procedure. In his Lewis Gollub (Harvard University); Bernard Pyron and H. Philip Zeigler (University of doctoral dissertation he noted that, Wisconsin); Stephen Kendall, Donald J. Levis, ‘‘We shall adopt the term ‘observing and William Deckner (Emory University); response’ (RO) to refer to any re- Donald Tosti, Roger Addison, Roger Stein- sponse which results in exposure to horst, Clifton Chadwick, and David Shields (Teaching Machines Inc.); Jerome Berlin, the pair of discriminative stimuli Gene Ruyle, and Schlomo Friedlander (Hu- involved’’ (Wyckoff, 1951b, p. 1). man Development Institute); Lizette Royer Many behavior analysts are famil- (Archives of the History of American Psy- iar with the above statement, because chology); Constance Carter (Library of Con- gress); and Kathleen Shoemaker (Emory Wyckoff’s observing-response proce- University Manuscript, Archives, and Rare dure is still widely used for investi- Book Library). gating conditioned or secondary re- Address correspondence to Rogelio Esco- inforcement (hereafter, conditioned bar, Laboratorio de Condicionamiento Oper- ante, Facultad de Psicologı´a, Universidad reinforcement) (e.g., DeFulio & Nacional Auto´noma de Me´xico. Av. Univer- Hackenberg, 2008; Escobar & Bru- sidad 3004, Col. Copilco-Universidad, Me´xi- ner, 2009; Fantino & Silberberg, co, D.F. C.P. 04510 (e-mail: rescobar@ 2010; Lieving, Reilly, & Lattal, servidor.unam.mx), or to Kennon A. Lattal, Department of Psychology, West Virginia 2006; Pessoa, Huziwara, Perez, En- University, Morgantown, West Virginia demann, & Tomanari, 2009; Shahan 26506 (e-mail: [email protected]). & Podlesnik, 2008).

149 150 ROGELIO ESCOBAR & KENNON A. LATTAL

Wyckoff’s observing-response pro- libertarians. Wyckoff’s journey cedure offered a means of demon- through behavior analysis may be strating the acquisition and main- likened to the journey described by tenance of an operant, the only Skinner in ‘‘A Case History in programmed consequence of which Scientific Method’’ (1956): an induc- is the production of discriminative tive approach in which the scientist stimuli, without altering the rate of follows the data wherever they hap- primary reinforcement (for reviews, pen to lead. This approach in Wyck- see Dinsmoor, 1983; Fantino, 1977). off’s career is an instructive counter- The procedure typically consists of point to the numerous examples of two components alternating random- gaps among basic research, applied ly. During one component, responses research, and service delivery aspects are reinforced according to an inter- of behavior analysis (e.g., Michael, mittent schedule (e.g., a variable 1980; Pierce & Epling, 1980). The interval); during the other compo- purpose of this historical review is to nent, responses are not reinforced. A trace the course of Wyckoff’s journey. response on a second operandum briefly produces the stimulus associ- EARLY ACADEMIC LIFE AND ated with the ongoing component INDIANA UNIVERSITY (e.g., a green light during the reinforcement component and a red light Wyckoff attended Antioch College during the extinction component). from 1941 to 1947, having completed The observing-response procedure is 5 years of premed courses. Following one of most useful methods for graduation, he applied to Harvard studying conditioned reinforcement Medical School, but he changed his (see, e.g., Williams, 1994, for a mind and pursued a PhD in psychol- review). ogy instead. His wife, Estelle, noted A year after he finished his doc- that he learned of Skinner’s work and toral studies, the theoretical Part 1 of decided to study under his supervi- his dissertation was published in sion at Indiana University (E. Wyck- the Psychological Review (Wyckoff, off, personal communication, March 1952). The Social Sciences Citation 22, 2009). Index indicates that this article has According to J. Cotton (personal been cited in over 250 other research communication, February 24, 2009), and theoretical articles in psychology. who became friends with Wyckoff at The procedural details of the observ- Indiana, he and Wyckoff took an ing-response procedure, which con- introductory course on the topic of stituted Part 2 of the dissertation, the experimental analysis of behavior were not published until 1969, how- with Skinner, circa 1947. The goal of ever, in the volume entitled Condi- the course was to familiarize gradu- tioned Reinforcement edited by Hen- ate students with Skinner’s 1938 dry (Wyckoff, 1969). book, The Behavior of Organisms. Designing the observing-response Even before Skinner arrived in procedure is alone sufficient to secure Bloomington in 1945, Indiana Uni- Wyckoff mention in any history of versity was fertile ground for behav- the experimental analysis of behav- iorism and behaviorists. Under the ior. That seminal work, however, influence of J. R. Kantor, W. N. perhaps overshadowed his less well- Kellogg, and R. C. Davis, graduate known work in behavior analysis students were well acquainted with from the mid-1960s that invites the learning theories. As part of the task attention of an even broader audi- to rebuild the Department of Psy- ence of applied behavior analysts, chology after the war, Kantor at- educational technologists, educators, tracted Skinner to Indiana University computer enthusiasts, and even civil (Hearst & Capshew, 1988). In the OBSERVING BEN WYCKOFF 151 following years, the addition to the ing with him on experiments on faculty of W. O. Verplanck, W. K. discrimination in which reinforce- Estes, S. W. Bijou, D. G. Ellson, C. J. ment was contingent on two proper- Burke, W. O. Jenkins, and I. J. ties of the stimuli. Skinner’s notes Saltzman favored the melding of (1983, p. 415) suggest that the ratio- different behavioristic approaches to nale for these experiments was to the study of learning. study concept formation. The origins of behavior analysis From his interactions with Skinner are closely tied to Bloomington, at Indiana, Wyckoff earned an invi- Indiana. The first and second Con- tation to work with Skinner at ferences on the Experimental Analy- Harvard during the summer of sis of Behavior were held there, at 1950. In Cambridge, Wyckoff imme- Indiana University, in 1947 and 1948 diately became good friends with (see Dinsmoor, 1987). Wyckoff may Charles Ferster, with whom he spent be seen in a photograph of the many evenings playing music, some- attendees of the first conference (see times joined by Skinner. Wyckoff p. 456 of Vol. 5 of the Journal of the was a fine pianist (E. Wyckoff, Experimental Analysis of Behavior personal communication, June 18, [JEAB]; see also Hearst & Capshew, 2009). 1988; Skinner, 1979). Wyckoff also According to Skinner (1983, pp. 37, attended the third conference at 415), during his time at Harvard, Columbia University, where a pho- Wyckoff continued their research on tograph (Dinsmoor, 1990) shows him concept formation that was started at standing next to another graduate Indiana. It was supposed to employ a student, Lloyd Homme, who also single-key procedure in which rein- was one of his closest friends in forcement would be contingent on Bloomington. In the ‘‘Conference two properties of the stimuli. Skinner on the Experimental Analysis of wanted Wyckoff to present four Behavior Notes’’ dated February 1, stimuli (red flashing, red continuous, 1949, they appeared as coauthors of green flashing, and green continuous) one presentation on the effects of to test for the formation of the deprivation on feeding behavior. As concepts of color and flashing or will be described below, their collab- not flashing. Wyckoff, however, ap- oration and friendship continued for parently modified the procedure. many years. Skinner (1983) observed in his auto- Wyckoff made a good impression biography that ‘‘Ben Wyckoff was his on Skinner. For one thing, he shared own man and did the work his own Skinner’s passion for building appa- way’’ (p. 415). Although the details of ratus. For another, he and Skinner Wyckoff’s modifications to Skinner’s had mutual research interests. During design are lost, Skinner does refer to his time at Indiana, Skinner was Wyckoff’s experiments in ‘‘The Sci- working on the experiments on ence of Learning and the Art of matching to sample described in Teaching’’: Skinner (1950, 1979, p. 320). Skinner described how these experiments, In a special case first investigated by L. B. using pigeon boxes displaying three Wyckoff, Jr., the organism responds to one response keys, led to studies of stimulus where the reinforcement consists of the clarification of the stimulus controlling a attention and abstraction or concept second response. The first response becomes, formation. It is unclear on which of so to speak, an objective form of ‘‘paying these experiments Skinner and Wyck- attention’’ to the second stimulus. In one off collaborated. Based on Skinner’s important version of this experiment, as yet unpublished, we could say that the pigeon is notes in his autobiography (1983, telling us whether it is ‘‘paying attention’’ to p. 37), however, before Skinner left the shape of a spot of light or to its color. Indiana in 1948, Wyckoff was work- (Skinner, 1954, p. 89) 152 ROGELIO ESCOBAR & KENNON A. LATTAL

According to L. Gollub (personal band and confronted her. She, how- communication, February 1, 2008), ever, had no idea what Skinner was the experiments on ‘‘attention’’ to talking about (E. Wyckoff, personal shape and color, which Skinner communication, June 18, 2009). suggested were started by Wyckoff, When Wyckoff returned to Indi- continued at Harvard, and culminat- ana, he obtained his MA with W. K. ed in Reynolds’s (1961) paper on Estes. He was always grateful for the attention in the pigeon. The similarity high standards in writing imposed by between Reynolds’s procedures and Estes (E. Wyckoff, personal commu- the experiments that Wyckoff con- nication, March 22, 2009). Given the ducted in Skinner’s lab, in terms of influence of Estes and the training in the use of two-element discriminative quantitative analysis in his courses stimuli, is notable. with the recently hired Cletus J. Skinner (1950) stated that he had Burke, it is not surprising that his difficulty training pigeons to produce master’s thesis focused on the quan- and ‘‘attend’’ to a sample stimulus titative relation between the number in his experiments on matching to of lever presses during extinction and sample. It seems reasonable to con- the number of reinforcers delivered clude that Skinner started to discuss during conditioning (Wyckoff, 1950, with Wyckoff, while he was at 1951a). Wyckoff chose Burke to be Harvard, the idea of measuring his adviser for the doctoral disserta- responses that make contact with tion. Burke obtained his PhD in 1949 the discriminative stimuli, the line of at the State University of Iowa with research that culminated in Wyck- Kenneth Spence (Burke, 1949). He off’s doctoral dissertation. This sug- had solid training in statistics and in gestion is supported by the fact that a the Hull-Spence theories. Verplanck few years later Wyckoff acknowl- described Burke in 1950 as the best edged Skinner in his dissertation for statistician he knew, a very good and his help in the experiments that led to creative experimental psychologist, the observing-response procedure. and as an informal but effective Although Skinner invited Wyckoff teacher.1 to stay at Harvard to complete his Wyckoff’s doctoral dissertation re- studies, Wyckoff decided to return to flected well his two main influences: Indiana University. According to Skinner’s experimental analysis of Estelle Wyckoff, the primary reason behavior with free-operant proce- for returning to Indiana was that dures and Burke’s strong focus on Wyckoff knew that Skinner’s domi- quantitative methods, anchored in nant personality would compel him Spence’s theory of discrimination. to be devoted exclusively to Skinner’s reinforcement theory. For Wyckoff, Wyckoff’s observing-response proce- as he learned at Indiana, reinforce- dure allowed recording of a response ment theory was only a part of what that exposed an animal to discrimi- are generally referred to as learning native stimuli. Recording these re- theories, and he wanted to take what sponses, which were characterized at he thought was useful from all of the time as ‘‘attending’’ and were them. For example, as his subsequent described previously by Spence research showed, he was not pre- (1940), was crucial in the continuity pared to abandon the mathematical vs discontinuity debate in discrimina- descriptions of learning. To avoid a tion learning (see also Spence, 1945). confrontation with Skinner, however, when he had to explain his reasons 1 Letter of recommendation to Edwin B. Newman. January 10, 1950. William S. for leaving, he used his wife as an Verplanck Papers, Box M1878, Folder 4. excuse. Skinner was not pleased Archives of the History of American Psychol- about Estelle’s pressure on her hus- ogy, University of Akron. OBSERVING BEN WYCKOFF 153

Wyckoff primarily was interested in tor at the University of Wisconsin– Spence’s continuity theory, but he Madison. He and Estelle moved to took the explanation for the occur- Madison in 1951 (E. Wyckoff, per- rence of observing behavior from sonal communication, March 22, Skinner’s (1938) The Behavior of 2009). Psychology at the University Organisms: Discriminative stimuli of Wisconsin at that time was dom- can function as conditioned reinforc- inated by the strong presence of ers. This double influence of Skin- Harry Harlow. The department fa- nerian and Hullian-Spencerian tradi- vored empiricism with a general tions was central in the subsequent approach to psychology and a strong positive impact of Wyckoff’s observ- emphasis on statistics (B. Pyron, ing-response procedure. personal communication, January Wyckoff’s procedure was adopted 18, 2009). Wyckoff had a strong by researchers in the Hull-Spence background in statistics, but it is tradition, like Perkins’s group, which not clear how his interests in behav- studied observing behavior in mazes iorism and operant procedures were (Levis & Perkins, 1965; Lutz & received by the department. L. Gol- Perkins, 1960; Prokasy, 1956). In an lub (personal communication, Janu- E maze, for example, an observing ary 31, 2008) recalled that, in one of response is recorded when the animal Wyckoff’s visits to the Harvard runs to the side in which the color of Pigeon Lab when he was on the a small waiting chamber correlates faculty at Wisconsin, he mentioned with the presentation or with the that he was given a broom closet absence of food instead of running to under the flight of the stairs as his the side in which the color of the office. Whether this office assignment waiting chamber does not correlate was a matter of space, as it was in with the outcome. many departments of psychology, or The observing-response procedure something less benign is unknown. was used extensively by researchers in During his tenure at Wisconsin, the Skinnerian tradition like Kelleher Wyckoff published three articles that (1958), whose research on response- dealt with conditioned reinforcement produced stimuli appeared in the first and discrimination learning (Wyck- issue of JEAB, and Holland, where it off, 1954, 1959; Wyckoff, Sidowski, appears in his vigilance tasks (Hol- & Chambliss, 1958). In his 1954 land, 1957, 1958). Also, Killeen (1982) paper, he used a linear-operator suggested that Herrnstein’s experi- model to describe the change in reflex ments on conditioned reinforcement strength as a function of the reinforc- (Herrnstein, 1964) were at least par- ing value of the stimulus conditions tially inspired by Wyckoff’s proce- presented after the response. He dure. Furthermore, the quantitative based this model on the idea of the model describing observing responses, dual function of stimuli as discrimi- which was central in Part 1 of Wyck- native and reinforcing (Dinsmoor, off’s dissertation, appealed to behav- 1950; Skinner, 1938). An interesting ioral researchers who were developing detail of this paper is that he correct- and extending mathematical theories ed his mathematical model after testing it with an electronic model of learning and attention (e.g., Atkin- that he constructed (see also Cor- son, 1961; Atkinson & Estes, 1963; deschi, 2002, pp. 169, 183; ‘‘Good Bush, 1965; Norman, 1968). Turn Deserves Another,’’ 1953). The electronic model was based on UNIVERSITY OF timer circuits using vacuum tubes. It WISCONSIN–MADISON simulated choices in a T maze with or After completing his PhD, Wyck- without a previous observing re- off accepted a position as an instruc- sponse. In the model, Wyckoff sim- 154 ROGELIO ESCOBAR & KENNON A. LATTAL ulated four conditions that resulted discussion). He also coauthored a from choosing to observe or not, and number of publications with col- choosing the reinforced or the unre- leagues and students (Auerbach, inforced side of the maze. For Waisman, & Wyckoff, 1958; Pyron example, if a simulated observing & Wyckoff, 1961; Sidowski, Wyck- response occurred in combination off, & Tabory, 1956; Wyckoff & with choosing the reinforced side of Page, 1954; Wyckoff & Sidowski, the maze, the reinforcing properties 1955; Zeigler & Wyckoff, 1961). His of the stimuli on the next trial interest in apparatus again was dem- increased by changing the voltage in onstrated in his 1954 paper with a condenser via relay connections. Page. They described a grid for This increased the probability of a administering electric shocks to rats. simulated observing response on the One problem with previous devices following trial. The experimenter was that the rats could avoid the recorded the simulated responses by shock by standing on two bars that illuminating different sequences of were on the same side of the electric four lights that represented the two circuit. A few years before, Skinner responses in the two choice phases2 and Campbell (1947) designed the (see also Minsky, 1961). first device that avoided the problem Earlier electronic models of learn- by repeatedly changing the polarity ing in mazes had been developed of the bars (i.e., a shock scrambler). (e.g., Ross and Smith’s 1937 ‘‘robot Rats could still detect the pattern of rat,’’ Ross, 1938; Shannon’s 1951 changes in the polarity in the bars, ‘‘maze-solving rat,’’ Cordeschi, 2002; however (see Sloane, 1964, for de- and Wallace’s, 1952, ‘‘maze-solving tails). Wyckoff and Page’s device was computer’’), but Wyckoff was the the first modification to Skinner and first to consider the principles of Campbell’s scrambler that permitted discrimination and conditioned rein- a more rapid, uniform, and reliable forcement in constructing such an polarity alternation (see Azrin, Hop- electronic model. Contemporary wood, & Powell, 1967). computer models of operant condi- Wyckoff was also coauthor of a tioning such as the ‘‘Skinnerbots’’ paper on the influence of punishment (Touretzky & Saksida, 1997) and and reinforcement in a minimal social Sniffy the Virtual Rat (Alloway, situation (Sidowski et al., 1956), Wilson, & Graham, 2005) now rou- which received particular attention tinely incorporate these principles. from social psychologists (e.g., Col- Wyckoff’s other accomplishments man, 1995; Gergen & Barton, 1974). during this time included a 1-year The experiment demonstrated that fellowship, in 1955, at the Center for two participants can work together Advanced Study in the Behavioral to produce reinforcement and reduce Sciences at Stanford University punishment for both of them even (‘‘Faculty: Honored and Appointed,’’ when they are unaware of being in a social interaction. This paper was 1955). From his work at the center, reprinted and discussed in McGinnies he refined his theory of conditioned and Ferster’s volume on the rein- reinforcement by addressing the forcement of social behavior (Si- asymmetry of the reinforcing and dowski, Wyckoff, & Tabory, 1971) aversive properties of the stimuli and in Staats’s edited textbook on produced by observing (Wyckoff, human learning (Sidowski, Wyckoff, 1959; see Dinsmoor, 1983, for a & Tabory, 1964). During Wyckoff’s last year at 2 The diagrams of the electronic model and the list of parts are stored as Document Wisconsin, a generalized discontent ADI4160 in the Library of Congress, Wash- with the policies of the American ington, D.C. Psychological Association (APA) to- OBSERVING BEN WYCKOFF 155 ward experimental psychology led to second one, thereby showing up for the creation of the Psychonomic class wearing both ties simultaneously. Society in 1959. The primary issue When Wyckoff was offered a was APA’s strong focus on profes- position of associate professor with sionalization and applied psychology tenure at Emory University in 1960, at the expense of basic research (see he accepted it without hesitation, Dewsbury & Bolles, 1995, for a leaving Madison with his wife and detailed history of the Psychonomic two sons, Daniel and Andrew (E. Society). The society started formally Wyckoff, personal communication, when a group of psychologists from March 22, 2009). Despite their hav- the University of Wisconsin sent ing talked only a few times, he letters inviting W. K. Estes, F. A. persuaded Stephen Kendall, who Geldard, C. H. Graham, N. E. Miller, was a talented graduate student at C. T. Morgan, W. D. Neff, K. W. Wisconsin, to apply to Emory and Spence, S. S. Stevens, and B. J. study with him. Kendall had taken a Underwood to serve on the organiz- few undergraduate courses with ing committee of the American Fed- Dinsmoor at Indiana, but his gradu- eration of Experimental Psycholo- ate assistantship at Wisconsin was gists.3 Among the psychologists who reduced essentially to training and sent the invitation letters, and who testing cats. Kendall’s impression was should be acknowledged for their role that this tedious task was unlikely to in starting what would be called later change in the near future. As a result, the Psychonomic Society, were Ver- he considered the invitation to trans- planck, who was at the time a visiting fer to Emory with Wyckoff a good professor at the University of Wis- opportunity, and took it (S. Kendall, consin, and Wyckoff. personal communication, March 7, Wyckoff’sloathingofadministrative 2009). work and academic politics may have contributed to his discontent at Wis- EMORY UNIVERSITY AND consin. According to Zeigler (personal TEACHING MACHINES, INC. communication, December 18, 2008), The environment at Emory was he has been described as neither prac- complicated. On the one hand, it was tical nor political enough to survive in favorable to conducting behavioral the academic ‘‘jungle.’’ E. Wyckoff research (D. Levis, personal commu- (personal communication, March 22, nication, March 17, 2009). On the 2009) described him as ‘‘resistant to other hand, the political issues authority and rebellious, introverted, surrounding the choice of a new andsomewhatofaloner.’’Accordingto department chair were completely J. Cotton (personal communication, disrupting (S. Kendall, personal com- February 15, 2009), another issue that munication, March 7, 2009). Appar- may not have helped Wyckoff’s teach- ently, an unspoken reason for hiring ing career was his well known absent- Wyckoff at Emory was the need for a mindedness. In a humorous example, new chair of the department. Con- Cotton recalled that Wyckoff kept an vincing Wyckoff to take the job, extra tie in the lab because wearing a tie however, ended up being impossible was mandatory while teaching. On (E. Wyckoff, personal communica- some occasions, he apparently forgot tion, March 22, 2009). Some months the tie he was wearing and put on a later, Charles C. Perkins joined the faculty and took the chair position, but all the pressure exerted on Wyck- 3 Letter. July 23, 1959. David Grant Papers, Box M1026, Folder: ‘‘Correspondence, 1942– off in the previous months adversely 1971.’’ Archives of the History of American affected his motivation to pursue an Psychology, University of Akron. academic career. 156 ROGELIO ESCOBAR & KENNON A. LATTAL

Wyckoff’s dislike of academic pol- ble for the design of instructional itics and administrative work con- programs that could be presented in trasted with his profound curiosity books rather than by machines. and love of research. D. Levis (per- Although instructional devices had sonal communication, March 17, been described by Pressey in the 2009), who was at the time a graduate 1920s (Pressey, 1926, 1927; see Ben- student of Perkins, remembered that jamin, 1988), teaching machines re- Wyckoff spent hours peering into his ceived more public attention after experimental chamber to see what his Skinner’s (1954) influential paper pigeons were doing. He investigated ‘‘The Science of Learning and the delayed reinforcement and supersti- Art of Teaching.’’ In it, Skinner tious behavior in pigeons, and started described a teaching device that to develop a procedure akin to a would integrate the advances in the synthetic schedule of reinforcement experimental analysis of behavior that generated different rates of into education. An important feature responding without altering rein- of the machine was that correct forcement frequency (S. Kendall, responses were immediately rein- personal communication, March 12, forced. The device presented pro- 2009). None of his research conduct- grammed material to individual stu- ed at Emory, however, was pub- dents in the form of questions, lished. Perhaps an important reason exercises, or problems to be solved. for this was that a new enterprise, The student was required to provide teaching machines, proposed by his an answer, which could be compared old friend Lloyd Homme, became immediately with the correct re- most appealing and seems to have sponse. In this way, each student distracted him from his academic practiced and was tested for his or pursuits. her advances with the material to be After obtaining his PhD at Indiana learned and advanced at his or her University in 1953 under the super- own pace (see also Lumsdaine, vision of Estes (Hearst & Capshew, 1960b). 1988), Homme took a position at the Skinner’s project attracted media University of Pittsburgh. He later attention (Skinner, 1983, p. 132) and was invited to work with Skinner at originated the golden age of teaching Harvard in his teaching machine machines. Several companies started program in the late 1950s (Skinner, to build and distribute teaching 1983, p. 119). This invitation fol- machines. Some of them were more lowed Skinner’s first presentation of successful in disseminating the tech- teaching machines at the University nology than Skinner, who had tried of Pittsburgh in 1954 (Skinner, 1954). to distribute his machines with IBM One of the achievements of Homme’s but could not close the deal (Skinner, collaborative work with Skinner was 1983, pp. 97–98). In 1959, Homme the development of instructional pro- and James Evans, one of his col- grams to be used in teaching ma- leagues from the University of Pitts- chines (Tosti & Kaufman, 2007). The burgh, started their own company, programs ultimately would become Teaching Machines Inc. (TMI), to the most important element in the develop and distribute teaching ma- teaching machine movement, far chines and programmed books. more important than the iconic Homme invited his old friends from hardware. From his work with Skin- Indiana University, Robert Glaser, ner, Homme learned that the pro- who also was a colleague at the grams were the key to making University of Pittsburgh, and Wyck- teaching machines successful. Indeed, off to join them in the enterprise. For Tosti (in Tosti & Kaufman, 2007) Wyckoff, it seemed to be a good observed that Homme was responsi- opportunity to put his acumen with OBSERVING BEN WYCKOFF 157

Figure 1. The picture shows from left to right, the Min/Max I, the Min/Max II, and the Min/ Max III teaching machines. apparatus to practical use. In the writing (D. Tosti, personal commu- beginning, the company was head- nication, March 17, 2009). In a short quartered in Pittsburgh, but in 1960 time, TMI developed programs that its headquarters moved to Albuquer- covered topics in English, basic and que, New Mexico (D. Tosti, personal advanced mathematics, general sci- communication, March 18, 2009). ence, languages, and music. The Although the programmed instruc- assortment of programs, in combination industry was more concentrated tion with a low-price teaching ma- on manufacturing teaching machines chine, was the key to obtaining a than on the programs for the devices major distribution agreement with (cf. Hechinger, 1966), TMI had the Grolier Inc. in November 1960. opposite agenda. As noted above, the Grolier, which was one of the largest first programmed instruction books publishers of encyclopedias in the were developed by Homme, along U.S., used their 5,000 door-to-door with Evans, and Glaser, at the salesmen to sell their encyclopedia University of Pittsburgh, between sets, the Book of Knowledge and the 1958 and 1960 (Glaser, Homme, & Grolier Encyclopedia (Klaw, 1962; Evans, 1960; Lumsdaine, 1960b). ‘‘The Truth About Those Teaching Dudley E. Cornell was hired by Machines,’’ 1962), with the added TMI to design what would become incentive of receiving a complemen- one of the most successful teaching tary Min/Max teaching machine and machines, the Min/Max (Cornell, its programmed courses with any 1963). The name was short for purchase. Of course, it was also minimum time/maximum learning. possible to purchase the machine The original version, shown in Fig- and the programs independently of ure 1, was made of metal, was the encyclopedia sets. slightly larger than a typewriter and By 1962, the Min/Max was made could hold several pages of programs more efficient and portable. Cornell at a time. Through a window, the had designed two new versions of the student could see the question, and a Min/Max, also shown in Figure 1. space to write the answer was acces- The Min/Max II and III were made sible in another window. After writ- of plastic and also about the size of a ing the answer, the student could see small typewriter. Both models includ- the correct responses and advance the ed two knobs on the sides that served paper by pushing it with a pencil to advance the paper, but the Min/ eraser (Fine, 1962). Max III was smaller and lighter As TMI’s Chairman of the Board, (Cornell, 1966). Wyckoff made frequent trips to The advertisements of TMI pro- Pittsburgh and later to Albuquerque moting a low-cost teaching machine from Atlanta to supervise the devel- and a variety of programs can be opment of the programs for the Min/ found in many newspapers, educa- Max, which Tosti was in charge of tion journals, and journals of that era 158 ROGELIO ESCOBAR & KENNON A. LATTAL focused on programmed instruction a more sophisticated teaching ma- and behavior (e.g., American Behav- chine than the Min/Max series. The ioral Scientist, Grade Teacher, Har- first version of the machine, which vard Educational Review, Journal of Kendall helped construct, was con- Educational Research, Journal of Pro- trolled with five keys and presented grammed Instruction, JEAB, The the instructional material on a screen. American School Board Journal, The material, designed to teach chil- School Management, The Elementary dren to read, was encoded on 35-mm School Journal, The New York Times, slides. The slides showed incomplete The Science Teacher, Today’s Educa- words that the children were required tion, among others). Even articles in to complete, one letter at a time. the general media described the Various combinations of presses on ‘‘revolutionary’’ teaching machines the five keys made up the alphabet. sold by TMI-Grolier (e.g., Klaw, Only when the keystrokes matched 1962; ‘‘The Teaching Machines,’’ the pattern in the film strip was the 1960). In the first 2 years, TMI- answer coded as correct and the film Grolier sold over 150,000 teaching allowed to advance, presenting the machines, thereby becoming the most correct letter on the screen. successful company devoted to pro- The machine was labeled the grammed instruction and teaching Wyckoff film-tutor and was first machines (Klaw, 1962). available in 1959 (Kopstein & Shil- After the enormous success of the lestad, 1961). Wyckoff filed for a Min/Max and the company’s pro- patent in 1960 (Wyckoff, 1964). grammed texts, TMI’s focus widened Figure 2 shows a diagram of the when they offered a version of one of machine based on the patent descrip- Skinner’s most controversial gadgets, tion. A serious problem with the film- the air crib or baby tender (Skinner, tutor was that, to use the machine, 1945/1972). In 1962, TMI began students had to learn a new code (i.e., selling its own version of Skinner’s the key combinations for each letter) air crib. The first model, named the or use a conversion table. Wyckoff Incu-crib, sold for $499. It was first subsequently designed improved ver- advertised in the April 1962 issue of sions of the machine that included a JEAB. Skinner recalled in his auto- full keyboard, eliminating the need to biography that he was asked to learn a code. Two versions of the comment on the crib, and he disliked machine are also shown in Figure 2. the use of a plastic bubble because of The Wyckoff film-tutor solved two the resonating acoustics (Skinner, problems posed by the Min/Max. 1983, pp. 250, 251). A few months One was the amount of paper re- later the model was modified to make quired to present the programs. In a the child more visible, improve the single year, TMI consumed 1,000 tons acoustics inside the crib, and lower its of paper in printing programs for the cost. After Skinner approved the Min/Max (Donnelly, 1964). The oth- modifications, the name of the crib er problem was that the length of was changed to TMI-Aircrib and it each question item was constrained sold for $250. D. Tosti (personal by the size of the window in the Min/ communication, March 17, 2009) Max. The Wyckoff film-tutor al- recalled that he used this crib for lowed the presentation of question three of his children. An advertise- items of various sizes (Lumsdaine, ment in the April 1963 issue of JEAB 1960a). portrays Jill A. Cornell, Dudley At TMI, Roger Steinhorst was in Cornell’s daughter, as a happy child charge of testing Wyckoff’s teaching in a TMI-Aircrib. machine in the laboratories in New Even before the Min/Max I was on Mexico. Steinhorst recalled that it the market, Wyckoff began designing worked flawlessly (R. Steinhorst, OBSERVING BEN WYCKOFF 159

Figure 2. Three versions of the Wyckoff film-tutor. The diagram on the left shows the first version of the machine available in 1959. The machine was controlled with five keys, projected the material on a screen, and could be folded to make it portable (drawing based on U.S. Patent No. 3,137,948). The middle drawing shows an improved version with a full keyboard. This drawing was used as the TMI logo in commercial letters circa 1960 (William S.Verplanck Papers, Box M1887, Folder 1. Archives of the History of American Psychology, The University of Akron). The picture on the right shows the commercial version of the Wyckoff film-tutor (picture based on Malpass et al., 1963). personal communication, March 19, the procedure of stimulus equivalence 2009). When TMI offered the Wyck- used by Sidman (1971), more than off film-tutor to Grolier for distribu- 10 years later, to teach children with tion, Grolier executives refused (R. disabilities spoken and written words. Steinhorst, personal communication, Afterwards, using a procedure that March 19, 2009). The executives was similar to the method of stimulus considered it to be too expensive. At fading used by Skinner and Holland $445, it was more than 20 times the (e.g., Holland, 1960), parts of the price of the Min/Max and weighed word were removed. The child had to close to 30 lb, too heavy to be carried choose or type the letters that com- and sold door to door. Although pleted the missing segments of the TMI advertised the Wyckoff film- word. After the word was trained, it tutor independently of Grolier (see, was incorporated into a sentence and e.g., JEAB, 1961, Vol. 3, No. 4; 1962, the procedure was repeated. Vol. 4, No. 1), it was never a Based on the tests conducted at commercial success. As might be TMI, the time required to complete expected, Wyckoff was demoralized the course ranged from 15 to by this outcome (J. Berlin, personal 35 hours. Fine (1962, p. 77) witnessed communication, March 18, 2009). the success of the program with 4- to Nevertheless, he developed a new 5-year-old children when he visited model that presented an audible TMI’s facilities. The program also version of the instructional material. proved to be successful for teaching Although the Wyckoff film-tutor children with learning disabilities to was not distributed by Grolier, the read (Malpass, Gilmore, Hardy, & program inside the machine was Williams, 1963; ‘‘Teaching Machines translated into a TMI-Grolier pro- Speed Progress,’’ 1964). Malpass et grammed course for the Min/Max al. described how children with dis- (see Andrego, de Baca, Fullilove, abilities, using Wyckoff’s method, Stranczek, & Wyckoff, 1962). The learned from two to six times more purpose of the program was to teach words in 8 weeks than the children children reading skills more rapidly who used the conventional classroom than the conventional methods. In- methods learned in 4 years. They also stead of teaching the child to read suggested that the method was as letter by letter, the program first effective as personalized instruction, trained the child to match an image with the advantage that it could be with a written word. This is similar to used with several children concur- 160 ROGELIO ESCOBAR & KENNON A. LATTAL rently. The method used by Wyckoff tional positive regard throughout his also had an impact on subsequent or her interaction with the speaker. programs developed by TMI. For Wyckoff might have been expected to example, D. Tosti (personal commu- disagree with his colleague’s analysis, nication, March 17, 2009) recalled but instead, he carefully considered it that he and Homme used it in a (J. Berlin, personal communication, course on letter writing. March 18, 2009). Wyckoff’s method for teaching Wyckoff went on to suggest that reading skills reached a diverse audi- communication can be analyzed as a ence. For example, James Cook learning process. People communi- Brown discussed with Wyckoff the cate on the basis of interlocking idea of using the modified film-tutor contingencies (Skinner, 1957, p. 40); for teaching Loglan, an artificial that is, speaking provides discrimina- human language created by Brown tive and reinforcing stimuli for one (‘‘Eight Languages into One,’’ 1960). another. A person communicates Loglan was created originally to test with another if his or her verbal the Sapir-Whorf hypothesis of lan- responses are reinforced, and the guage relativity. Brown’s idea was to person stops communicating if his create a language that could be or her verbal responses are ex- learned independently of cultural tinguished or punished. The three differences and used a grammar elements Berlin considered to be based on logical rules to eliminate important in achieving good commu- ambiguities (Brown, 1989). As will be nication were analyzed by Wyckoff in described below, Wyckoff’s method terms of positive reinforcement made also came to the attention of Robert contingent on the verbal responses of Moses, who planned to use it to teach the other person (see also Truax, illiterate African-American citizens 1966a, 1966b, for a similar analysis). from Mississippi to read (see section For Wyckoff, Rogers had identified on Human Rights, below). reinforcers for adults’ verbal behavior (J. Berlin, personal communica- PERSONAL RELATIONS AND tion, March 18, 2009). Wyckoff PROGRAMMED INSTRUCTION: noted the complexity of such rein- THE HUMAN forcers, because he firmly believed DEVELOPMENT INSTITUTE that simply acknowledging what another person has to say is not an During his tenure at Emory, effective way to increase communica- Wyckoff befriended Jerome Berlin, a tion, even if it is contingent on a faculty colleague in clinical psychol- verbal response. Deckner, one of ogy who, as a former student of Carl Wyckoff’s graduate students at Em- Rogers at the University of Chicago, ory, recalled that, as a joke, Wyckoff was a proponent of client-centered built a small apparatus with a lever, therapy. The two frequently dis- that, when pressed, put up a little sign cussed their seemingly different ap- that said, ‘‘mmm hmm’’ (W. Deck- proaches to psychology. The most ner, personal communication, March fruitful of these interactions may 6, 2009). Of course, no one expected have been around the topic of how it to work. personal relationships are main- Wyckoff went further in analyzing tained. They both agreed that an communication by pointing out that, important element in such mainte- if it involves a learning process, it can nance is communication. For Berlin, be shaped. He reasoned that commu- effective communication was accom- nication could be systematically plished if the listener resembled a trained using programmed instruc- client-centered therapist by showing tion with two persons simultaneously empathy, congruence, and uncondi- (G. Ruyle, personal communication, OBSERVING BEN WYCKOFF 161

January 28, 2009; J. Berlin, personal no market for this product (D. Tosti, communication, March 18, 2009). personal communication, March 17, Wyckoff’s analysis was convincing 2008). After this rejection, Wyckoff enough to persuade Berlin to test and Berlin turned to selling it on their these ideas. own. Wyckoff resigned as Chairman Several of Wyckoff’s students of the Board of TMI to have more served as participants in the first time for the new project (J. Berlin, experiments in which spontaneous, personal communication, March 18, and later preestablished, dialogues 2009), but he continued as a member were scrutinized (G. Ruyle, personal of the board of directors. communication, January 28, 2009; J. To embark on the new enterprise, Berlin, personal communication, both Wyckoff and Berlin also re- March 18, 2009). Wyckoff and Berlin signed their faculty positions at Em- asked the participants to interact in ory (Wyckoff was allowed to contin- dyads (R. Addison, personal commu- ue as Kendall’s adviser until he nication, March 18, 2009). They obtained his PhD in 1963; S. Kendall, assessed whether communication personal communication, March 7, was improved by training each par- 2009). They rented a small building ticipant to make positive statements on West Peachtree Street in Atlanta, about the previous verbal response of and founded their company. The the other. For subsequent experi- HDI was nominally started in March ments, they created examples of 1962, but was not fully operational appropriate and inappropriate dia- until 1964. Figure 4 shows a part of a logues to serve as the basis for brochure describing the institute. In discriminating the correct and incor- their new environs, they continued rect verbal responses of the other their work on interpersonal commu- participant. nication and, during 1963 and 1964 From these experiments, Wyckoff they wrote the first programmed and Berlin derived systematic exam- courses to improve communication ples of everyday dialogues and a in business and organizations, and series of questions that they incorpo- others to improve communication in rated into the first programmed marriages. Figure 3 also shows a instruction course completed in 1963 fragment of the fourth edition of the and designed to improve communi- latter (HDI, 1970). cation in general relationships. Fig- The strong emphasis on research in ure 3 shows a fragment of the sixth developing the training programs edition of the manual (Human De- resulted in HDI becoming a great velopment Institute [HDI], 1972), success. The training programs were which was designed to train each used to improve communication be- individual in a dyad to identify and tween employees at United Airlines, reinforce the appropriate verbal re- Union Carbide, Lockheed Aircraft, sponses of the other individual. In Space Technology Laboratories, this way, communication was expect- Baylor University Hospital, the So- ed to improve in the absence of a cial Security Administration, Western therapist (see HDI, 1972). These Electric, Procter and Gamble, volun- ideas, the program, and the results teers at the Atlanta Chapter of of several tests were presented at the Hadassah, and graduate students 1963 and 1964 APA conventions at both the University of Alabama (Berlin & Wyckoff, 1964; Wyckoff and Georgia State College (Berlin & & Berlin, 1963). Wyckoff, 1964; Nicholson, 1963; Wyckoff suggested that the manual Roalman, 1963). be edited by TMI and distributed by Originally created to improve com- Grolier, but the executives of Grolier munication between two persons, the again declined, noting that there was training programs started to gain 162 ROGELIO ESCOBAR & KENNON A. LATTAL

Figure 3. Sample items from the General Relationship Improvement Program and Improving Communication in Marriage (courtesy of Jerome Berlin). The first edition of the former was written by the Human Development Institute in 1963. The top three items were designed to teach people how to avoid punishing others’ verbal responses. The bottom three items were used to teach people how to distinguish different situations in a marital relation that require different types of conversations. The terminology was taken from Rogers’s client-centered therapy. popularity as an alternative to coun- theory (Lewin, 1951) grounded t- seling (cf. Eisenberg & Zingle, 1975; groups in 1947 (Margulies & Raia, Ellis, 1966; Hickman & Baldwin, 1978). Because of programmed in- 1971). Although Hickman and Bald- struction, however, Wyckoff and win found that HDI’s program for Berlin’s manuals were more system- improving communication in mar- atic than t-groups, and could be used riage positively influenced the atti- in the absence of a therapist or tudes of couples toward marriage, it observer. was not found to be more effective According to J. Berlin (personal than personalized marriage counsel- communication, March 18, 2009), ing. Nevertheless, it is worth noting revising and developing new pro- that Wyckoff and Berlin’s attempt to grams at HDI, although successful, use psychological principles to train was exhausting. The first years of the two persons simultaneously in im- company were difficult. Each pro- proving their relationship was pre- gram took months and even years to ceded only by the creation of field develop. To survive, the work at HDI OBSERVING BEN WYCKOFF 163

Figure 4. Fragments of a brochure distributed by the Human Development Institute in 1964 (courtesy of Jerome Berlin). The left section shows a teaching machine, designed by Wyckoff, that could be used to work through HDI’s manuals. The right section shows hand drawings of Jerome Berlin and Benjamin Wyckoff. was transformed from a research About the same time, TMI was enterprise into a business, which collapsing. To begin with, teaching was completely different from what machines were no longer thought to Wyckoff and Berlin envisioned in the be revolutionary devices (see Benja- beginning. Once again exasperated min, 1988, for a history of teaching by administrative issues, Wyckoff left machines). Criticisms raised by edu- HDI in 1965. cators had reduced the sales of both 164 ROGELIO ESCOBAR & KENNON A. LATTAL the Min/Max and programmed ences Institute in La Jolla, California. books. These criticisms had two At that time, Rogers was a resident thrusts. One was the dehumaniza- fellow at the Institute. Berzon moved tion inherent in the use of machines. to Atlanta to learn about pro- Although Skinner never intended grammed instruction at HDI, from teaching machines to be a substitute which the tapes then developed (e.g., for teachers, this idea surfaced in Berzon & Solomon, 1964). the general media as a threat to the From interactions with employees development of creative and free- of several different companies, Berlin thinking students (e.g., Boroff, 1960; observed that most often workplace Gilmore, 1961). The other criticism structures consisted of white supervi- pointed to the functionality of sors and African-American or His- teaching machines. It was suggested panic employees. He adapted HDI’s that teaching machines were nothing programs to train the supervisors to more than expensive page turners relate to the employees with empathy (e.g., ‘‘The Truth About Those and respect. The final product was Teaching Machines,’’ 1962). One named the Sensitivity Kit or S-Kit solid criticism was that the primitive (‘‘How to Succeed in Hard-Core state of the technology available for Hiring,’’ 1968; James, 1969). Fried- teaching machines was limiting the lander, who arrived at HDI in 1967, evolution of programmed instruc- recalled: tion (Gilbert, 1960). TMI survived for some years because, by 1963, it Iwasde facto leader of the team that had shifted from the massive door- produced the breakthrough product, called to-door teaching machine business the S-Kit, which was used to train 130,000 to developing custom courses for ‘‘front-line supervisors’’ in industries around the country to manage culturally different large organizations such as IBM, workers. In my opinion this was the true the U.S. Navy, and others. In 1963, ‘‘proof of concept’’ that prosocial behavior in Glaser left TMI to start an inde- adults could be systematically modified pendent project, and Homme did through the application of the principles of operant conditioning to the design of a the same in 1964. In 1966, Evans sequence of INTERACTIVE interpersonal and Cornell declared bankruptcy, exercises. (S. Friedlander, personal communi- and TMI’s assets were purchased cation, March 16, 2009) by Grolier. By the mid 1960s, HDI also was Although HDI was enormously suc- struggling for survival. To be relieved cessful, Berlin decided, as Wyckoff of some of the administrative and had earlier, that he had had enough financial issues, in 1967, Berlin was of administrative issues and left the forced to negotiate establishing HDI organization in the early 1970s. The as a subsidiary of Bell & Howell assets of HDI then were sold entirely (‘‘Bell-Howell Plans to Buy,’’ 1967; to Bell & Howell, and the institute J. Berlin, personal communication, was moved to Chicago in 1974 (G. March 18, 2009). Bell & Howell Ruyle, personal communication, Jan- capitalized on the commonalities uary 28, 2009). Given the debates between the work done at HDI and between Rogers and Skinner (e.g., Rogers’s client-centered psychothera- Rogers & Skinner, 1956) in which py to develop and distribute audio- they each strongly asserted their tapes designed to train groups of points of view of psychology, it is people to interact effectively. These somewhat ironic that the work of tapes were known as encountertapes Berlin and Wyckoff, firmly anchored in the Human Potential Movement in the principles of operant condi- (Klemesrud, 1970). The encounter- tioning, was one of the starting points tapes were developed by Betty Ber- of the expansion of client-centered zon at the Western Behavioral Sci- therapy into what became the Hu- OBSERVING BEN WYCKOFF 165 man Potential Movement (e.g., How- a plan to use programmed instruction ard, 1970). to teach illiterate African-American Mississippians to read, thereby in- creasing the likelihood of their pass- HUMAN RIGHTS ing the literacy test. Wyckoff strove to treat everyone In the spring of 1963, Wyckoff with empathy, dignity, and respect and Robert Moses, field secretary (A. Weiskoff, personal communica- for SNCC, met at Wyckoff’s home tion, March 22, 2009). His experi- in Atlanta to discuss programmed ence of working in the early 1960s instruction. One of the topics was in industries in which African-Amer- how reinforcement theory was used in programmed instruction. Wyckoff ican employees had conflicts with reportedly noted the importance of white supervisors, and women were reinforcement immediacy, a princi- paid only a fraction of what men ple that ‘‘fermented within Bob earned, seemed to have strongly Moses’’ (Holt, 1965, p. 153). Holt affected him. Several of the people describes how Moses came to con- interviewed for this review corrobo- ceptualize the many obstacles im- rated that Wyckoff was a passionate posed to register to vote as delays of supporter of human rights. He was reinforcement. remembered for talking in the TMI As a result of the seed planted by offices and among friends about the discussion with Wyckoff, Moses the successful civil rights protests and his colleagues conceived a plan in the U.S. South in the 1960s, and to capitalize on the principle of of the importance of supporting immediate reinforcement. This in- both the civil rights and the wo- volved establishing the 1963 Free- men’s rights movements (D. Tosti, dom Vote Campaign in Mississippi. personal communication, March 17, Holt (1965) noted that ‘‘The Free- 2009). He was a member of the dom Vote Campaign provided pos- National Association for the Ad- itive and quick reinforcement’’ vancement of Colored People and (p. 153). It was as a result of this, hosted civil rights meetings at his according to Holt, that the Missis- home (S. Kendall, personal commu- sippi Freedom Democratic Party nication, March 7, 2009). (MFDP) became established as an Holt (1965) described Wyckoff’s alternative to the Mississippi Dem- rather unique involvement in the ocratic Party. The former, following Freedom Vote Campaign in 1963. the principle of immediacy of rein- One of the political issues in Mis- forcement, allowed any citizen to sissippi was that in 1962, only 6.7% vote in its delegate elections, regard- of Mississippi’s African-American less of whether they met state adults were eligible to vote. The eligibility requirements. The out- reason was that voter eligibility de- come was the massive media atten- manded the citizen meet a series of tion that the MFDP received at the strict requirements (e.g., paying a poll 1964 Democratic National Conven- tax, fulfilling a residency condition, tion (see Smith, 2004) that resulted and passing a literacy test). The result in the MFDP delegation receiving of such requirements was that, prior two at-large seats at the convention. to the 1965 Voting Rights Act that Holt relates these outcomes directly abolished them, an African-Ameri- back to Wyckoff and the principle can citizen’s opportunity to vote was of immediacy of reinforcement. The long delayed after the initial attempt ultimate outcome, according to to register. Holt noted that the Holt, was that the campaign dem- Student Nonviolent Coordinating onstrated convincingly to the coun- Committee (SNCC) had considered try as a whole that African-Ameri- 166 ROGELIO ESCOBAR & KENNON A. LATTAL cans in Mississippi were eager to influenced Deckner’s work on educa- exert their right to vote.4 tional games based on programmed instruction (Deckner, Deckner, & LATER CAREER AND LEGACY Davis, 2007), and Kendall’s signifi- cant contributions to the literature on Although the enterprises in which conditioned reinforcement using the Wyckoff was involved were success- observing-response procedure (e.g., ful in terms of their impact on Kendall, 1965, 1972, 1973). education and improving human Wyckoff saw that all psychologists, relationships, none of them were independent of their particular orien- financial successes. Both TMI and tation, are concerned with the study HDI generated large profits, but of behavior. He was open to different most of those profits were reinvested approaches to the study of behavior, to improve the machines and pro- rather than being identified strictly as grams. After Wyckoff left HDI, he a Skinnerian, Hullian, or Spencian. worked as a systems analyst at the This openness contributed to his Georgia Department of Vocational enduring contributions to both Skin- Rehabilitation. He also obtained a nerian and Hull-Spence theories of position as director of research and behavior through the observing-re- evaluation at the Georgia Depart- sponse procedure, his mathematical ment of Corrections, where later theory of secondary reinforcement, created a program used to keep track and programmed instruction. Like of the records (A. Weiskoff, personal Skinner, he was committed to the communication, March 22, 2009). He application of behavioral research to retired from that position in 1987. improving the human condition. His Even though Wyckoff had only a active involvement with TMI and few doctoral graduates during his HDI exemplifies the translation of career (Benjamin Pubols, H. Philip basic principles of behavior analysis Zeigler, Jerry Tate, William Deckner, to applied settings and social issues. and Stephen Kendall), his influence is Wyckoff’s most important message evident in their academic careers. was that, by leaving aside theoretical Although Pubols was primarily in- and conceptual biases, it is possible to volved in neuroscience, he conducted integrate psychological knowledge to research on the parameters of rein- improve the quality of peoples’ lives. forcement (e.g., Pubols, 1958, 1960, 1962). Zeigler’s primary interest was REFERENCES comparative cognition. He modified Wyckoff’s observing-response proce- Alloway, T., Wilson, G., & Graham, J. (2005). dure such that a pair of stimuli was Sniffy the virtual rat: Pro version 2.0. presented after an observing response Belmont, CA: Thomson West. Andrego, P., de Baca, P. C., Fullilove, J., was emitted. In this way, he used Stranczek, T., & Wyckoff, L. B. (1962). the procedure as an analogue of Modern English series: Remedial reading. Harlow’s Wisconsin General Test In Programs ’62: A Guide to programmed Apparatus (Zeigler, 1958; Zeigler & instructional materials available to educators (pp. 53–55). Washington, DC: Center for Wyckoff, 1961). Wyckoff directly Programmed Instruction. Atkinson, R. C. (1961). The observing re- 4 Wyckoff’s interest in social issues contin- sponse in discrimination learning. Journal of ued throughout his life. In 2006 he completed Experimental Psychology, 62, 253–262. the script of a musical (music and lyrics by Atkinson, R. C., & Estes, W. K. (1963). Thomas A. Gaines; U.S. Copyright Office, Stimulus sampling theory. In R. D. Luce, R. Registration Number: PAu003079245) based R. Bush, & E. Galanter (Eds.), Handbook of on the 1960s novel Peaceable Lane by Keith mathematical psychology (Vol. 2, pp. 121– Wheeler. The novel narrates the segregation 268). New York: Wiley. issues faced by an African-American family Auerbach, V. H., Waisman, H. A., & Wyck- who moved to a white community in the off, L. B., Jr. (1958). Phenylketonuria in the suburbs of Westchester, New York. rat associated with decreased temporal OBSERVING BEN WYCKOFF 167

discrimination learning. Nature, 182, Dewsbury, D. A., & Bolles, R. C. (1995). The 871–872. founding of the Psychonomic Society. Psy- Azrin, N. H., Hopwood, J., & Powell, J. chonomic Bulletin & Review, 2, 216–233. (1967). A rat chamber and electrode proce- Dinsmoor, J. A. (1950). A quantitative com- dure for avoidance conditioning. Journal of parison of the discriminative and reinforc- the Experimental Analysis of Behavior, 10, ing functions of a stimulus. Journal of 291–298. Experimental Psychology, 40, 458–472. Bell-Howell plans to buy Atlanta educational Dinsmoor, J. A. (1983). Observing and firm. (1967, June 3). Wall Street Journal, conditioned reinforcement. Behavioral and p. 7. Brain Sciences, 6, 693–728. (includes com- Benjamin, L. T., Jr. (1988). A history of mentary) teaching machines. American Psychologist, Dinsmoor, J. A. (1987). A visit to Blooming- 43, 703–712. ton: The first Conference on the Experi- Berlin, J. I., & Wyckoff, L. B., Jr. (1964, mental Analysis of Behavior. Journal of September). Human relations training the Experimental Analysis of Behavior, 48, through dyadic programmed instruction. 441–445. Paper presented at the convention of the Dinsmoor, J. A. (1990). Academic roots: American Psychological Association, Los Columbia University, 1943–1951. Journal Angeles. of the Experimental Analysis of Behavior, 54, Berzon, B., & Solomon, L. N. (1964). The self- 129–149. directed therapeutic group: An exploratory Donnelly, C. J. (1964). Problems in publishing study. International Journal of Group Psy- programmed materials. In G. D. Ofiesh & chotherapy, 14, 366–369. W. C. Meierhenry (Eds.), Trends in pro- Boroff, D. (1960, September 25). The three grammed instruction: Papers from the first R’s and pushbuttons. New York Times,pp. annual convention of the National Society 36, 66, 68, 70, 72. for Programmed Instruction (pp. 269–271). Brown, J. C. (1989). Loglan 1: A logical Washington, DC: National Education As- language (4th ed.). Fredericksburg, VA: sociation. BookCrafters. Eight languages into one is Floridian’s ambi- Burke, C. J. (1949). The function relating tion. (1960, July 2). The Miami News,p.7A. momentary effective reaction potential to Eisenberg, J. M., & Zingle, H. W. (1975). running-time in a straight runway. Unpub- Marital adjustment and irrational ideas. lished doctoral dissertation. The University Journal of Marriage and Family Counseling, of Iowa, Iowa City. 1, 81–91. Bush, R. R. (1965). Identification learning. In Ellis, A. (1966). The nature of disturbed R. D. Luce, R. R. Bush, & E. Galanter marital interaction. Rational Living, 1, (Eds.), Handbook of mathematical psycholo- 22–26. gy (Vol. 3, pp. 161–203). New York: Wiley. Escobar, R., & Bruner, C. A. (2009). Observ- Colman, A. M. (1995). Game theory and its ing responses and serial stimuli: Searching applications in the social and biological for the reinforcing properties of the S-. sciences (2nd ed.). Oxford, UK: Butter- Journal of the Experimental Analysis of worth-Heinemann. Behavior, 92, 215–231. Conference on the Experimental Analysis of Faculty: Honored and appointed. (1955, Behavior Notes. (1948, June). Bloomington, November 15). Wisconsin Alumnus, 57,p. IN: mimeograph. 18. Retrieved from http://digicoll.library. Cordeschi, R. (2002). The discovery of the wisc.edu/cgi-bin/UW/UW-idx?type5article& artificial: Behaviour, mind and machines did5UW.V57I6.I0013&id5UW.v57i6&isize5 before and beyond cybernetics. Dordrecht, text The Netherlands: Kluwer Academic. Fantino, E. (1977). Conditioned reinforce- Cornell, D. E., III. (1963). U.S. Patent ment: Choice and information. In W. K. No. 3,105,307. Washington, DC: U.S. Pat- Honig & J. E. R. Staddon (Eds.), Handbook ent and Trademark Office. of operant behavior (pp. 313–339). Engle- Cornell, D. E., III. (1966). U.S. Patent wood Cliffs, NJ: Prentice Hall. No. 3,274,704. Washington, DC: U.S. Pat- Fantino, E., & Silberberg, A. (2010). Revisit- ent and Trademark Office. ing the role of bad news in maintaining Deckner, C. W., Deckner, P., & Davis, D. F. human observing behavior. Journal of the (2007, May). Behaviorally-based educational Experimental Analysis of Behavior, 93, games: ‘‘Steady learning.’’ Paper presented 157–170. at the annual meeting of the Association for Fine, B. (1962). Teaching machines.New Behavior Analysis, San Diego. York: Sterling. DeFulio, A., & Hackenberg, T. D. (2008). Gergen, K. J., & Barton, W. (1974). Social Combinations of response-dependent and psychology: Explorations in understanding. response-independent schedule-correlated Del Mar, CA: CRM Books. stimulus presentation in an observing pro- Gilbert, T. A. (1960). On the relevance of cedure. Journal of the Experimental Analysis laboratory investigation of learning to self- of Behavior, 89, 299–309. instructional programming. In A. A. Lums- 168 ROGELIO ESCOBAR & KENNON A. LATTAL

daine & R. Glaser (Eds.), Teaching machines fixed-interval schedule. Journal of the Ex- and programmed learning: A source book perimental Analysis of Behavior, 17, (pp. 475–485). Washington, DC: National 161–168. Education Association of the United States. Kendall, S. B. (1973). Redundant information Gilmore, K. (1961). Teaching machines— in an observing-response procedure. Journal Blessing or curse? Science Digest, 49, 76–80. of the Experimental Analysis of Behavior, 19, Glaser, R., Homme, L. E., & Evans, J. L. 81–82. (1960). An evaluation of textbooks in terms Killeen, P. R. (1982). Incentive theory: II. of learning principles. In A. A. Lumsdaine Models for choice. Journal of the Experi- & R. Glaser (Eds.), Teaching machines and mental Analysis of Behavior, 38, 217–232. programmed learning: A source book (pp. Klaw, S. (1962, July 19). What can we learn 437–445). Washington, DC: National Edu- from teaching machines? The Reporter, cation Association of the United States. 27(2), 19–26. Good turn deserves another, electronic brain Klemesrud, J. (1970, December 20). Having ‘‘discovers’’: Machine learns to imitate test wonderful encounter. New York Times,pp. rats perform functions that bring reward. 9, 16, 17. (1953, September 8). Toledo Blade,p.3. Kopstein, F. F., & Shillestad, I. J. (1961). Hearst, E., & Capshew, J. H. (1988). Psychol- A survey of auto-instructional devices (ASD ogy at Indiana: A centennial review and Technical Report). Wright-Patterson AFB, compendium. Bloomington: Indiana Univer- OH: Aeronautical Systems Division, Air sity, Department of Psychology. Force Systems Command, United States Hechinger, F. M. (1966, June 19). Education: Air Force. Retrieved from http://www.dtic. On the automated classroom. New York mil/cgi-bin/GetTRDoc?AD5AD268223& Times,p.170. Location5U2&doc5GetTRDoc.pdf Herrnstein, R. J. (1964). Secondary reinforce- Levis, D. J., & Perkins, C. C., Jr. (1965). ment and rate of primary reinforcement. Acquisition of observing responses (RO) Journal of the Experimental Analysis of with water reward. Psychological Reports, Behavior, 7, 27–36. 16,114. Hickman, M. E., & Baldwin, B. A. (1971). Use Lewin, K. (1951). Field theory in social science. of programmed instruction to improve New York: Harper and Row. communication in marriage. The Family Lieving, G. A., Reilly, M. P., & Lattal, K. A. Coordinator, 20, 121–125. (2006). Disruption of responding main- Holland, J. G. (1957). Technique for behav- tained by conditioned reinforcement: Alter- ioral analysis of human observing. Science, ations in response-conditioned reinforcer 125, 348–350. relations. Journal of the Experimental Anal- Holland, J. G. (1958). Human vigilance. ysis of Behavior, 86, 197–209. Science, 128, 61–67. Lumsdaine, A. A. (1960a). Some issues Holland, J. G. (1960). Teaching machines: An concerning devices and programs for auto- application of principles from the laborato- mated learning. In A. A. Lumsdaine & R. ry. Journal of the Experimental Analysis of Glaser (Eds.), Teaching machines and pro- Behavior, 3, 275–287. grammed learning: A source book (pp. 517– Holt, L. (1965). The summer that didn’t end. 539). Washington, DC: National Education New York: Morrow. Association of the United States. How to succeed in hard-core hiring. (1968, Lumsdaine, A. A. (1960b). Teaching ma- August 24). Business Week, pp. 64–66. chines: An introductory overview. In A. A. Howard, J. (1970). Please touch: A guided tour Lumsdaine & R. Glaser (Eds.), Teaching of the human potential movement.New machines and programmed learning: A source York: McGraw-Hill. book (pp. 5–22). Washington, DC: National Human Development Institute. (1970). Im- Education Association of the United States. proving communication in marriage (4th ed.). Lutz, R. E., & Perkins, C. C., Jr. (1960). A Atlanta, GA: Author. time variable in the acquisition of observing Human Development Institute. (1972). Gener- responses. Journal of Comparative and al relationship improvement program (6th Physiological Psychology, 53, 180–182. ed.). Atlanta, GA: Author. Malpass, L. F., Gilmore, A. S., Hardy, M. W., James, R. D. (1969). Helping out, and making & Williams, C. F. (1963). Comparison of two some money too. Wall Street Journal,p.18. automated teaching procedures for retarded Kelleher, R. T. (1958). Stimulus-producing children (Cooperative Research Project responses in chimpanzees. Journal of the No. 1267). Tampa: University of South Experimental Analysis of Behavior, 1,87– Florida. 102. Margulies, N., & Raia, A. P. (1978). Concep- Kendall, S. B. (1965). The distribution of tual foundations of organizational develop- observing responses in a mixed FI-FR ment. New York: McGraw-Hill. schedule. Journal of the Experimental Anal- Michael, J. L. (1980). Flight from behavior ysis of Behavior, 8, 305–312. analysis. The Behavior Analyst, 3, 1–24. Kendall, S. B. (1972). Some effects of Minsky, M. (1961). A selected descriptor- response-dependent clock stimuli in a indexed bibliography to the literature on OBSERVING BEN WYCKOFF 169

artificial intelligence. IRE Transactions on Sidowski, J. B., Wyckoff, L. B., & Tabory, L. Human Factors in Electronics, 2, 39–55. (1956). The influence of reinforcement and Nicholson, D. (1963, May 5). Series improves punishment in a minimal social situation. leadership. Atlanta Journal and Constitution, Journal of Abnormal and Social Psychology, E7. 52, 115–119. Norman, M. F. (1968). Mathematical learning Sidowski, J. B., Wyckoff, L. B., & Tabory, L. theory. In G. B. Dantzig & A. F. Veinott (1964). The influence of reinforcement and (Eds.), Mathematics of the decision sciences punishment in a minimal social situation. In (Part 2, pp. 283–313). Providence, RI: A. W. Staats (Ed.), Human learning: Studies American Mathematical Society. extending conditioning principles to complex Pessoa, C. V. B. B., Huziwara, E. M., Perez, behavior (pp. 347–353). New York: Holt, W. F., Endemann, P., & Tomanari, G. Y. Rinehart, and Winston. (2009). Eye fixations to figures in a four- Sidowski, J. B., Wyckoff, L. B., & Tabory, L. choice situation with luminance balanced (1971). The influence of reinforcement and areas: Evaluating practice effects. Journal of punishment in a minimal social situation. In Eye Movement Research, 3,1–6. E. McGinnies & C. B. Ferster (Eds.), The Pierce, W. D., & Epling, W. F. (1980). What reinforcement of social behavior (pp. 34–38). happened to analysis in applied behavior Boston: Houghton-Mifflin. analysis? The Behavior Analyst, 3, 1–10. Skinner, B. F. (1938). The behavior of Pressey, S. L. (1926). A simple apparatus organisms: An experimental analysis. New which gives tests and scores and teaches. York: Appleton-Century-Crofts. School & Society, 23, 373–376. Skinner, B. F. (1950). Are theories of learning Pressey, S. L. (1927). A machine for automatic necessary? Psychological Review, 57, 193–216. teaching of drill material. School & Society, Skinner, B. F. (1954). The science of learning 25, 549–552. and the art of teaching. Harvard Educational Prokasy, W. F., Jr. (1956). The acquisition of Review, 24, 86–97. observing responses in the absence of Skinner, B. F. (1956). A case history in differential external reinforcement. Journal scientific method. American Psychologist, of Comparative and Physiological Psycholo- 11, 221–233. gy, 49, 131–134. Skinner, B. F. (1957). Verbal behavior. New Pubols, B. H., Jr. (1958). Delay of reinforce- York: Appleton-Century-Crofts. ment, response perseveration, and discrim- Skinner, B. F. (1972). Baby in a box. In ination reversal. Journal of Experimental Cumulative record (3rd ed., pp. 567–573). Psychology, 56, 32–40. New York: Appleton-Century-Crofts. Pubols, B. H., Jr. (1960). Incentive magnitude, (Original work published 1945) learning, and performance in animals. Skinner, B. F. (1979). The shaping of a Psychological Bulletin, 57, 89–115. behaviorist. New York: New York Uni- Pubols, B. H., Jr. (1962). Constant versus versity Press. variable delay of reinforcement. Journal of Skinner, B. F. (1983). A matter of consequenc- Comparative and Physiological Psychology, es. New York: New York University Press. 55, 52–56. Skinner, B. F., & Campbell, S. L. (1947). An Pyron, B., & Wyckoff, L. B. (1961). The rate automatic shocking-grid apparatus for con- of pecking by pigeons as a function of tinuous use. Journal of Comparative and amount and frequency of reinforcement. Physiological Psychology, 40, 305–307. American Journal of Psychology, 74,278– Sloane, H. (1964). Scramble patterns and 282. escape learning. Journal of the Experimental Reynolds, G. S. (1961). Attention in the Analysis of Behavior, 7,336. pigeon. Journal of the Experimental Analysis Smith, V. (2004). A delegate situation. The of Behavior, 4, 203–208. Crisis, 111(4), 30–35. Roalman, P. (1963, December 2). It’s a Spence, K. W. (1940). Continuous versus non- ‘‘teaching machine’’: How flipping through continuous interpretations of discrimination unusual book can help improve human learning. Psychological Review, 47, 271–288. relations. The National Observer,p.9. Spence, K. W. (1945). An experimental test of Rogers, C. R., & Skinner, B. F. (1956). Some the continuity and non-continuity theories issues concerning the control of human of discrimination learning. Journal of Ex- behavior. Science, 124, 1057–1066. perimental Psychology, 35, 253–266. Ross, T. (1938). The synthesis of intelligence— The teaching machines. (1960, November 7). Its implications. Psychological Review, 45, Time, 86, 19. Retrieved from http://www. 185–189. time.com/time/magazine/article/0,9171, Shahan, T. A., & Podlesnik, C. A. (2008). 826729-1,00.html Conditioned reinforcement value and resis- Teaching machines speed progress of the tance to change. Journal of the Experimental retarded. (1964, March 23). New York Analysis of Behavior, 89, 263–298. Times,p.26. Sidman, M. (1971). Reading and auditory- Tosti, D. T., & Kaufman, R. (2007). Who is visual equivalences. Journal of Speech and the ‘‘real’’ father of HPT? Performance Hearing Research, 14, 5–13. Improvement, 46(7), 5–8. 170 ROGELIO ESCOBAR & KENNON A. LATTAL

Touretzky, D. S., & Saksida, L. M. (1997). Wyckoff, L. B., Jr. (1954). A mathematical Operant conditioning in Skinnerbots. Adap- model and an electronic model for learning. tive Behavior, 5, 219–248. Psychological Review, 61, 89–97. Truax, C. B. (1966a). Reinforcement and Wyckoff, L. B., Jr. (1959). Toward a quanti- nonreinforcement in Rogerian psychothera- tative theory of secondary reinforcement. py. Journal of Abnormal Psychology, 71, Psychological Review, 66, 68–78. 1–9. Wyckoff, L. B., Jr. (1964). U.S. Patent Truax, C. B. (1966b). Some implications of No. 3,137,948. Washington, DC: U.S. Pat- behavior therapy for psychotherapy. Jour- ent and Trademark Office. nal of Counseling Psychology, 13, 160–170. Wyckoff, L. B., Jr. (1969). The role of The truth about those teaching machines. observing responses in discrimination learn- (1962, February). Changing Times, 16(2), ing: Part II. In D. P. Hendry (Ed.), 15–18. Conditioned reinforcement (pp. 237–260). Wallace, R. A. (1952, May 2). The maze Homewood, IL: Dorsey. solving computer. In Proceedings of the Wyckoff, L. B., Jr., & Berlin, J. I. (1963, 1952 Association for Computing Machinery August). The teaching of improved interper- national meeting (pp. 119–125). Retrieved sonal relations through programmed instruc- from http://portal.acm.org/citation.cfm?id5 tion for two people working together. Paper 609800 presented at the convention of the American Psychological Association, Philadelphia. Williams, B. A. (1994). Conditioned reinforce- Wyckoff, L. B., & Page, H. A. (1954). A grid ment: Experimental and theoretical issues. for administering shock. The American The Behavior Analyst, 17, 261–285. Journal of Psychology, 67, 154. Wyckoff, L. B., Jr. (1950). Resistance to Wyckoff, L. B., & Sidowski, J. B. (1955). extinction of a lever pressing response in Probability discrimination in a motor task. white rats as a function of number of Journal of Experimental Psychology, 50, reinforcers Unpublished master’s thesis, 225–231. Indiana University, Bloomington. Wyckoff, L. B., Sidowski, J. B., & Chambliss, Wyckoff, L. B. (1951a). Resistance to extinc- D. J. (1958). An experimental study of the tion of a lever pressing response in white relationship between secondary reinforcing rats as a function of number of reinforcers. and cue effects of a stimulus. Journal of Proceedings of the Indiana Academy of Comparative and Physiological Psychology, Science, 60,308. 51, 103–109. Wyckoff, L. B., Jr. (1951b). The role of Zeigler, H. P. (1958). Observing responses and observing responses in discrimination learn- discrimination learning. Unpublished doc- ing. Unpublished doctoral dissertation, In- toral dissertation, University of Wisconsin, diana University, Bloomington. Madison. Wyckoff, L. B., Jr. (1952). The role of Zeigler, H. P., & Wyckoff, L. B., Jr. (1961). observing responses in discrimination learn- Observing responses and discrimination ing: Part I. Psychological Review, 59, learning. The Quarterly Journal of Experi- 431–442. mental Psychology, 13, 129–140.