Popularizing the New Psychology in the Americas: From Scripture's Chain Reaction to the Polygraph, Biofeedback, and Computer Imaging Technologies

Table 1 (after Kusch, 1995) focuses on the power relationships which are present in various forms of psychological research. A superior or inferior location in the diagram, of the subject 'S' or experimenter 'E,' indicates a higher or lower social status within the experimental situation. Brackets '( )' indicate a severely depleted importance or merely formal presence of a participant within the experimental situation. This analysis complements Danziger's analysis which focuses on the knowledge goals of the psychological research.

Part I: Scripture and the Intradisciplinary Popularization of Mental Chronometry

By the mid-1890's, the New Psychology of mental contents was off to a running start in the Americas. Edward Scripture, now the founding head of the Yale laboratory (Sokal, 1980), helped familiarize a second generation of psychology students with the basic phenomena of mental chronometry. His 1898 text defined the new psychology as the "science of mental life which employs methods hitherto peculiar to the physical sciences" (p.ix).

Scripture's other popular work Thinking, Acting and Doing, (1895, 1907) showed, in detail, how the Runner's reaction time might be studied in the psychological laboratory (see fig 4).

Figure 4: Runner's Reaction time. Scripture, (1895).

The psychologist fires a gun which is connected to a chronoscope by means of a special switch located on the end of the muzzle. The "runner," who is tied at the waist by a rope, switches off the chronoscope by lunging forward in response to the sound of the pistol.

One of the more theatrical devices Scripture used was the "chain reaction" example, featuring the heads of state of various countries (fig 5).

Figure 5: Chain Reaction. Scripture, (1895).

When the seconds-hand of the American looking experimenter's watch is at the beginning of a minute, he presses the head of the first "head of state," and notes the time which passes before he is, in turn, touched by the last head of state. Dividing that time by the number of participants, yields the average reaction time. The message, here, is that reaction time phenomena can be empirically investigated even with commonly available equipment such as a pocket time piece.

As historians of psychology, we should note how this popularizing device is reminiscent of that applied to the reflex arc concept (which itself has a long philosophical and physiological tradition behind it). That tradition stretches back to the 'man and the fire' example used by Descartes (1662) and right up to the 'child and the flame' example used by both the Austrian physiologist Theodore Meynert (1885), and William James (1890).

Just as Scripture's books were becoming popular, the young John Dewey wrote a critique of the reflex arc concept which complained that such a small unit of analysis was too removed from daily psychological life to be of much relevance (Dewey, 1896). As if to allay such potential critiques of reaction time phenomena, Scripture carefully and graphically depict the supposed everyday nature of the reaction time (see fig 6).

Figure 6: Street Car Chain Reaction. Scripture, (1895).

Dewey, of course, never claimed that reflex arcs and reaction times don't exist, but only that such a unit of analysis was too small to have much ecological validity (i.e., relevance to human conduct). In order to complete the parallel between the popularization of reaction time and reflex arc concepts we need only note Howard Warren's thoroughly Americanized version of the reflex arc diagram (see fig 7).

Figure 7: Seeing and Acting. Warren, (1930).

This diagram occurred in slightly modified forms between 1918 and 1930. What could be more ecologically valid to an American than baseball?

What of the social relationships assumed by reaction time research? As indicated by the above descriptions of figures 3 and 4, the participants in the early psychological laboratory were referred to in terms of the activity they had been assigned. That is, what we would now call the "subject" was often referred to as the "discriminator," "actor," "associator," "perceiver," etc. What we would now call the experimenter was then referred to as the "manipulator," the "signaller," the "reader," etc. As time went on, these descriptive terms were replaced and the social roles of the participants became more firmly fixed.

Although, as Danziger (1990) points out, the Wundtian interchangeability of subjects and experimenter roles was already on the wane by 1890, the simple reaction time situation did continue to be depicted in the next generation of American introductory/experimental texts. Robert Givler's Psychology: The Science of Human Behavior (1920), for instance, showed just such an experimental situation (see fig 8). In describing this simple reaction time situation, Givler used the term "sender" and "receiver" for the two participants seated at the same table and separated by a screen.

Figure 8: Simple Reaction Time. Givler, (1920).

Figure 9: Choice Reaction Time. Murphy, (1935).

Similarly, Gardner Murphy (1935) includes a choice reaction time situation in his introductory text (see fig 9). By this time, however, the fixed division of labor had taken hold and it shows in the way the participants are described. Murphy describes the top participant as the "subject" and the bottom one as the "experimenter." He makes a point of the fact that the experimenter is in "another room of the building" and that the American made, Dunlap chronoscope (bottom right) is being used.

The American Context for Reaction Time and Mental Testing

To understand the American context for reaction time research, we must look at the influence of J. Mck. Cattell. The point here, is that even in such basic demonstrations as provided by Givler and Murphy, the equipment is Wundtian but the inspiration is Galtonian.

At Leipzig, Cattell was often his own data source. Cattell presented verbal stimuli and required verbal responses (i.e., the time it takes to name objects, shapes, letters, and also the time it takes to translate words). He also noticed that, for a given individual, the reading time varies for different languages.

After his graduation from Leipzig in 1886, Cattell spent a year in Britain where he met and was profoundly influenced by Francis Galton. Upon his return to America Cattell wasted no time setting up an explicit program for the study of individual differences. In 1890 he suggested that a series of ten tests be given to all the students of Experimental Psychology at the University of Pennsylvania and "all who present themselves" (Cattell, 1937 in R.I. Watson, 1979, p. 213).

Cattell's list of tests (including: The Rate of Movement, Sensation-areas, Pressure causing Pain, Least noticeable differences in Weight, Reaction-time for Sound, Time for naming Colours, Bi-section of a 50 cm. line, Judgement of 10 seconds time, Number of Letters remembered on once Hearing) combined those performed by Galton and those of the Leipzig laboratory. The motivation for taking down such data, however, was significantly different from the Wundtian investigation into the generalized human mind. As Boring (1950) pointed out, Cattell's psychology is a psychology of human capacity. It seeks "a description of human nature in respect of its range and variability" (p. 539).

Cattell went on to establish the Psychological Corporation (an organization dedicated to the promotion of mental testing) in 1921. Shortly before that, however, he was dismissed from his Columbia position for his pacifist views regarding WWI. Ironically, it was during WWI that extensive use was made of "mental testing" on groups of recruits in the form of the Army Alpha and Beta tests. A committee of American psychologists, under the chairmanship of R.M. Yerkes, had devised these tests at the Vineland Training School in New Jersey (see fig 10).

Figure 10: Alpha and Beta test committee (Upper left to right: F.L.Wells, G.M. Whipple, R.M. Yerkes (the chairman), W.V. Bingham, L. Terman. Lower, left to right: E. A. Doll, H. Godard, T. M. Haines).

The allocation of men was, traditionally, the task of the officer class, but Yerkes somehow persuaded the army to establish a school of military psychology at Fort Oglethorpe, Georgia. Most of these military psychology trainees were graduate students in education or psychology. Upon completion of training, they were dispersed to the various army camps in which over 1.7 million men were hurriedly subjected to mental testing.

The Army Alpha consisted of a printed booklet containing a variety of questions and tasks, all of which could be answered or carried out on the printed page.

Figure 11: Army Alpha Test. Engle, (1946).

The recruits in fig 11 have their hands raised so that, when instructed to do so, they will all start the test at the same moment. The aim of this group testing procedure was to quickly evaluate the relative performance of each man against the performance of the rest of the men (Gould, 1981; Minton, 1988; Fancher, 1985).

Between the wars, mental testing was increasingly applied to classroom evaluation of students. One of the best known tests designed for this purpose was the Otis Self-Administering Test of Mental Ability. A national `intelligence testing' movement was also set into motion by Yerkes and Terman under the auspices of the American World Book Company (Gould, 1981). Both the American public and the military was becoming more familiar with the concept of mental testing.

Between the years 1941-1946 over 10 million American men and women were given various forms of General Classification Tests (e.g., Army GCT, Navy GCT, etc.). Although this testing movement employed mostly paper and pencil methods, some aspects of the traditional experimental technology did survive within the new framework. During World War II, for instance, reaction time tasks had been incorporated into the process of selecting air force personnel.

Figure 12: Discrimination Reaction Time. Engle, (1945); Melton, (1947); Popplestone & McPherson, (1994).

Figure 13: Testing record of Harley Manning. From the Walter and Catharine Cox Miles Papers. Official photographs of the U.S. Army Air Forces. Archives of the History of American Psychology.

Fig 12 shows complex discrimination reaction times being performed by U.S. Air Force Cadets (Melton, 1947).This was one task in a standard battery of aptitude tests aimed at determining relative skill or ability. The advantage is that groups of subjects can be tested, with the record of performance being stored in the service file of each individual (fig 13). This sort of training continues up to the present in the form of flight simulation and police weapons training (e.g., shoot-don't-shoot tests). In both occupations, quick and accurate decisions can be important.

It is clear that in their use of the reaction time aspect of Wundt's physiological psychology, the Americans embraced the comparison of Individual Differences over the German tradition of pursuing the general aspects of the human mind. As we will see, however, the American use of the more psycho/physical aspect of the Wundtian tradition remained more loyal to the generalized human mind definition of the subject matter.

Part II: American Outcome of Wundt's Sensationist Aspect and the Extradisciplinary popularization of Psychology.

The second aspect of the Wundtian physiological psychology originated in the Weber/Fechner tradition of psychophysics. For Wundt, however, psychology is a separate science in which it is not necessary to refer to physiological processes. His emphasis was on interpreting such results as a relation between sensation and judgement rather than between physiological stimulus and psychological sensation. His aim was to establish the specific properties of psychological causation within fully conscious human experience. With the exception of Cornell, the American emphasis was more physiological but the methods used retained the flavour of the Wundtian tradition.

American variations on the sensationist theme

As an example of the Wundtian methods retained in this new context, Givler (1920) presents a traditional `just noticeable difference of weights' study (see fig 14).

Figure 14: Weight Judgement Experiment. Givler, (1920).

Figure 15: Paradoxical Heat Experiment. Murphy, (1935).

The manipulator (Givler in this case) presents the perceiver with a series of different sized weights by moving a rotation mechanism which is shielded from sight by a wooden screen. The perceiver's arm is limited in motion by a fixed rest. The verbal reports made by the perceiver consist of a prearranged wording (i.e., "heavier than," or "equal to," the stimulus lifted just before).

Similarly, Murphy provides a somewhat more complicated `two in one' example (see fig 15). On the one hand, it depicts the set up for a typical Wundtian sensation experiment. A metal instrument is heated by passing water through it to a temperature of 43'C (107'F) and is brought down lightly on the various squares of a grid, which had been previously stamped, onto the perceiver's hand. The perceiver says `warm' if he notes a warmth sensation; 'touch' if he becomes aware of touch but does not get warmth. The point of doing the experiment is to distinguish between skin locations which are differentially sensitive to warmth and touch, from those that cannot differentiate cold from touch (Heiser & McNair, 1934).

Karl Dallenbach (a former student of Titchener at Cornell) presented a number of variations on this sensationist theme between 1927 and 1937. One of those variations was the `paradoxical heat' experiment which used the two upright, pieces of equipment in the foreground of fig 15. Both warm and cold water is passed through alternate tubes of this `thermo-aestesiometer' apparatus. When a bare arm is laid across the grill there is physically the application of warm and cold stimulation to nearby areas of the skin. The psychological result of this combined physical stimulation is the paradoxical experience of heat (Burnett & Dallenbach, 1927; 1928; Dallenbach, 1927).

The sensationist line of American research retained the generalized human mind definition of the Wundtian tradition, but even from the time of Scripture's texts, it also contained a hint of a practical, applied, streak.

For example, Scripture (1895/ 1907) showed experimental situations devised to investigate the unsteadiness of the sportsman's aim, the speed of the pugilist's blows and the steadiness of the fencer's arm (see fig 16).

Figure 16: Unsteadiness of the Sportsman's aim. Scripture (1895).

These crude beginnings, however, are as far as things went within the confines of the Wundtian Physiological Psychology tradition.

Part of the later, more openly practical, influence on research, came from an individual who was imported to the U.S. by William James. Hugo Munsterberg met James at the first international congress of psychology in 1889. He took over the Harvard lab in 1892 and stayed until 1916 when he collapsed while lecturing.

Munsterberg and Joseph Jastrow (the first American psychology PhD.) set up a demonstration of psychological apparatus for the World's Fair at Chicago, 1893 (see figs 17-18).

Figure 17: Psychology Laboratory, World Fair, Chicago, 1893. American Archives of the History of Psychology.

Figure 18: Floor Plan of the Chicago World Fair Psychology Lab. American Archives of the History of Psychology.

Jastrow handled the apparatus, and Munsterberg wrote up a propaganda booklet for the occasion. For a nominal fee, there was a testing laboratory for those who were curious about their own mental capacities. As was the case with the Galtonian labs, the emphasis was on demonstrating the usefulness of the new empirical psychology to the public.

Munsterberg and his American students (Robach, Marston, Givler and Murphy), however, went far beyond this Galtonian gimmick to promote the cause. Together, they opened up the area of Applied psychology (including: Industrial, Business, and Legal applications).

One of these applied research areas led, in successive stages, to the development of the lie detector. William Marston (1924) claimed that changes of systolic blood pressure are correlated with changes in emotional states and that these changes might be used to detect deceit during questioning. He used a medical instrument called a sphygmomanometer as a measurement instrument (see fig 19).

Figure 19: Marston's Sphygmomanometic Lie Detection. Givler, (1920).

A rubber pressure bag, contained in a silk envelope, is wrapped around the subject's left arm in order to monitor the pressure of the brachial artery. The pulse in the radial artery of the left arm is also monitored during the procedure (Marston, 1924; Marston, King & Marston, 1931).

While Marston did his work with Harvard students in experimentally devised situations, others brought lie detector technology to actual criminal investigations. A team of three Californians: August Vollmer (police chief in Berkeley), John Larson and Leonarde Keeler (both police employees) developed the version of the lie detector currently in use (Block, 1977). Larson is shown in fig 20 with this `polygraph' instrument which measures changes in a suspect's heart beat, blood pressure, respiration, and electrical skin conductance.

Figure 20: Larson and the Polygraph. Larson, (1932).

The pens of the polygraph produce a permanent record of these measurements onto a moving strip of graph paper. The record of "relevant" questions is then compared with that of interspersed "neutral" questions (Larson, 1932).

During his career as a polygraph examiner, Leonarde Keeler worked on cases ranging from horse stealing to mass murder (E. Keeler, 1984).

Figure 21: Keeler and the Polygraph. Prothro & Teska, (1950).

Fig 21, originally from a Chicago newspaper report, shows Keeler, standing at the right, as he questions a prison guard about the Touhy prison break in 1942. When the photo just mentioned was reproduced in a well known general psychology text, it was accompanied by the following statement:

The argument behind the `lie detector' (that lying will produce an emotional response) also hinges on the assumption that emotional processes are automatic, autonomic and uncontrollable. This basic argument was graphically undermined, by animal research done by Neal Miller and his colleagues in the mid-to-late 1960s. They showed that various autonomic processes can be instrumentally conditioned.

Figure 22: Miller and DiCara with subject. Kagan & Havemann, (1972).

Fig 22 shows Miller and Leo DiCara with one of their experimental subjects. Miller & DiCara (1967) immobilized the rat's voluntary movements by administering the paralysing drug curare (see fig 23).

Figure 23: Miller's Curarized Rat experiment.

These conscious, but immobilized, rats received stimulation in their brain `pleasure centers' whenever their digestive muscles contracted or relaxed (depending on the assigned condition). These contractions were recorded on an EMG machine. This and other studies (Miller, 1968, 1969; Miller & Banuazizi, 1968; Miller & Dworkin, 1969) indicated that rats were able to retain their new abilities after recovering from the effects of the curare.

Up until that time, the separation of voluntary and autonomic processes was a matter of disciplinary entrenchment. During the period when Miller was developing the techniques used in his later studies (e.g., Miller, & Carmona, 1967) he encountered considerable resistance from his early research assistants, who, seemed to be doing little more than a half-hearted job.

Into the fray of this debate jumped Nigel Calder, a BBC television reporter. He received carte blanche to conduct interviews and filming of brain research in eight countries. It was by way of a now famous, staged for television replication, of earlier East Indian research, that Calder presented the much needed evidence regarding human control of autonomic processes to the Western audience.

In this replication of earlier research, conducted at the All-India Institute of Medical Sciences New Delhi, researchers placed a Yogi into cramped, but controlled, conditions to test his ability to adjust his autonomic processes (see fig 24).

Figure 24: Yogi in Control Box. Calder (1970).

The results indicated that the Yogi was able to decrease his oxygen intake by four times the baseline without adverse effects. Calder's much cited text combined such results with other parallel developments in the brain sciences (Calder, 1970).

The implication of the research into human control of autonomic processes was that there is an unsurmountable logical contradiction in advocating the use of polygraph technology for detecting deceit. In a trained or hardened individual, lying will not necessarily produce an emotional response. The critical literature has claimed to show just that effect of training (e.g, Lykken, 1981). The predominant conclusion seems to be that the effective use of a polygraph as a `lie detector' depends on the power of a modern superstition: it compels honesty because people believe it works, not because it really does (Kleinmmuntz & Szucko, 1984).

It is ironic, that, despite these developments, both the lie detector and the control of autonomic processes were exploited by popular American culture. The lie-detector, for instance, became the centre piece for a hit American television show where the host F. Lee Bailey (a famous trial lawyer) would offer guests a chance to prove whether or not they are telling the truth (see fig 25). Likewise, the control of autonomic processes, became popularized in the biofeedback technology of the mid-1970s.

Figure 25: F. Lee Bailey. Landy, (1984).

Figure 26: Alpha Wave Experiment (Kamiya, 1972).

Jeo Kamiya's investigations into Alpha brain wave training consisted of three phases; the second of which is presented in fig 26. During the baseline phase, the subject's EEG, in a totally darkened room with eyes closed, is taken. During the Alpha training phase, a tone from an intercom (right) gets louder or softer as alpha waves become more, or less, prevalent. Every two minutes, the tone cuts out and the subject opens her eyes to view a five second long, digital, score display. This score indicates the exact amount of alpha rhythms obtained in the previous interval.

In the assessment phase of the experiment, Kamiya tested each trained subject against a yoked control. He also collected phenomenological data regarding the subject's judgement of what is involved "internally" in handling the tone. Kamiya suggested that the control over Alpha brain waves was a matter of learning and a process of personal growth (Kamiya in Evans, 1976, p.126; Kamiya, 1972).

The popular dissemination of such findings led Californians to climb into plastic, sound proofed, enclosures with the expressed goal of obtaining the 'calming effects' of Alpha brain waves (see fig 27).

Figure 27: Group Biofeedback. Houston et al., (1979).

A more serious, and probably longer lived, application of biofeedback is in the clinical control of blood pressure and tension headaches (Brener & Kleinman, 1970; Richter-Heinrich & Miller, 1982). Relaxation of the forehead muscles usually ensures relaxation of scalp and neck muscles. Studies indicate that after four to eight training sessions, subjects learn to recognize the onset of tension and to reduce it without feedback from the machine (e.g., Tarler-Benlolo, 1978).

Although the modern forms of lie detector apparatus, and biofeedback technologies, are both outgrowths of the Americanized version of the Wundtian sensationist tradition, the assumed social relations between the psychologist and layperson are quite different. For the lie-detector, the `psychological expert' and `naive subject' (or suspect) relationship was assumed. This corresponds to the standard Applied Psychology relationship as listed in Table 1. In this case, the expertise of psychotechnology is supposed to triumph over any possible deceit. In the biofeedback application, however, the subject (now the self-learner) is given a progressively greater role in reaching their own goals (tranquilness, control of blood pressure, etc.). The experimenter is the facilitator but not the source of this self learning. In that sense, the biofeedback research situation is a return to the more egalitarian research relationships that were characteristic of the Wurzburg school (see Table 1). As in the Wurzburg tradition, the biofeedback experimenter treated the subject as a co-discoverer, or co-producer of the phenomena under study.

This return to a more egalitarian model of research was in complete agreement with what George Miller (1972) called "giving psychology away to the unwashed." Miller was resisting the then hegemonic behaviorist notion of research as a method of establishing complete control over the environment and behavior.

Part III: Behaviorism, Artificial Intelligence, and Integration.

In the behaviorist research situation the presence of the subject was a mere formality (see Table 1). That is, it mattered not what organism was placed into the learning situation because the goal of research was to obtain universal and often mathematical laws which applied to all organisms. This was the primary commonality between Watson's famous `give me a hundred healthy infants' speech, Hull's successive mathematical motivation formulas, and Skinner's claims for the identity of cross-species learning curves.

The 'Artificial Intelligence' movement pushed this conception of the all-powerful psychotechnical experimenter one step further by doing away with the organism altogether. The aim of such research was to produce machines that could: (1) perform tasks previously considered to require human intelligence; and (2) pass for people within limited settings.

The experimental 'subject' in such research was merely a mechanical devise or, sometimes, a theoretical (non-existent) "universal machine" which functioned according to a predesigned set of rules. Under such a research program, the `cutting edge' of psychotechnology was passed back into the hands of experts (including mathematicians, electrical engineers, computer scientists).

The Computational Metaphor vs the Expert Systems Movement

The Artificial Intelligence argument suggested that by programming computers to perform more and more intelligently, we might learn more about human intelligence as well. The most radical forms of AI (called strong AI) suggested that we might even be able to develop mechanical networks which worked on the same principles as neural networks and therefore could think. Whether in the weak or strong form, the A.I. position shares with Hobbes the basic assumption that `reasoning is nothing but reckoning.' This 'computational' metaphor retained some degree of influence until the mid-1980s when figures such as Howard Gardner provided carefully worded, but effective, critiques. Gardner suggested that an adequate understanding of human thought has to incorporate aspects of mechanical, biological, and social systems. His conclusion was that the search for an integrated understanding of human abilities will entail "important but enormously difficult undertakings for which traditional computational considerations may provide scant help" (1985, p. 387, emphasis added).

While the strong AI position has retained an institutional base (e.g., M.I.T., Carnegie-Mellon, etc.) it has also been theoretically marginalized and replaced with a rapidly expanding `integrated approach' to studying human cognition. As evidence of the negative component of the above statement, I offer the following quotations from Pratt's (1987) history of the so-called Artificial Intelligence research. In the closing chapters of his book Pratt contrasts the `brain project' (strong AI) with the onrush of a second major project in AI called the `expert systems' movement (one form of weak AI). Instead of trying to model the human brain, computers are best used for what they are: (1) complex mechanical calculating machines; and (2) digitized repositories for the storage and retrieval of human expertise. Pratt suggests that the expert systems movement "dominates the horizon current AI workers look towards" (p. 237).

Pratt's final few words on the issue combines this distinction with a further argument for some sort of qualitative difference between human thinking and that which is able to be operationalized in formal system machines:

Psychology and the Chess-playing Computer Example

This shift in the motivation behind AI research has been reflected in the way such technology has been used and popularized in psychology. The new, integrated approach to the study of human cognition (the positive component of my above argument) uses computerized imaging technology (e.g., Magnetic Resonance, Computer Axial Tomography, and Positron Emission Tomography) as expert systems which work in conjunction with phenomenological and observational methods. The approach holds great promise for providing a truly ecologically valid explanation of the developmental and functional features of higher mental processes.

This transition from computational to integrative analysis can be illustrated by looking at changes in the way that introductory texts have portrayed the 'chess playing machine vs human chess player' example (see figs 28-32).

That is, the movement from strong AI approach toward the expert system use of computers was reflected in the way the chess playing machine example was popularized in psychological texts.

Figure 28: Alex Bernstein playing chess with an IBM 704 computer. Lindgren et al, (1966).

Fig 28 shows Alex Bernstein, an experienced chess player who had programming access to an IBM 704 machine in 1957. At this time it was a laborious task for an experimenter to learn how the appropriate computer switches might be flipped in order to signify movements on a chess board. At least initially, the game itself, was a secondary concern to such technical difficulties (see fig 29).

Figure 29: Advice Taking Chess Program. Zobrist & Carlson (1973).

Computer technology was far from user friendly and early critics suggested that chess playing programs might never match the complexity of a competent human player (eg., Lindgren, et al., 1967, pp. 237-238). This was clearly a state of the art argument.

The cumbersome appearance of this early attempt does not, in fact, do justice to the important breakthrough which it represents. That breakthrough was a matter of providing the machine with mechanical rules of thumb which govern the `choice' among allowable moves of the chess pieces.

Pratt explains as follows: There is no problem in getting a machine to simulate a player of chess, if a `player' is taken to mean that which is able to move the pieces in accordance with the rules. The challenge comes in equipping the machine with the ability to `judge' the merits of one move over another. For a game of 40 moves (on each side), a machine calculating at the rate of one possible position per millionth of a second would require over 10 to the 95th power years to decide on its first move. To speed things up, a successful program would have to do more than merely perform exhaustive searches of all possible moves. It would it would have to include short-cut procedures.

Bernstein (1958) had surprised the critics by including such shortcuts in his program and producing a mechanical chess program that could rival the novice human player. Subsequent programs were provided with both faster processing hardware and more sophisticated, mechanical, forms of heuristic protocols.

Figure 30: A tough match between Chess Master and Chess Program (Rubin & McNeil, 1985).

By the early to mid-1970s, psychology texts began emphasizing the possibility of programming and teaching new moves to chess programs. In fig 30, an American chess master (C. Kalme) is shown providing his expertise to an 'advice taking' chess program. Such 'teaching' activities seemed to be narrowing the gap between human and machine chess playing abilities. The former 'state of the art' argument was falling apart.

To shore up the anti-computational camp, a more sound argument regarding the qualitative difference in kind between machine operations and human thought, began to be emphasized in psychology texts. Rubinstein (1977) for instance, suggested that a computer "cannot generate new concepts [or heuristics] of its own accord," and also lacks "such uniquely human characteristics as intuition" (p. 383).

Finally, in the early 1980s chess programs were fast enough to defeat all but the top few hundred human players in the world (see fig 31). They also included 'self-improvement' subprograms which searched out heuristics to improve performance. Under these circumstances, the state of the art argument was completely debunked.

This state of affairs, however, did not effect the more sound aspects of the 'qualitative difference' argument. Even within the limited arena of highly structured logic-governed games, the "methods used by the best computer chess playing program are entirely different than those used by the human chess master" (Rubin & McNeil, 1985, p. 220). That is, the highly sophisticated, mechanical, forms of heuristic protocols which the machine has been provided are not similar in any way to the contextualized, organic, and goal-directed intentions of the human player (Dreyfus, 1979; Dreyfus & Dreyfus, 1986).

The recognition of aspects of human thought that are more than formal has always been a tricky business. In 1980, however, John Searle pointed out that a human being, acting purely like a formal system program, could conceivably transcribe Chinese characters into another language with no knowledge of Chinese (Searle, 1980). In this case, as in all computer programs, the meaning of the symbols being transcribed is completely external to the entity that is transcribing the symbols (be that entity mechanical or human).

This `Chinese speaking room' formally imitates a Chinese speaker but the person in the room does not understand Chinese. This example proves that carrying out a formal program is not sufficient for understanding. In other words, only the weak form of A.I. (which postulates only simulation of human abilities) remains. One can make a computer look like it understands (or is intentional, or has feelings), but it does not.

Historically speaking, however, the most influential factor in the psychologist's abandonment of the computational metaphor was not the old argument from qualitative differences. In fact, the Achilles heal of general psychology has always been its lack of adequate categories to express the nature of these qualitative differences against those who would argue for biological, mechanical, or formal systems reduction.

Rather, the shift away from the computational metaphor was brought about by the introduction, and dissemination, of the expert system usage of computer technology in psychology (see fig 31).

Figure 31: Computers as mechanical expert systems (a medical database in this case).

In particular, a long acquaintance with teaching machines made psychology texts readily embrace the possibility of computer-assisted instruction (CAI). Similar applications of computerized expert systems (i.e., statistical calculating machines and computerized repositories for human knowledge) were already common in medicine, engineering, geology, law, manufacturing, and business (Pratt, 1987, pp. 136-173).

The recent widespread usage of computer imaging technology in psychology bears witnesses to the success of the expert systems applications in psychology. R. J. Haier and his colleagues, for instance, found that the brains of people who perform well on mental mathematical tasks consume less energy than the brains of poor performers (Haier et al., 1988). Haier measured brain activity with a technique called positron emission tomography, or PET. A PET scan records the amount of glucose used by brain cells. The harder the neurons work, the more sugar they metabolize. By using harmless, radioactively labelled glucose, it is possible to record an image of how hard the brain is working. Although we might assume that smart brains are hard working brains, the reverse appears to be true. That is, an expert experimental subject allocates less cognitive effort to a task than does an inexperienced subject.

Figure 32: Playing Several Chess Games Simultaneously (Dember, et al., 1984).

This PET scan research helps explain how a master chess player is able to carry out several games simultaneously (see fig 32). Such an extension of the PET scan data contains convergent validity with previous phenomenological and observational evidence. Chase and Simon (1973), for instance, found that chess masters were able to remember piece positions better than novices if the pieces they had been shown represented an arrangement that had been reached as a result of a game. That is, chess experts perceive game position arrangements in large units, or chunks, which necessitate less cognitive effort.

For the first time in the history of psychological practice we have a basis for integrating various forms of empirical data and research situations into an explanatory and ecologically relevant account of higher mental processes. All of the above empirical evidence has been present in both cognitive psychology texts and introductory texts for some years now. The presentation of this evidence in its proper integrated perspective seems the next logical move for general psychology.

Summary

Scripture, Givler and Murphy were all intradisciplinary popularizers of the New Psychology to an American audience. What survived the trip from Germany to North America? It was the emphasis on experimental (empirical) investigation into psychological processes. The particular slant to North American psychological work, however, was toward establishing the functional utility of psychological processes, and beyond that, in exploiting any potential military or commercial advantage of such knowledge.

In the mean time, the New Psychology was transformed from an academically based investigation of the generalized human mind, into a loosely connected band of psychotechnologies and research projects which often, at least initially, flew in the face of previous psychological doctrine. Given the applied focus, is it surprising that some of the extradisciplinary popularizers of these new technologies (e.g., Keeler and Calder) were not psychologists at all?

The fall of the computational metaphor was largely due to the widespread use and eventual recognition of computers as expert systems. Computer aided imaging techniques can now be combined with introspective, situational, and experimental evidence into an ecologically valid and explanatory account of higher mental processes.

Evidently, this transformation toward an ecologically valid psychotechnology did not occur quickly enough for Edward Scripture. In his contribution to the Murchison series, a contribution marked by its intentional dearth of biographical material, Scripture did provide one telling statement for posterity: "After twenty-five years of active service as a physician, I cannot read the usual book on psychology without being bored" (1936, p. 231). It is doubtful that Scripture would have made the same statement today.

Bibliography

Block, E. (1977). Lie Detectors: Their History and Use. New York: D. McKay Pub.

Bernstein, A. et al. (1958). A Chess-player program fo the IBM 704 computer. Proceedings of the Western Joint Computer Conference pp. 157-159.

Boring, E. (1950). A History of Experimental Psychology. (2nd ed.). New York: Appleton-Century-Crofts.

Brener, J., & Kleinman, R. (1970). Learned control of decreases in systolic blood pressure. Nature. Vol. 226, pp. 1063-1064. Bringmann, W. & Tweney, R. (Eds). (1980). Wundt Studies: A Centennial Collection. Toronto: C.J. Hogrefe.

Burnett, N. & Dallenbach, K. (1927). The experience of Heat. American J. of Psyc. Vol.38, pp.418-431.

Burnett, N. & Dallenbach, K. (1928). Heat intensity. American J. of Psychology, Vol.40, pp.484-494.

Calder, N. (1971). Mind of Man: An Investigation into Current Research on the Brain and Human Nature. New York: Viking.

Cattell, J. (1937). Retrospect: Psychology as a Profession. In R. Watson (Ed.). Basic Writings in the History of Psychology. (1979). New York: Oxford University Press.

Chase, W. & Simon, H. (1973). Perception in chess. Cognitive Psychology. Vol. 4, pp.271-280.

Dallenbach, K. (1927). The temperature spots and end organs. Amer. J. of Psych. Vol.39, pp.402-427.

Danziger, K. (1990). Constructing the Subject. Cambridge: Cambridge University Press.

Dember, W., Jenkins, J., & Teyler, T. (1984). General Psychology. (2nd. ed.). New Jersey: LEA.

Dewey, J. (1896). The reflex arc concept in psychology. Psychological Review. Vol. 3 , pp.357-370.

DiCara, L. & Miller, N. (1968). Instrumental learning of systolic blood pressure responses by curarized rats. Psychosomatic Medicine. Vol.30, pp.489-494.

DiCara, L. (1970). Learning in the autonomous nervous system. Scientific American, Vol. 222, pp.30-39.

Dreyfus, H. (1979). What Computers Can't Do: The Limits of Artificial Intelligence. (Rev ed., 1st edition published 1972). New York: Harper Colophon.

Dreyfus, H. & Dreyfus, S. (1986). Mind Over Matter. New York: Blackwell.

Engle, T.L. (1945). Psychology. New York: World Book Co.

Fancher, R. (1985). The Intelligence Men. New York: Norton.

Fancher, R. (1990). Pioneers of Psychology. New York: Norton.

Gardner, H. (1988). The Mind's New Science: A History of the Cognitive Revolution. New York: Basic Books.

Givler, C.G. (1920/22). Psychology: The Science of Human Behavior. New York: Harper.

Gould, S. (1981). The Mismeasure of Man. New York: Norton.

Haier, R. et al., (1988). Cortical glucose metabolic rate correlates of abstract reasoning and attention studied with positron emission tomography. Intelligence. Vol.12, 199-217.

Heron, W., Doane, B., & Scott, T. (1956). Visual disturbances after prolonged perceptual isolation. Can. J. of Psych, Vol.10, pp.13-16.

Heron, W. (1957). The pathology of boredom. Scientific American. Vol.196, pp.52-56.

Hebb, D. (1966). A textbook of psychology. Phil: W. B. Saunders.

Hebb, D. (1980). Understanding psychological man: A state-of-the-science report. Psychology Today. No.16, pp.52-53.

Heiser, F. & McNair, W. (1934). Stimulus-Pressure and thermal sensation. American J. of Psych. Vol.43, pp.580-589.

Holroyd, K. & Lazarus, R. (1982). Stress, coping and somatic adaptation. In L. Goldberger & S. Breznitz (Eds.). handbook of Stress: Theoretical and Clinical Aspects. NY: Free Press.

Houston, J. et al., (1979). Invitation to Psychology. New York: Academic Press.

James, W. (1890). Principles of Psychology. New York: Holt.

Kagan, J. & Havemann, E. (1972). Psychology. (2nd ed.). NY: HBJ.

Kamiya, J. (1968). Conscious control of brain waves. Psychology Today. No.1, pp.57-60.

Kamiya, J. (1969). Operant control of the EEG alpha rhythm and some of its reported effects on consciousness. In C. Tart (Ed.). Altered states of consciousness: A book of readings. New York: Wiley. pp.507-515

Kamiya, J. & Nowlis, D. (1970). The control of EEG alpha rhythms through auditory feedback and the associated mental activity. Psychophysiology. Vol.6, pp.476-483.

Kamiya, J. (1972). Conscious control of brain waves. Readings in Psychology Today. Del Mar: CRM Books, pp.415-418.

Kamiya, J. (1974). Autoregulation of the EEG alpha: A program for the study of consciousness. In M. Chase (Ed.). Perspectives in the Brain Sciences. L.A.: University of California.

Keeler, E. (1984). The Lie Detector Man: The Career and Cases of Leonarde Keeler. New York: Telshare Publ.

Kleinmuntz, B. & Szucko, J. (1984). Lie detection in ancient and modern times: A call for contemporary scientific study. American Psychologist. Vol.39, pp.766-777.

Kusch, M. (1995 in Press). Upcoming article. Isis. Vol.86.

Landy, F. (1984). Psychology: The Science of People. New Jersey: Prentice Hall.

Larson, J. (1932/69). Lying and its Detection Montclari: Patterson Smith.

Lykken, D. (1981). A Tremor in the Blood: Uses and Abuses of the Lie Detector. New York: McGraw-Hill.

Marston, W. (1924). A theory of Emotions and affection based upon systolic blood pressure studies. American Journal of Psychology. Vol.35, pp.469-506.

Marston, W.; King, C.; & Marston, E. (1931). Integrative Psychology. London: Kegan Paul.

Melton, A. (1947). Apparatus tests. In C. Flanagan (Ed.). Army Air Forces Aviation Psychology Program Report. (No.4). Washington: Government Printing Office.

Meynert, T. (1885/1968). Psychiatry. New York: Hafner.

Miller, N. & Carmona, A. (1967). Modification of a visceral response, salivation, in thirsty dogs, by instrumental training with water. J. of Comp. & Phys. Psyc. Vol.63, 1-6.

Miller, N, & DiCara, L. (1967). Instrumental learning of heart rate changes in curarized rats: Shaping and specificity to discriminative stimulus. J. of Comp. & Physiol. Psyc. Vol.63, pp.12-19.

Miller, N., & Banuazizi, A. (1968). Instrumental learning by curarized rats of a specific visceral response, intestinal or cardiac. J. of comp. & Physiological Psychology. Vol.65, pp.1-17.

Miller, N. (1969). Learning of visceral and glandular responses. Science. Vol.163, pp.434-445.

Miller, N & Dworkin, B. (1974). Visceral learning: Recent difficulties with curarized rats and significant problems for human research. In P. Obrist, et al. (Eds). Cardiovascular Psychophysiology. Chicago: Aldine.

Minton, H. (1988). Lewis M Terman. New York: NYU Press.

Richter-Heinrich, E. & Miller, N. (Eds.). (1982). Biofeedback: Basic Problems and Clinical Applications. NY: North-Holland.

Miller, G. (1972). On turning psychology over to the unwashed. Readings in Psychology Today. Del Mar: CRM Books, pp.11-18.

Murphy, G. (1935). A Briefer General Psychology. New York: Harper.

Popplestone, J. & McPherson, M. (1994). An Illustrated History of American Psychology. New York: Brown & Brown.

Pratt, V. (1987). Thinking Machines: The Evolution of Artificial Intelligence. New York: B. Blackwell.

Prothro, B. & Teska, P. (1950). Psychology: A biosocial Study of Behavior. New York: Ginn & Co.

Robach, A. (1952). History of American Psychology. New York: Library Publs.

Rubin, Z. & McNeil, E. (1985). Psychology: Being Human. (4th ed). New York: Harper & Row.

Rubinstein, J. (1975). The Study of Psychology. Guilford Conn: Duskin Publishing.

Scripture, E. (1895/1907). Thinking, Acting, and Doing. New York: Putnam's.

Scripture, E. (1898). The New Psychology. New York: Scribners.

Scripture, E. (1936). E.W. Scripture. In C. Murchison (Ed.). A history of psychology in Autobiography. Vol.III, pp.231-260.

Searle, J. (1982). Minds, brains, and programs. The Behavioral and Brain Sciences. Vol.3, pp. 417-457.

Sokal, M. (1980). Biographical approach to the history of psychology: The Career of E. W. Scripture. In J. Brozec, L. Pongratz (Eds). Historiography of Modern Psychology. pp.255-278.

Tarler-Benlolo, L. (1978). The role of relaxation in biofeedback training. Psych. Bull. Vol. 85, pp. 727-755.

Warren, H. (1930). Elements of Psychology (Rev. ed.). NY: Harper.

Zobrist, A & Carlson, F. (1973). Advice taking computer. Scientific American. Vol. 228, pp.92-105.

Illustrations

Table 1: Social Relationship Between Subject and Experimenter. (after Kusch, 1995).

Figure 1-3: Three Traditions of Psychological Research Danziger, (1990); Fancher, (1990).

Figure 4: Runner's Reaction time. Scripture, (1895).

Figure 5: Chain Reaction. Scripture, (1895).

Figure 6: Street Car Chain Reaction. Scripture, (1895).

Figure 7: Seeing and Acting. Warren, (1930).

Figure 8: Simple Reaction Time. Givler, (1920).

Figure 9: Choice Reaction Time. Murphy, (1935).

Figure 10: Alpha and Beta test committee (Upper left to right: F.L.Wells, G.M. Whipple, R.M. Yerkes (the chairman), W.V. Bingham, L. Terman. Lower, left to right: E. A. Doll, H. Godard, T. M. Haines).

Figure 11: Army Alpha Test. Engle, (1945).

Figure 12: Discrimination Reaction Time. Engle, (1945); Melton, (1947); Popplestone & McPherson, (1994).

Figure 13: Testing record of Harley Manning. From the Walter and Catharine Cox Miles Papers. Official photographs of the U.S. Army Air Forces. Archives of the History of American Psychology.

Figure 14: Weight Judgement Experiment. Givler, (1920).

Figure 15: Paradoxical Heat Experiment. Murphy, (1935).

Figure 16: Unsteadiness of the Sportsman's aim. Scripture (1895).

Figure 17: Psychology Laboratory, World Fair, Chicago, 1893. American Archives of the History of Psychology.

Figure 18: Floor Plan of the Chicago World Fair Psychology Lab. American Archives of the History of Psychology.

Figure 19: Marston's Sphygmomanometic Lie Detection. Givler, (1920).

Figure 20: Larson and the Polygraph. Larson, (1932).

Figure 21: Keeler and the Polygraph. Prothro & Teska, (1950).

Figure 22: Miller and DiCara with subject. Kagan & Havemann, (1972).

Figure 23: Miller's Curarized Rat experiment.

Figure 24: Yogi in Control Box. Calder (1970).

Figure 25: F. Lee Bailey. Landy, (1984).

Figure 26: Alpha Wave Experiment. Kamiya, (1972).

Figure 27: Group Biofeedback. Houston et al., (1979).

Figure 28: Alex Bernstein playing chess with an IBM 704 computer. Lindgren et al, (1966).

Figure 29: Advice Taking Chess Program. Zobrist & Carlson (1973).

Figure 30: A tough match between Chess Master and Chess Program. Rubin & McNeil (1985).

Figure 31: Computers as mechanical expert systems (medical database example).

Figure 32: Playing Several Chess Games Simultaneously. Dember, et al., (1984).