The Experimental Examination of Process

The present chapter attempts to illustrate the Whiteheadian concept of process within the frame of a psychological laboratory. The phenomenological (or macro) concept of process directs attention to a mutually connected series of events over time referring to one and the same subject and occasion. Even if only a limited section of such a series is cut out for study it is important to determine where the cut is made. Usually, however, the experimenter would prefer to include the entire process, from beginning to end, in the study. These two endpoints thus should be carefully defined.

Ontogeny may, of course, be considered as a process. Heinz Werner was intrigued by the formal correspondence between ontogeny and microgeny, the latter termed microgenesis in his vocabulary (Werner, 1956). For the present occasion, however, we have to focus on processes unfolding in a discernible present, open to inspection and skillful measurement. Many possibilities have been brought to notice and tried.

Two groups of methods have been used to study these microprocesses: (1) adaptive serials and (2) perceptgeneses. Adaptive serials are supposed to reflect how an individual adjusts to the demands of a new and strange situation, the process being registered with predetermined intervals. Perceptgeneses are reconstructions of the mostly hidden pulses assumed to construct our outside world. The reconstruction is accomplished by fragmenting the stimulus presentation. After a brief glimpse of the stimulus, just at the threshold of visibility, the viewer reports what he/she has seen. After a prolonged second presentation the viewer reports again, and so on until correct recognition. The series of reports constitutes a perceptgenesis.

1. Adaptive Serials

The usefulness of a serial approach was discovered long ago in experiments with mirror-drawing (Smith, 1952). A situation had been arranged where participants were requested to draw diagonals in squares, first using direct vision, thereafter viewing their movements in a mirror. The difference in drawing time between the two conditions was planned to be used as a measure of flexibility, as also assessed by seasoned clinicians. There was no correlation. But if, instead, you assumed a serial perspective, i.e., tried to follow how the initial drawing difficulties, be they small or considerable, were overcome as the subject tried again and again, a sizeable correlation could be ascertained between experimental and general behavior. People judged as inflexible or “adhesive” improved more slowly in the mirror-drawing situation.

1.1. Visual Afterimages

The distinctive character of serial change was first discovered in experiments with twins and projected visual afterimages. It was the change in size from one measurement to the following one that correlated within pairs of identical twins but not within pairs of fraternal, same-sexed twins (Smith, 1949). Afterimage serials used for diagnostic purposes, with compulsive neurotics particularly in focus, were later published by Kragh (1955). He was able to define when an adaptive process came to an end: this was when directed change ceased and turned into randomized change or relative stability.

Visual afterimages have continued to be important in process research over the years. Since the experimenters have made use of a standardized contrivance, one description of method will be sufficient. As their stimulus they have used a schematized red face projected on a dimly lit grey, transparent screen placed 40cm from the viewer. After fixation of the face for 20 seconds the screen was moved to a distance of 60 cm where the projected afterimage was measured by two movable rulers, handled by the viewer. The color was also described and the intensity estimated according to a scale demonstrated beforehand. The experiment was usually repeated 15-20 times (only 10 times with children). Traces of lingering afterimages were extinguished between trials.

A review of a long series of experiments was published in Smith (2001). These experiments have clearly demonstrated that visual afterimages are not fixed products of a stimulated retina but depend, to a high degree, on how they are appreciated by the participant. This state of things makes serial experimentation possible and fruitful. Not until the viewer has handled this subjective phenomenon in the world of objects (screen, rulers, etc.) will he/she be able to realize its true nature; in the case of cognitively immature children, never during the course of the experiment.

Afterimages produced by immature children differ from those of most adults. The color does not change from positive to negative, i.e., from red to blue-green in most of our experiments, and the size of the image does not expand in relation to an increased projection distance. If, thus, the child cannot clearly distinguish between the stimulus and the afterimage, the latter is likely to retain color as well as size when the projection distance is increased, just as a picture glued to the screen would do.

If nine-year-old children were told that the experimenter was able, by means of some abstruse trick, to produce an afterimage when they fixated the projection screen, the immature characteristics of the images were enhanced for the children who told that the images were the result of processes in their own eyes, i.e., not at all objective phenomena. It was even possible to influence adult participants in a similar manner, and not only unsophisticated persons.

When, in another study, medical students were instructed, in a semi-professional vocabulary, that afterimages were either strictly peripheral phenomena, products of the retina or, in a comparison group, influenced by neural centers adjacent to areas where personality characteristics were determined, distinctly different afterimage products emerged. The retinal instruction group limited the scope for individual involvement in afterimage production while the centralized theory facilitated such reactions. The latter images resembled the variegated, sometimes primitive, early reports in normal afterimage serials while the former ones more resembled the late, stereotyped products in the same series. The instructional influence, naturally, interacted with the participants’ personal characteristics.

A result, particularly interesting for general process theory, emerged when the experimenters tried to cause regressions by means of extraneous stimulation. An afterimage serial produced by means of the arrangements described above usually lasted 15-20 trials from beginning to end, the latter defined as the termination of structured change. If a loud signal was introduced, after seven trials the serial pattern from 1 to 7 was repeated, although in an abbreviated version. But when, in a new study, the signal was sounded late in the series, no such regression could be registered. Once established, the distinctive character of the process could not be repeated.

Given the susceptibility of afterimages to various individual characteristics it is not surprising that afterimage data have been used as diagnostic aids. Anxiety can thus be reflected in exaggeratedly large and/or dark images. But such manifestations may also be mitigated by defensive manoevers. One such intervention is change of afterimage color from blue-green to green, a hue resistant to over-saturation and blackening. In other cases the edges of the image seem fuzzy, thus preventing correct size estimation. The most drastic defensive measure is, of course, to prevent the image from appearing.

It remains to be asked whether serials as these correspond to the perceptgeneses produced by methods to be treated later in the present chapter. One rather drastic occasion of parallelisms concerns intermittent regressions in protocols of schizophrenic patients. In afterimage serials these regressions manifest themselves as sudden lapses from mature to immature images. In their perceptgeneses the progress from early to late interpretations is suddenly interrupted by empty phases, called zerophases, or by returns to interpretations registered in the introductory phases of the microprocess, phases often stamped by obviously personal reminiscences.

1.2. Spiral Aftereffects

Aftereffects of perceived movement have been utilized by Andersson (e.g. Andersson & Bengtsson, 1986). In the typical laboratory situation the participant is placed facing an arithmetic, black spiral with 2 1/2 turns drawn on a white background. Its diameter is 180 mm. Rotating 100 rpm it seems to contract inwards. After fixating it for 45 sec the participant is asked to look at a stationary circle, which replaces the spiral. The effect is an expanding or approaching circle. The participant has to indicate when this effect ceases. An ordinary test session, preceded by two trials for instruction purposes, includes ten trials.

These ten duration values serve as the foundation for a number of indices characterizing the aftereffect serial. Initial duration level refers to the first two trials and final level to the last two. In accordance with the statistical treatment of data from the Serial Color-Word Test (S-CWT, cf. below) two further units were used to describe the variability. R thus represents the linear change over ten trials and V the residual variability, not ascribable to linear increase or decrease of duration values.

Andersson (1983) has presented a comprehensive theoretical background to his research using the Serial Aftereffect Technique (SAT), a background colored by a psychoanalytic approach to ontogeneses, anxiety and defense against anxiety. Moving closer to his data base he presumes that aftereffect duration reproduces the balance between intraceptive and extraceptive determinants of the viewer’s perceptual world. Denial of the subjective contributions to the aftereffect results in brief, often hardly noticeable durations, whereas long durations represent the opposite tendency, allowing free rein to subjective determinants. Medium durations thus represent a stabilized balance between self and non-self.

Like the visual afterimages, the spiral aftereffects are sensitive to anxiety. Long effects were, according to Andersson, typical of participants with pronounced symptoms of anxiety. But in a study of fear of flying Amnér (1997) found that individuals with an insufficient differentiation of self and non-self, i.e., individuals at both the extreme ends of duration values, were more afflicted with fear of flying than individuals with medium values. Thus even total suppression of self factors was detrimental to handling fear of flying in a mature manner. Only those with medium values were capable of differentiating self from non-self, the latter represented by the aircraft.

In a similar vein Hansson and Rydén (1987) found that a gradual detachment of self from non-self in the course of aftereffect serials was typical of individuals who in another task were most successful in overcoming the distorting effects of a visual illusion.

1.3. The Serial Color-Word Test (S-CWT)

This is a method designed to find out how test subjects, step by step, accommodate to a contradiction. The basic idea dates back to E.R. Jaensch´s laboratory in Marburg which was later on exploited by Stroop (1935). The arrangement used here was first tried by Smith and Klein (1953). The test material consists of 100 color-words (red, blue, green, yellow) printed in incongruent hues (blue in red, green, or yellow, but not in blue, etc.). The participant is asked to name the color and disregard the printed word. The main unit is the time taken to name the color of 20 color-words. This gives five time values to the test sheet. The test is presented altogether five times.

Instead of using the time difference between naming a series of color slabs, on the one hand, and the color of incongruent color-words, on the other, as was the original purpose of the test, the serial approach implies a change in the focus of interest, to a study of how, over a series of attempts, the participant endeavors to master the introductory difficulties, be they modest or considerable. For this purpose the constructors used the basic statistics mentioned above to distinguish between two kinds of variation, one implying linear change (R), i.e. gradually increasing or decreasing time values, and the other the residual (V) when the linear variation was subtracted from the gross variance.

These two numerical values were calculated for each of the five subtrials. The participant was thus clocked for five time values in a subtrial. The two different aspects of variations in naming speed could then be combined to represent four adaptive types: low variation in both dimensions, low linear variation but high residual variation, etc. The dominating adaptive style over the five trials was estimated. In addition, the change over time including the entire sequence was calculated utilizing the same kind of typology as in the case of the subtests. This overall treatment could reveal if, for instance, an initial tendency to decelerating naming speed in the early subtrials was eventually replaced by a more irregular variation, or vice versa, and so on.

The test situation, including instructions and statistical treatment, are described in detail in the manual (Smith, Nyman, Hentschel, & Rubino, 2001) as are also attempts at utilizing the test results for diagnostic purposes. In the present context I will limit the presentation to two groups of comparisons between test data and clinical diagnostics: (1) the intermittent collapse of adaptive efforts in patients with psychosis; and (2) the adaptive peculiarities of hypertensive 68-year-old men who later suffer from either heart or brain infarction.

The test protocols of schizophrenic and other psychotic patients were characterized by sudden increases in reading time values as if the participant’s command of the test situation had completely collapsed. This adaptive style parallels that described above for psychotics taking the afterimage test and, in addition, the intermittent anxiety shocks breaking through their perceptgeneses. What for a period of time has seemed relatively normal in the adaptive process of these people is abruptly thwarted by an ostensible blocking of their grip of the situation. In the afterimage test the change could be defined as regression. Here it seems like an involuntary loss of reality control, with no outside disturbance to account for it.

Lena Andre-Pettersson et al. (2003) used S-CWT data in a group of 68-year-old men suffering from hypertension to predict their fate five years later. It was possible, on the basis of different adaptive styles, to foresee if they would either suffer a cardiac or cerebral injury, in many cases with fatal consequences. In both instances the participant intermittently demonstrated his ability to resume a normal, efficient adaptive style, but in the course of the adaptive process this proved to be impossible. The heart infarction people aggravated an adaptive style reflecting emotional problems, the cerebral infarction people a style more closely reflecting stress problems.

Besides these more dramatic instances of the diagnostic power of the S-CWT numerous studies have demonstrated correlations between, for instance, more ordinary variations in adaptive style and various forms of neurotic disturbances. Even perfectly normal variations in cognitive style can be traced in the S-CWT curves.

2. Perceptgeneses

The basics for the reconstruction of microgeneses (here: perceptgeneses) were outlined in the preceding text. Some requirements should be particularly underlined. Stimuli have to be meaningful: geometrical figures devoid of meaning or projective test figures with no definite signification, e.g., Rorschach cards, are more or less unavailing. The perceptgenesis needs a definable end stage. Stimuli should also be fresh to the viewer; familiar motifs produce truncated geneses or none at all. We have also learned that motifs made by professional artists are more fertile than amateurish products. Finally, dominating details may distort the reconstruction and make the definition of an end-stage difficult.

There is a tacit agreement between constructors of various perceptgenetic tests to unveil the process in 15-20 steps, starting just below the visibility threshold and ending when the viewer has given a correct report of the stimulus picture. But methods of reporting may vary. In the Defence Mechanism Test (DMT) the viewer is asked to draw or, at least, sketch what he/she has seen or glimpsed on the projection screen. The Meta-Contrast Technique (MCT) has no such requirement, mainly because the original test subjects, psychiatric patients, were often unwilling even to try their hand on a piece of paper. Common to all methods is an aim to represent the reporting in detail and to avoid the use of predetermined response alternatives.

The techniques of presentation vary somewhat between different tests and have also changed with the introduction of new electronic devices. This topic will be touched upon briefly with the test presentations below.

2.1. How to Understand a Perceptgenesis?

A perceptgenesis is not simply a series of reports differing only in degree of resemblance to a common stimulus theme. Early impressions in a series of presentations are not only more indistinct and uncertain, they are also more ambiguous, comprising more significations than impressions close to the endstage. The course of a perceptgenesis thus implies a progressive disqualification of competing developmental possibilities. At the same time the genesis gradually abandons its close contact with the personal reminiscences and becomes more objective, independent of the viewer’s private memories (see particularly Brown, 1991). This qualitative change over the course of a perceptgenesis entails a differential interpretation of early and late signs. Early signs must be considered as more fundamental, more typical of enduring response tendencies, late signs as more marked by the present stimulus context.

2.2. The Creative Functioning Test (CFT)

The stimuli used in this test are only mildly provocative, still lifes drawn by engaged artists, and thus intended to produce an optimal variety of interpretations. The testing starts in the usual perceptgenetic way with gradually prolonged exposure times and carefully recorded descriptions. When the viewer has reached the final, correct interpretation of the stimulus theme, the procedure is turned upside-down, the viewer not knowing. In the second half of the testing the viewer is thus presented with gradually abbreviated exposures of the stimulus.

There are two main response avenues in the second portion of the CFT. One is to stick to the interpretation now known to be the correct one, the other is to gradually return to more idiosyncratic alternatives, often the same as used in the first section of the test. It was assumed that the latter avenue would be typical of creative persons. This proposition, using a more refined categorization, has been tried in a variety of samples including scientists, children and adolescents, professional artists, etc., and compared with carefully selected criteria (see the manual by Smith & Carlsson, 2000).

In this context it may be particularly instructive to consider the qualitative difference between successive stages of the perceptgenesis. By returning to early interpretations of the motif the viewer attains an increased associative freedom, often termed metaphoric logic as compared to the more restricted, abstract logic of late process stages. By abandoning the correct interpretation the person not only avoids the risk of automatizing his/her constructive processes but also gains the advantage of a richer context of alternative process trajectories.

2.3. Tracing Defensive Strategies

In spite of waning interest in psychoanalysis, its theory of defense has attracted lasting and even increasing interest, as attested by a recent survey (Hentschel et al., 2004). Process researchers have been active in the area for almost half a century. By introducing threatening stimulation in perceptgenetic experiments they hoped to be able to represent existential problems in the miniature world of the laboratory. If it could be demonstrated that the manner in which a test person handles a threatening stimulus motif reflected his handling of experienced danger in real life, a set of efficient diagnostic tools could be constructed.

The DMT (Kragh, 1985) is the method most closely associated with classical psychoanalysis. The typical stimulus depicts a centrally placed young person, supposed to represent the viewer and called the hero in the manual, and a more peripherally placed threatening figure. To begin with the DMT was intended for sifting out unsuitable fighter plane pilots. The assumption guiding this grandiose attempt was that an excess of defensive reactions, particularly defenses of an inflexible variety, would hamper a pilot’s ability to master a dangerous situation. The assumption proved to be promising and led to further applications in other fields, e.g., handling of automobiles and lorries (Svensson & Trygg, 1994). A more decisive step into clinical psychology was taken by Elisabet Sundbom and her associates at Umeå University (e.g. Sundbom et al., 2002). Among defensive varieties let us just pick out repression implying that the threat lurking behind the hero is made impotent by turning it into a lifeless object.

The MCT represents a different device for introducing a threat into an innocuous situation. Here the viewer is first adapted to the picture of a young person sitting in a room with a prominent window in the background (or, in a parallel version, a young person standing beside an opening in a wall). After that a threatening face is made gradually more visible in the window, being introduced perceptgenetic fashion. This required two combined tachistoscopes until the entire test could be computerized and the stimuli presented on a TV screen. Already from the beginning MCT was intended for clinical application. Even if a representative defensive repertoire could be mapped by this test, the sign variants covered a broader specter, including depressive stereotypy, projective transformations and psychotic regressions. The MCT is described in detail in a manual (Smith, Johnson, Almgren, & Johanson, 2002). In a typical psychotic protocol the response series is suddenly interrupted by impressions of chaos or total blackness, a type of regression also described in the afterimage section above.

3. Concluding Remarks

The transfer of process in real life to a manageable laboratory situation is no utopian scheme but an attainable possibility. Several alternative devices for making the transfer possible have been outlined in the preceding text. Most of them were tried in extensive investigations and even adapted for practical use. There are two major types of processes offered for laboratory application: adaptive serials and perceptgeneses. Although they represent differing perspectives to process psychology, adaptive serials depicting them from without and perceptgeneses also from within, they share important formal characteristics. For a comprehensive description of process the two approaches might be combined.

Although the validity of several of the process methods has proved to be high, the reservations expressed by some critics should not be ignored. The most salient hesitation about process-oriented tests concerns the possibly distorting influence of the methodology itself. It does not seem natural, for instance, for a process to be incessantly interrupted for reporting or measurement. But this kind of criticism seems to take for granted that processes in a supposedly natural condition advance undisturbed. It has long been known, instead, that perception relies on repeated fixations, not on a long, undisturbed gaze. But the possibility that the experimental methodology may generate artefacts rather than the real thing should always be kept in mind.

The reader may have felt a certain confusion about the terminology, particularly about the use of microgenesis and perceptgenesis interchangeably. The term microgenesis was introduced by Heinz Werner as a substitute for the German Aktualgenese. It should be understood within a frame of reference of general psychology. Neither Werner nor his predecessor in Germany, Friedrich Sander, had any inclination for speculation in the orbit of personality. When in the year of 1960 Ulf Kragh and the present author suggested, in Sander’s home, that psychoanalysis might be a frame of reference for Aktualgenese, Sander was visibly shocked. Perceptgenesis, on the other hand, was a term reserved for a theory about perception as a peep-hole into personality. The connection between perception and personality was developmental, as originally stressed by Kragh. A perceptgenesis thus ideally proceeds from stages enclosed by personal experiences to a final stage of pure perception. This agrees with Brown’s (1991) approach.

Users of process methods as diagnostic tools are at risk of confronting a particular danger—that of converting process characteristics into substances. A particular sign of defense in the DMT or MCT, say isolation, may thus be designated as a trait in the subject’s personality outfit. There is nothing inherently wrong with realizing that isolation will probably recur as a typical behavior in other situations. But substance thinking is likely to block further analysis of the role of a particular behavioral style in the broader context of personality, an analysis necessary for the practical utilization of diagnostic findings. Substance thinking may seem to be an almost irresistible temptation. But in the long run it will tend to obstruct the development of functional theorizing in psychology, dealing with growth and reality construction.

Works Cited and Further Readings

Amnér, G. 1997. Fear of flying. A manifold phenomenon with various motivational roots (Lund, Studentlitteratur).

Andersson, A.L. 1983. “Cognitive growth, psychoanalytic conceptions of the mind, after-effect experience and disavowal as a defense against perceptgenetic threat,” Archives of Psychology, 135, 103-114.

Andersson, A.L. and M. Bengtsson. 1986. “Perceptgenetic defenses against anxiety and a threatened sense of self as seen in terms of the Spiral Aftereffect Technique,” in The Roots of Perception, edited by U. Hentschel, G. Smith, and J.G. Draguns (Amsterdam, North-Holland), 217-46.

André-Pettersson, L., B. Hagberg, L. Janzon, and G. Steen. 2002. “Adaptive behavior in stressful situations in relation to postinfarction mortality. Results from prospective cohort study: Men born in 1914, in Malmö, Sweden,” International Journal of Behavioral Medicine, 10, 79-92.

Brown, J.W. 1991. Self and process. Brain states and the conscious present (New York, Springer).

Hansson, S. B. and O. Rydén. 1987. “Relationship between differentiation and integration of self and nonself in terms of modes of perceptual adaptation,” Perceptual and Motor Skills, 64, 523-38.

Hentschel, U., G. Smith, J.G. Draguns, and W. Ehlers (eds.). 2004. Defense mechanisms. Theoretical research and clinical perspectives (Amsterdam, Elsevier).

Kragh, U. 1955. The Aktual-genetic model of perception and personality (Lund, Gleerup).

Kragh, U. 1985. Defence Mechanism Test, DMT. Manual (Stockholm, Persona).

Smith, G.J.W. 1949. Psychological studies in twin differences (Lund, Gleerup).

Smith, G.J.W. 1952. Interpretations of behavior sequences (Lund, Gleerup).

Smith, G.J.W. 2001. The process approach to personality. Perceptgenesen and kindred approaches in focus (New York, Kluwer/Plenum).

Smith, G.J.W. and G. S. Klein. 1953. “Cognitive controls in serial behavior patterns,” Journal of Personality, 22, 188-213.

Smith, G.J.W. and I. Carlsson. 2000. The Creative Functioning test (CFT). Manual (Lund University, Department of Psychology).

Smith, G.J.W., G.E. Nyman, U. Hentschel, and I.A. Rubino (2001). The Serial Color-Word Test (S-CWT). Manual (Frankfurt, Swets & Zeitlinger).

Smith, G.J.W., G. Johnson, P.-E. Almgren, and A.M. Johanson. 2002. MCT—the Meta-Contrast Technique. Manual (Lund University, Department of Psychology).

Stroop, J.R. 1935. “Studies of interference. Serial verbal reactions,” Journal of experimental Psychology, 18, 643-61.

Sundbom, E., M. Henningsson, U. Holm, S. Söderberg, and B. Evengård. 2002. “Possible impact of defenses and negative life events on patients with Chronic Fatigue Syndrom,” Psychological Reports, 91, 963-78.

Svensson, B. and L. Trygg. 1994. Personlighet, olycksbenägenhet och yrkesanpassning. [Personality, accident proneness, and work adaptation] (Stockholm, Almqvist & Wiksell).

Werner, H. 1956. “Microgenesis and aphasia,” Journal of Abnormal Social Psychology, 52, 347-53.

Author Information

Gudmund Smith
Professor emeritus, Department of Psychology
Lund University, Box 213, 22100, Lund, Sweden

How to Cite this Article

Smith, Gudmund, “The Experimental Examination of Process”, last modified 2008, The Whitehead Encyclopedia, Brian G. Henning and Joseph Petek (eds.), originally edited by Michel Weber and Will Desmond, URL = <>.