Gestalt theory: its past, stranding and future...

Abstract

This article focuses on the contributions of the founders of Gestalt theory, not only for the high value they carried even back then, but also for the strong relevance they have today. The main purpose is to point to the deficient, even wrong transmission of this perspective particularly in the past 50 years and to highlight its potential to connect the immense amount of accumulated but disconnected scientific facts and pieces within psychology as of today. The first part of this article discusses Max Wertheimer’s important 1912 “phi phenomenon” article and recounts the Gestalt theorists’ launch of their influential journal Psychologische Forschung in 1922, the rise of the oppressive and violent Nazi regime in Germany, and the resulting emigration of the Gestalt founders to the US where they had to face a radically different perspective to psychology. The second part discusses the main postulates of the theory, focusing on how the movement emerged, its main theoretical perspective, and its work on perception. In a second and third article (Mungan, 2021a; 2021b), I will review their intriguing research and conceptualizations on memory and productive thinking, respectively. Hence, the current article should be read as the first in a series of three.

Gestalt Theory:  Its Past, Stranding, and Future...

Giovanni Strazza

(From: Wiki Commons; Credits: Scientific American “Unveiling the Illusion”)

"Homo sum, humani nihil a me alienum puto"

[I am human, and I think nothing human is alien to me]

Publius Terentius Afer

1. Introduction

As an extensively misquoted and misunderstood theory, I felt the need to write this comprehensive article to bring to the surface what Gestalt theory --in all its simplicity as well as complexity-- is actually about. This interesting almost enigmatic theory emerged in the 1910s in Germany, was about to become an alternative to Wundt’s structuralism and the Anglo-North American functionalist schools (including behaviorism) in the 1920s, only to disappear from mainstream psychology around the 1950s and 60s when the US-based “cognitive revolution” took over all there was.

There were two factors that pushed me to “excavate” this theory. One of these was the consistent curiosity that the “Gestalt School” (as it is mostly referred to in psychology textbooks) would trigger in students. Yet, this curiosity would remain "up in the air" because something major was missing in mainstream textbook descriptions. All that was presented was a descriptive, “ungrounded” theory that listed certain laws of perceptual organization, no more. Another factor was that despite being pushed away and turned into a caricaturized, decomposed and decontextualized theory that would cover no more than 1-2 pages in a given textbook of psychology, it nonetheless obstinately resisted to be fully buried among the "dusty pages of history" as it happened to structuralism or behaviorism. Perhaps the strongest indication of this survival were the two 2012 articles which Wagemans at al. wrote for the centennial of the theory. Whereas the first article (ca. 90 pages) focuses on the fundamental propositions of the Gestalt theory, the second one (ca. 35 pages) analyzes how these propositions relate to today’s primarily vision science research. As such, the Wagemans et al. articles clearly served as a reminder, even a wake-up call, rather than a memoir of some past and “passé” theory.

What is most essential to understand any theory, is to read the actual works of the theorists. With respect to Gestalt theory, at least two things stand out. Firstly, unlike, say, Freud’s theory, Gestalt theory was prominently launched not by one but by three founders. Secondly, many of the sources were written in German, the mother tongue of the theorists, and quite few of the sources that were translated into English -which became the new world language of science since the mid-20th century-, were abbreviated, even incorrectly translated (cf. Ellis, 1938). Some of their most critical works have not been translated up until 2012 (e.g., Wertheimer/Spillmann, 2012) and some are yet to be translated (see Steinman, Pizlo, & Pizlo, 2000). With the Nazis coming to power in 1933, the founders had to flee Germany at a point where their theoretical and empirical work was steadily growing. Hence their ultimate move to the United States caused a harsh discontinuance not only in terms of empirical output but also in terms of the challenge of translating their conceptualizations into a non-native language.

Yet, I also think that in addition to the language obstacle there was an obstinate philosophical and epistemological barrier that made it impossible for them to communicate their phenomenological groundings to a pragmatist, functionalistic, behavioristic Anglo-North American climate. Wolfgang Köhler’s 1959 address at the 66th Annual Conference of the American Psychological Association (APA), as the president of the association, elegantly expresses this profound difference in how science is to be understood. In his speech, he speaks of the fussiness of the Anglo-North American perspective to a point that it becomes myopic, and of its lack of conceptual boldness. He observes that very meticulous work had been done with great skill in the Anglo-North American tradition of psychology, but that, because of this meticulousness, anything that could not yet be subjected to such methodology was left “unsearched”.

In fact, as of today, I might say that there is a much greater danger caused by a prevailing and invasive North American-born perspective of forcing scientists to publish as much as possible as fast as possible. Moreover, what is desired is that published articles can be read “instantly” and provide information that is almost as simple as a newspaper headline. This view has become so much of a norm that no even dares to question it. Even worse, today there are clues that the experimental rigor that Köhler praised is also eroding (see. Zwaan, Etz, Lucas et al., 2018). The “publish or perish” dictum has turned fiercer particularly in the past 20 years, which in turn has jeopardized the possibility of (1) asking daring questions and doing “fringe” research, the kind of research which typically brings about scientific breakthroughs, (2) examining and reporting the phenomenal experiences of the participants in addition to the aggregated numerical data summaries in experimental studies. I will re-address these concerns again in the final section of the article since their critical importance, I believe, will become more evident after the exposition of this theory.

When reading the original sources of Gestalt theory, I was perplexed at the depth and continuing relevancy of their main tenets. I think the most surprising aspect was to see that the important questions posed by the theorists still remain unanswered, even “unasked”. But foremost, I was impressed to encounter an extremely broad theory that was not at all restricted to perception or even cognition in general. Instead, there was a theory that has the potential to connect for the first time all the disjointed and fragmented fields within psychology.

The article consists of two main parts (see Table 1). The first part describes the coming about of the theory and its first appearance in Turkey. The second part presents its main postulates as well as its philosophical grounding. As much as its figure-ground and grouping principles are widely described in the psychology literature, its conceptual, philosophical outlook is either mentioned in passing or completely overlooked. I will conclude with a general evaluation.

Table 1

Outline of the Article

CONTENTS

1.Introduction

Table 1 9

Outline of the Article 9

2. The Emergence of Gestalt Theory 10

2. 1 Max Wertheimer's 1912 Article on the Phi Phenomenon 10

2.2 Journals: Zeitschrift für Psychologie and the founding of Psychologische Forschung in 1922 14

2.3 Nazi Germany, Moving to the United States, Followers of Theory in America 16

2.4 An Article on Gestalt in 1938's Turkey: "Geştalt Nazariyesi" 20

3. Gestalt Theory 22

3.1 Main Tenets of the Theory 22

3.1.1 Gestaltqualitäten, Figure-Ground and Gestalt Principles: 22

3.1.2 Gestalt Principles of Grouping 31

3.1.3 The Role of Experience and Learning 52

3.1.4 Overall Conclusions from Wertheimer’s Seminal 1923 Article 54

3.1.5 Koffka's Discussions on Figure-Ground Relationships and Gottschaldt's 55

3.2 Conceptual Propositions of Gestalt Theory 61

3.2.1 Parts and the Whole: Microscopic and Macroscopic View 62

3.2.2 Context, Field, and Structure Against Randomness 66

3.2.3 Isomorphism 67

3.2.4 Geographical Environment, Phenomenal Environment, and Meaning and Values of Living Beings 68

4. In Place of a Conclusion: What Gestalt Theory Is and What It is Not… 71

5. References 77

2.The Emergence of Gestalt Theory

2. 1 Max Wertheimer's 1912 Article on the Phi Phenomenon

In general, sources point to Max Wertheimer's 104-page 1912 article titled “Experimentelle Studien über das Sehen von Bewegung (Experimental Studies on the Seeing of Motion[1] )”as the article that launched Gestalt theory[2]. His article deals with the issue of "apparent motion", a phenomenon already well-known not only in the world of cinema of the time but also in experimental laboratories of psychology. But there was an obvious difference in Wertheimer's analysis of this phenomenon. Unlike previous researchers who demonstrated this effect with a variety of stimuli (moving light beams, object, etc.) and a variety of means (stroboscopic display, tachistoscopic display), Wertheimer focused on the experiential side of seeing motion. In this section, I will highlight some of the critical points of this "founding" article, without going too much into the kind of details that a vision researcher might be interested in (for that, the reader is referred to Wertheimer/Spillmann, 1912/2012).

In his article, Wertheimer first described the current literature on apparent motion and then presented a series of experiments designed to better understand which critical parameters the effect depended on. In one of his experiments he used two equidistant horizontal lines which were flashed in succession (e.g., lines a and b in Figure 1). He found that at high speed, participants experienced two simultaneous horizontal lines whereas at low speed, they perceived two separate lines that flashed one after the other (which was what was indeed happening). However, at a critical 'intermediate speed’, something different happened: the experience of a single line that moved from a to b (or b to a). In a series of rigorous experiments, he then tested whether other factors played a role in this apparent motion experience. He looked at the specific configuration of the two stimuli (e.g., horizontal vs. vertical vs. radial location change), the role of participants’ attention (by manipulating how and where to focus their attention), and whether this illusion could be explained by eye movements. In each setup, some of the trials presented two separate stimuli flashed one after the other, others, a single stimulus that indeed moved from one location to the other. Participants’ task was to report whether they experienced the successive flashing of two separate stimuli or a single moving stimulus. Critically, Wertheimer not only collected data on whether or not participants succeeded in distinguishing between genuine motion and apparent motion, but he also asked participants to describe their visual experience, in other words, their phenomenal experience.

Wertheimer found that at a critical interstimulus interval (between the first and the second stimulus) of approximately 100 milliseconds[3], participants were no longer able to distinguish between actual and apparent motions. In fact, Wertheimer notes that in the rare occasion where participants would report a difference it was typically in the opposite direction, reporting a "stronger, more energetic and more obvious" motion in trials with apparent rather than real motion. In these trials, they would also frequently perceive the motion before they would perceive any of the two lines, and sometimes even, they would only see motion per se, without seeing the lines at all.

In another experiment, other than fixation instructions, a setup was devised to fully rule out the role of eye movements. Here, participants would see the simultaneous appearance of lines b and c after which, following their offsets, lines a and d would follow, again simultaneously (Figure 1). Since moving one’s eyes simultaneously in two opposite directions is impossible, any report of an apparent motion must have been due to other reasons. Even if one could gaze, say, at line b and then line a (or, line c to d), this cannot explain participants’ experience of lines b and c simultaneously “moving” to position a and d. In other words, this finding showed that the illusion of motion is not caused by eye movements.

Figure 1 (Prepared by the author after Wertheimer, 1912)

Had he stopped right there, Wertheimer would not have been any different from the other experimental psychologists of his time (and of today, for that matter). However, he did not and instead went on to search for what exactly was happening. Since in the no-motion trials there was only one stimulus and then another one in a different location, it was impossible for the participants to have ever sensed the intermediate stages that exist when a stimulus indeed moves from one point to the other in space-time[4]. In other words, unlike in the real motion trials, in the no-motion trials, those intermediate stages were nonexistent in the physical world. So, why then did participants report to have seen a moving single object? Is this experience of apparent motion simply a mental act of "logical inference", of “filling in” or is it a very real, physiological experience? Although, at first glance, it might look as if there is not much of a difference between the two, there is a major difference: the first is a judgmental event that happens "after the fact", whereas the other indicates a concurring phenomenal experience of motion that can be described as "seeing moment-by-moment motion". Participants’ phenomenal reports hinted to the latter rather than the former. Through a meticulous process of reasoning, Wertheimer concludes that what happened was a real-time experience of motion, hence he coined the effect the "phi effect" and symbolized it with the Greek letter Φ. When presenting line a and then line b in 100 ms succession, participants would not perceive two consecutive, isolated stimulus images in the form of “a then b”, they would also not perceive a still form composed of parts a and b in the form of “a and b”. Instead, according to Wertheimer, what is seen is "aΦb", that is, a singular unit of motion. The Φ dimension is a perceptual state that does not exist in external space but occurs under special conditions in physical space, such that now a and b are in a bonding state with each other in the phenomenal (and likely physiological[5]) space.

After a series of cleverly designed experiments, Wertheimer emphasizes that the phi effect also occurs when, for example, the shapes or colors of stimuli a and b are changed, the only difference being that the participants now declared that the shape or color changed during motion. As mentioned earlier, he also reflected on the role of attention. When participants were asked to focus their attention on a or the area between a and b, the phi effect still occurred. But it weakened or even disappear when asked to focus on b. In another experiment, a third fixed stimulus c was placed inside or outside the 'motion zone' of a and b. Wertheimer found that even when the third stimulus was in the 'motion zone', it did not perceptually enter into a motion relationship with stimulus a, that is, an aΦc did not emerge. Furthermore, instructing participants to direct their attention to places other than b, did not change the results. In the conclusion section of his article, Wertheimer uses the term "Gestaltqualität[6]" (gestalt quality) to describe this “boundedness”.

Wertheimer’s 1912 article is a stunning reading from start to end. As we mentioned before, the phenomenon of "apparent movement" was known in both the cinematic and experimental world. But no one approached the phenomenon the way Wertheimer did. This difference, I believe, is due to his firm philosophical training, and relatedly his close relationship to his mentor Carl Stumpf, who in turn was a student of Brentano. Also, contrary to the general claim that Gestalt psychology only did a series of demonstrations rather than experimentations, the entire paper proves otherwise showing both meticulous experimental setups as well as an intense curiosity and in-depth questioning of what is going on by including participants experience.

2.2 Journals: Zeitschrift für Psychologie and the founding of Psychologische Forschung in 1922

Wertheimer (1880 – 1943), Wolfgang Köhler (1887 – 1967), and Kurt Koffka (1886 – 1941) were the three founders of the Gestalt movement. All three received their training from Carl Stumpf, who, as a phenomenological philosopher and psychologist, managed to combine these two fields through the perception of sound and especially music. It is perhaps for this reason that all three frequently give examples from music in their writings. At that time, they published their articles in the Zeitschrift für Psychologie under the editorship of Hermann von Ebbinghaus, the leading psychology journal of the period (probably the only one in Germany). Some of their articles were written in response to the mainstream researchers, and we understand that these three young people began to raise their criticisms against the empirist[7] view that constituted the mainstream of the period. Their most central objection was to the unquestioned (and still unquestioned) presupposition that we can understand something as a whole by simply understanding each of its parts and their local relations and interactions. When Wertheimer (1880 – 1943), Wolfgang Köhler (1887 – 1967), and Kurt Koffka (1886 – 1941) were the three founders of the Gestalt movement. All three received their training from Carl Stumpf, who, as a phenomenological philosopher and psychologist, managed to combine these two fields through the perception of sound and especially music. It is perhaps for this reason that all three frequently give examples from music in their writings. At that time, they published their articles in the Zeitschrift für Psychologie under the editorship of Hermann von Ebbinghaus, the leading psychology journal of the period (probably the only one in Germany). Some of their articles were written in response to the mainstream researchers, and we understand that these three young people began to raise their criticisms against the empirist view that constituted the mainstream of the period. Their most central objection was to the unquestioned (and still unquestioned) presupposition that we can understand something as a whole by simply understanding each of its parts and their local relations and interactions. When psychology emerged as a young science in the second half of the 19th century, this was its "founding ideology". According to this view, all mental and behavioral experiences are built gradually from single sensations which come together to form perceptions, memories, and ideas. In other words, the most complex abilities are formed step by step, starting with the simplest sensations which through successive experiences get connected to each other only to generate increasingly large "complexes". Brilliant research published in Zeitschrift für Psychologie under Ebbinghaus's supervision constantly circled around this view. Naturally, articles that examined the most basic and "molecular" sensations to the finest detail covered most of the space in this journal. In such a mental climate, the three founders of Gestalt theory spoke out their criticisms and defended their ideas by responding to objections to their views. Finally, in 1922, Wertheimer, Koffka, and Köhler (together with the neurologist Kurt Goldstein and the psychiatrist Hans Walter Gruhle) founded their own journal Psychologische Forschung: Zeitschrift für Psychologie und ihre Grenzwissenschaften (Psychological Research: The Journal for Psychology and Its Bordering Sciences).

In his seminal 1923 article Untersuchungen zur Lehre von der Gestalt[8] , Wertheimer proposed what are today known as the most basic principles of perceptual organization. In the following sections I will review the main propositions of this article. What is critical here was that there was finally a breaking away from the associationistic viewpoint (even ideology) to a completely different conceptualization. Even the term "bordering sciences" in the journal's name can be seen as a sign of a comprehensiveness that this new perspective was striving for. The journal continued until 1974 but just 11 years after its 1922 start, Germany witnessed the most unfortunate and brutal period of fascism in its history, which led to a full collapse of what was until then likely to be Europe’s creative driving force in science and art. Wertheimer (who was also a close friend of Einstein) was Jewish on both sides of his family and Koffka half Jewish. Köhler, on the other hand, was the single non-Jewish German university professor who loudly spoke out against the Nazis. Sadly, it seems that all other non-Jewish academics in one way or the other submitted to the new regime without speaking out against the ever increasing oppression and mass dismissals in academia. Despite being regarded as one of the most respected scientists of the time at the Psychology Institute in Berlin, Köhler, too, faced extensive pressure, but more so the students who continued attending his lectures.

2.3 Nazi Germany, Moving to the United States, Followers of Theory in America

Max Wertheimer, Kurt Koffka, and Wolfgang Köhler all left Germany one by one[9]. Köhler was the last to leave, keeping his position until 1935 without compromising his firm opposition to Nazi ideology. Once in the United States, the newly thriving center of science, these three great minds attracted a lot of attention across the most prestigious universities there. In the end, all three were provided positions at different universities. Yet, they were, after all, torn apart from their own lands and intellectual habitus and forced to re-explain and re-convey all their accumulated knowledge in a different language and a completely different cultural environment. Despite this, they were able to influence North American psychology, which was then dominated by behaviorism, and this influence lasted at least up until the 1950s. Köhler, for example, was invited to Harvard to deliver the William James talk series. Koffka wrote, in my opinion, one of the best textbooks of psychology (1936), and Wertheimer, who died unexpectedly in 1943, is said to have made a strong impression despite his non-fluent English due to his inspiring personality and speeches (see Mandler, 2007). His book, Productive Thinking, had just been completed and was published two years after his death. In addition to looking at problems in geometry and algebra, the book examines the ways of creative thinking of Galileo, Gauss and his close friend Einstein within Gestalt theoretic framework. This mind-opening content makes it perfectly clear why the book was re-published repeatedly since its first edition in 1945. The most recent edition came out in 2020 with an extensive preface by Viktor Sarris.

A few points need to be made here. First of all, despite all the respect they received, the prevalent scientific ‘ideology’ they found themselves in was behaviorism. Hence, they were faced with an associationism in its extremist form which very much disliked European, continental ‘mentalism’. To try to convince such an audience of the importance of scrutinizing phenomenal experience in order to better understand human nature was utterly impossible. Even people like Karl Lashley who broke off from the behaviorist school and became part of what was later called the “biology revolution” remained strictly aloof to considering phenomenal experience as important. Even though this remains unmentioned, I believe that the main reason for this discord was a profound difference in ontological terms. Whereas the Anglo-North American perspective openly or latently committed to a monist materialist perspective, I believe that the Gestaltists were closer to a double aspect monism as proposed by Spinoza. The latter kind values the mental as much as the material and does not believe that the former can be reduced to or fully explained by the latter. Within this light, we can say that in the United States, the three Gestalt founders were faced with the deeply grounded discomfort of the Anglo-North American viewpoint against phenomenology. As mentioned earlier, Köhler elegantly expressed this in his 1959 address at the American Psychological Association. And in this climate, they could not find too many doctoral students at the universities they were placed in. Thus the inspiring studies which they could carry out with talented doctoral students during their homeland period from the 1910s to 1933 now came to a halt.

Among those doctoral students was Hedwig von Restorff, who quit psychology and started to work at the medical faculty after Köhler left in 1935, and the Institute in Berlin came under Nazi control (MacLeod, 2020). Karl Duncker, who wrote a very creative and influential thesis on thinking, was from a socialist family. Köhler succeeded in protecting him from being dismissed so long as he himself remained head of the institute. In 1935, Köhler refused to take the "Hitler oath" and was therefore dismissed. He left Germany together with Duncker, who after traveling to a few countries, also settled at a university in the United States and continued to work productively until his suicide in 1940 at the age of 38. Yet another successor was Wertheimer’s PhD student Rudolf Arnheim who made significant contributions to the field of visual perception, art and aesthetics, and Köhler’s student Hans Wallach who studied visual and auditory perception. Another noteworthy successor was Erich Goldmeier, who wrote one of the most inspiring books on memory (cf. Mungan, 2021a). Goldmeier is probably one of the least known Gestalt theorists. His mother and father were killed in Auschwitz and he probably settled in the United States as soon as the Nazis came to power.

Some successors in Germany who did not leave German academia upon its Nazification, continued their research. For example, Wolfgang Metzger, Wertheimer’s assistant in Frankfurt, immediately took over Wertheimer’s position and became a member of the Nazi party. Edwin Rausch and Kurt Gottschaldt were also among those who remained in Germany although Edwin Rausch and Kurt Gottschaldt were marginalized to some extent during the Nazi period since they were not as partisan as Metzger (for a different perspective on Gottschaldt’s relations with the Nazis though see Mastroianni, 2006). Despite that, the main axis of science had already shifted to the United States and hence, to English. But most importantly, Germany had lost its prestige once its academia was seized under Nazi power, particularly so with World War II[10].

It seems that there were only very few Gestaltists in North American psychology, possibly only Mary Henle, who worked closely with both Kurt Koffka and Wolfgang Köhler. Mary Henle tried hard to make the conceptual side of the theory understood and to prevent it from being reduced to “a simple theory with a few practical and easy-to-understand visual demos”. It seems that Ellis's quick, abbreviated translations in his 1938 A Source Book of Gestalt Psychology volume did not serve this purpose too well. It is likely that Henle, therefore, decided to publish an anthology titled Documents of Gestalt Psychology in 1961 which included carefully selected articles and translations of the main founders and successors of the theory. For the English speaking scientific community, this anthology is probably one of the most important reference sources to understand the philosophical grounding of this theory. On a last note, there are probably very few scholars even within social psychology, who know that Solomon Asch, a second-generation North American immigrant of Jewish-Polish origin, was a Gestaltist who developed his experiments, including his famous conformity experiment, fully from within this theory.

Apart from this, there were those among North American scholars who were not Gestaltists but who were nonetheless strongly inspired by this theory. This applies especially to those working on visual perception and creative/productive thinking. James J. Gibson, famously known for his theory of "direct perception", is said to have been a regular attendee of Köhler's seminars (Mandler, 2007). Indeed, in his ecological perception approach, that is, perception in true three-dimensional space that contains motion, Gibson suggested perhaps the most radical of perception theories. Like the Gestaltists, he suggested that this space contained a lot of meaningful information hence providing direct knowledge about objects in their environment without the constant necessity for intermediary processing levels or gradual memory accumulations. Finally, with the advent of computers in the 1950s, the "cognitive revolution", which emerged as a movement against the ideology of behaviorism, seemingly opened space for Gestalt theory, making use of its many visual demonstrations, which occupy a superficially descriptive 1-2 pages in every single introduction textbook in psychology. Yet, in reality, it never challenged the simple, linear, associationistic viewpoint of behaviorism. One could even claim that all it added was an additional emphasis on top-down processing where the top-down component was an information-processing memory formation that again would come about in a more or less simplistic, linear, associationist manner. In this article, I will try to clarify why the so-called “cognitive revolution” is not at all an extension and elaboration of Gestalt theory.

2.4 An Article on Gestalt in 1938's Turkey: "Geştalt Nazariyesi"

Ziyaeddin Fahri (Fındıkoğlu), one of the most prominent sociologists of the time, published a Turkish translation of physician, psychologist and psychotherapist Pierre Janet’s “Gestalt theory” article[11]. Figure 2 shows a section of the preface by Fahri whose first sentence reads "Very few publications have been made in our country about the theory of form, one of the newest philosophical theories of our century." (Ülkü, 1938, 11(66), p. 486).

Figure 2 (Ülkü Journal, 1938)

This article is important as it shows that the Gestalt theory was introduced to this part of the world with relatively little delay. For example, Janet's book that Ziyaeddin Fahri mentions in the preface of this translation was Les Debuts de l'Intelligence (The Beginnings of Intelligence), published in 1932. Interestingly, both Janet and Fahri were clearly aware of the importance of this theory. It is worth mentioning that Gestalt theory has had an influence as far as Japan (Zanforlin, 2004), yet it has never been able to have a place in the mainstream of psychology, which was either busy doing minute experimentalist research or following the paths of psychoanalysis.

When we look at the current literature on Gestalt theory in Turkey, we see that almost all of the articles are either on Gestalt therapy (according to Henle, 1978, and Schultz, 1981, at least the North American version of Gestalt therapy seems to share nothing with the actual theory except for the word “Gestalt”[12] ) or education (for example, Kaygusuz, 2014; Koç & Bulut, 2014; Zeren, 2008). The latter is not surprising because this theory is able to offer a fully new perspective to learning and comprehension.

Gestalt theory is also pervasive in architecture and design journals and theses (e.g. Akkurt, 2019). Furthermore, many Turkish webpages have information about some typical visual examples of the theory (e.g.; ).

To summarize, also in Turkey, the incredibly rich and diverse writings of the Gestalt theorists are basically overlooked or utterly unknown. Instead, we see a direct reliance on how mainstream North American psychology has understood, or rather misunderstood Gestalt theory. The purpose of this article is to fill this gap. In the next section, I will present and discuss the main tenets of the theory, not so much –as often done- as a mere listing of “laws of organization” but rather within their theoretical groundedness.

3.Gestalt Theory

3.1 Main Tenets of the Theory

3.1.1 Gestaltqualitäten, Figure-Ground and Gestalt Principles:

Von Ehrenfels (1859-1932), Rubin (1886-1951), and Wertheimer

Wertheimer begins his seminal 1923 article Investigations on the Teaching of

Gestalt II. with:

“I stand at the window and see a house, trees, sky. And I could, for theoretical reasons, attempt to count and say: here are … 327 shades of lightness (and colors). (…) And let's say that in this strange count the house has 120, the trees have 90 and the sky has 117 (light tones), then I have at least (for sure) this kind of a grouping, this segregation, rather than 127 and 100 and 100, or 150 and 177. (…)

(And what a strange process, if one succeeds to do so, what surprise, if, after a long time of watching, (…) I discovered in the most unrealistic stance, that parts of a window’s dark frame form a Latin N with a straight branch of a tree)... " (p. 301)[13]

These beginning sentences allude to a couple of things. First of all, the attempt of counting the many shades of lightness (or color) are a direct reference to the then dominant structuralist approach of Wilhelm Wundt and his followers, who believed that through introspection one should identify the elements of larger wholes so as to reach an understanding of the wholes. This method was in accordance with the empiricistic[14] philosophy of that period: external stimuli are manipulated by the researcher in a controlled manner and via a certain training in introspection the participant is made to "count" all elementary sensory experiences. Apart from its methodological problems, the main problem here is the meaninglessness of the research question. In his 1879 Leipzig laboratory, Wundt aimed to understand human cognition in terms of sensations. The only thing that changed in the periods following Wundt was that, apart from sensations, other experiences such as thought and memory were assumed to be “decipherable” using a similar approach of isolating local, elementary mechanisms (using this approach, Ebbinghaus studied memory formation and decay, Georg Elias Müller mechanisms such as memory inhibition, and Oswald Külpe more complex experiences such as problem solving). In all these pursuits, the main presupposition was the understanding that by isolating the 'sense particles' contained in any given experience and measuring them one by one, one could grasp something about the resulting experience.

In his 1923 article, Wertheimer continues with an example from the music:

Or: I hear a melody (17 tones!) with its accompaniment (32 sounds!). I hear the melody and accompaniment, not simply “49” or at least certainly not or arbitrarily 20 plus 20.”  (p. 103)

With these examples, Wertheimer draws attention to the phenomenon of grouping and segregation that spontaneously "permeates" all kinds of sensational and perceptual experiences, something that has been completely missed in experimental psychology. Even though today no one challenges this basic fact, we still do not know its exact mechanisms. For example, when we look out of the window, what is most striking is not so much the sub-components of the “tree” image, but the experience of a tree as a whole separated from all other objects (other trees, windows and sky). And so is the experience of the window in the same holistic and segregated way or the mysterious emergence of the letter "N" if a tree branch enters into an appropriate configuration with the similar colored, vertical borders of the window frame. Or, for example, when we effortlessly hear out a melody in a complex, polyphonic musical form such as a symphony or a fugue. The next part of the article focuses on spatio-temporal configurations that generate these “sudden”, “spontaneous” dynamics of grouping and the basic principles[15] underlying them.

In 1890 Christian von Ehrenfels, a Viennese philosopher, introduced the term "Gestaltqualitäten", which also was the very title of his paper. Based on Ernst Mach's conceptualizations of "space and sound forms[16]" , von Ehrenfels pointed out that these shapes could be “transposed” without losing their holistic features. The examples were again from music, where transposition is a well-known phenomenon, Transposition in music refers to the playing a given melody from a different tone where its pitch contour (i.e., its melodic rises and falls) but more importantly its exact pitch intervals between the notes are fully retained, thus sounding identical except from a higher or lower tone. In other words, if someone who has always heard a melody as played from the note la (A), then hears it as being played from the note sol (G) or si (B), they will have no difficulty whatsoever to identify the melody as same. This is so despite the fact there will not be a single note that remained “in place” because when listening, what they instantly and even involuntarily will have identified is its Gestalt, not its individual, exact notes.

Another critical publication is the 1921[17] Visually Perceived Figures book by Danish psychologist Edgar Rubin. In his book, Rubin makes reference to von Ehrenfels' concept of "Gestaltqualitäten" and directly raises the "figure-ground" issue. Although the experimental studies he mentions in his book are rather simple in terms of setup and data analysis (when compared, for example, to Wertheimer's outstanding 1912 research), the ingenuity the visual materials he created left a strong mark in psychology. He attentively notices, for instance, that in his Figure 3, people cannot help but see white shapes on a black background (and vice versa if colors are reversed). Even if we force ourselves to see the blackness as a single shape (say, with “holes” inside) we are not able to do so steadily, maybe only locally in the attended spots but then even only for a very brief time. In other words, what is important here is not just the instant segregation of figure and ground but also its stability.

Figure 3 (Rubin, 1921)

In Figure 4, on the other hand, we have a different case. Here, people may differ in terms of which of the two main shapes (the black one or the white one) they perceive as figure and which one as ground. Yet, typically everyone will be able to voluntarily switch back and forth between the two. What is important in this stimulus is that when the black part is perceived as figure, a claw-like shape appears, whereas when the white part is perceived as a figure, a rounded shape appears, perhaps reminiscent of two lips and a tongue. Hence, there is a clear contrast between the roundness of one and the sharpness of the other. In contrast, in Figure 5, this asymmetry is controlled for by having two potential figures which are more or less the same.

Figure 4 (Rubin, 1921)

Figure 5 (Rubin, 1921)

Another of his stimuli is the famous vase stimulus (Figure 6), which we know from hundreds of psychology textbooks, with rarely any reference to Rubin. Here, Rubin points out that participants would simply shrug their shoulders and say that they obviously saw a vase only to be utterly startled when their attention was drawn to the two faces looking at each other.

Figure 6 (Rubin, 1921)

3.1.2 Gestalt Principles of Grouping

In his seminal 1923 paper, Wertheimer closely scrutinizes these grouping/segregation dynamics and then goes on to propose, step by step, principles that seem to guide these. Wertheimer formulates the question as follows: if there are five consecutive stimuli, say “abcde”, what makes such as series be segmented as (abc)(de) as opposed to, say, (ab)(cde)? If it is grouped in one way or the other, how stable is that state of grouping, and how amenable is it to change? After posing these most basic questions, Wertheimer proceeds with the first principle, the principle of proximity.

3.1.2.1 Principle of Proximity

This principle states that those elements that are close to each other are prone to be grouped, and as such, segregated from those that are more distant from it. In his article, Wertheimer shows a series of dots which are either closer to or further away from each other (Figure 7). Such a stimulus (and let us consecutively label each dot with the letters of the alphabet, just as Wertheimer did) is instantaneously grouped as (ab) (cd) (ef) etc. When asked to try to group it as (a) (bc) (de) etc., participants either reported that they cannot, or if they can, only for a brief glimpse of a time. Meanwhile, Wertheimer reminds the reader of all the critical things to control in such experiments, such as the specific ordering of the stimuli, the speed at which they appear on the screen, problems such as the appearance of unwanted contour spots due to the projector etc[18].

Figure 7 (Wertheimer, 1923)

In the next figure, when the points are arranged diagonally (Figure 8a), he is not contented with simply stating that they are perceived as a series of 2-dot diagonals. Instead, he goes on to see whether participants could manage to group the stimulus as shown in Figure 8b. He once again noted that they either could not do this kind of grouping at all or just momentarily, only to turn back to the other grouping.

Figure 8a (Wertheimer, 1923)

Figure 8b (Wertheimer, 1923)

In Figure 9a, we present a snapshot from the original 1923 article. Here the reader will notice that they will almost inadvertently add the two dots that appear in the descriptive text below the image to the two diagonal dots of the figure so as to form an S-like group of 4 dots (Figure 9b).

Figure 9a (Wertheimer, 1923)

Figure 9b (Wertheimer, 1923; red circle added by author)

Wertheimer also mentions the impact of the number of points used (e.g., two points per group versus four or five points per group). He remarks that the higher the number of elements per grouping the less flexible the perceptual organization of the entire layout. On the other hand, while the number of elements per grouping affects the flexibility for alternative ways of grouping it does not affect the ease with which their “default” grouping, say their proximity-based grouping occurs. In other words, “default” grouping seems to occur with comparable ease, regardless of number of elements per group. Wertheimer then gives a rhythmic example of the same phenomenon. When presenting a sequence of knocks of alternating short and long between-knock durations (       ) a similar instantaneous grouping emerges. An auditory counterpart of the Figure 8a, in turn, would be grouped dual knocks with one being lower pitch and the other, higher pitch.

On page 308 of his article, he emphasizes that as much as grouping by proximity appears to be the most self-evident thing to be, it is not. Instead, what needs to be done is to carefully understand what exact temporal or spatial distances lead to segregation, whether there are absolute spatio-temporal distance values that predict segregation or whether the mechanisms are driven by proportional values. He then goes on to ask whether we might ever find a mathematical function to model proximity-based grouping or not. All these questions remain unanswered even today, which in a way is shocking, given that almost 100 years have passed. In their 1990 article, Rock and Palmer already mention that at the retinal level, there is not a single finding to explain, for instance, simple figure-ground grouping. The retina processes any given stimulation solely on the basis of wavelength and luminance, hence it processes color shades and shades of brightness and contrast but there is no mechanism to directly pinpoint/reflect distance, be it of the spatial or temporal kind.

In other words, what appears to be so naturally grouped in the outside world has no direct counterpart at the retinal level. This also applies to the cochlea, which at any given time responds only to the frequency and loudness of an incoming sound[19].

Moreover, we still do not know how exactly an auditory rhythmic sequence is processed by the brain since it is yet unknown how exactly the proportional relationships of temporal durations (which are also critical in the perception of spatial properties) are encoded in the brain. Despite decades of research in neuroscience, we still do not know how the brain “makes sense” of time (cf. Buhusi, 2020).

3.1.2.2 Principle of Sameness[20]

This principles states that in a layout of diverse elements, those that are same will be grouped and as such segregated from the rest. Here, too, Wertheimer not only looks at what is grouped but also at whether participants can make themselves perceive alternative groupings, and if so, how stable those are (see Figures 10a-d). Once again, the grouping that occurs instantly and effortlessly is the one where all same elements are grouped with each other. Alternative groupings such as trying to perceive Figure 10c as ten horizontal lines of an alternating series of unfilled and filled circles are not stable and remain restricted to local areas where attention is directed to (i.e., while we perceive six vertical series of unfilled circles alternating with six vertical series of filled circles as an emergent whole, this is not the case for their horizontal nonhomogeneous versions).

a)

b)

c)

d)

Figures 10a-d (Wertheimer, 1923)

When discussing the principle of sameness, Wertheimer once again provides comparable examples from the auditory domain (see  Figure 11a-d).

Figures 11ab (Wertheimer, 1923; “!” stands for stressed sounds, “” for unstressed sounds, hence yielding a binary grouping of an alternating series of stressed and unstressed groups in 11a and a ternary version of it in 11b)

Figure 11 cd (after Wertheimer 1923 examples)

3.1.2.3 Principles of Proximity and Sameness in Juxtaposition...

In the subsequent section of his paper, Wertheimer focuses on how the two basic principles interact with each other. He states that the principle of sameness seems more dominant when the two were put in juxtaposition (Figures 12b, 13b). He also remarks that even if observers seem more prone to group the unfilled and filled (smaller) circles together despite their spatial separation, the stimuli in Figures 12b and 13b seem to appear “erratic” to the person. Likewise, specific parameters (amount of distance, degree of dissimilarity between the segregated stimuli) also influence how the stimuli will be grouped and how stable or instable that grouping is. He points to a similar tendency when using auditory instead of visual stimuli (Figures 14a-b, 15a-b).

Figure 14 ab (Wertheimer, 1923; “!” stands for stressed sounds, “” for unstressed sounds; red circles added to indicate dominant groupings, arrows mark extended time intervals)

Figures 15ab (After Wertheimer 1923; the unfilled notes are played twice as long as the filled ones, red circles are added by the author to indicate groupings, and arrows to indicate elongated time interval)

3.1.2.4 Principle of Common Fate

This principle states that all those elements that move simultaneously are grouped together. Here, Wertheimer once again analyzes a principle in combination with another principle. As an example he presents Figure 16 to point out that the perception of a grouped unit that has structure-congruent motion (e.g., dots d-e-f moving up simultaneously) is phenomenally experienced as smoother compared to the case when a structure-incongruent motion occurs (e.g., dots c-d-e moving up simultaneously). He further remarks that the effect does not change depending on whether the simultaneously grouped movement is vertical, horizontal, diagonal, or radial.

Figure 16 (Wertheimer, 1923)

Wertheimer himself does not go into further detail but let me refer the reader to Albert S. Bregman’s (1994) masterwork on grouping principles in the auditory domain.

3.1.2.5 Principle of Good Continuity

This principle states that elements that form some sort of successive continuity (one might say a “directional flow”) will be grouped as one unit. Here, Wertheimer looks at different versions of continuity to see how resistant a given continuity is to different types of constellations (see Figure 17). In Abb.8 of Figure 17, obviously, what is grouped is AC as separate from B, that is, (AC)B, not (AB)C or A(BC). Abb.9 serves to show that such a grouping is independent of whether there is a continuous line or a sequence of (equidistant) dots. He then points to the typical grouping of (AD)(BC) in Abb.10 instead of (AB) and (CD) or (AC) and (BD) and poses the question as to whether the (AD)(BC) choice is due to the former having a curvilinear characteristic and the latter a linear characteristic. Hence he tests Abb.11 and 12 to prove otherwise. He also ponders on the possibility that the angle with which the separated unit (e.g., B in Abb.8 and 9) attaches to the other unit (e.g. AC in Abb.8 and 9) might change the grouping, but it does not. He further looks into a series of other factors, such as the degree of curvature of the arcs and experiments with those as well.

Hence, we see a very meticulous construction of a variety of stimuli with the purpose of understanding the essential factors that bring about a given grouping. In other words, even in this very article where Wertheimer introduces the various principles of grouping for the first time, we are not talking about a simple demonstration with the most obvious of stimuli, just the contrary. We see a very analytical and very perceptive mind who pays as much attention to the different aspects of a given stimulus layout as to the phenomenal experience of perceiving them. In conclusion, he remarks that what seems to determine grouping is a certain holistic directionality, an inner belongingness, a simplicity that emerges through grouping via good continuation.

Figure 17 (Wertheimer, 1923)

3.1.2.6 Principle of Closure

A contour surrounding a regular or irregular space (e.g. Figure 18) allows that section to be perceived as a single shape (Figure 18). Wertheimer again examines many different aspects to better understand why such grouping and segregation occurs. He points, for instance, to the fact that a closed area must not necessarily lead to a segregation (cf. Abb.31 and Abb.33 in Figure 19). The constellations in Figure 19 are intriguing in that they point to the importance of the overall constellation. In that sense, it is rather disappointing that the principle of closure has been reduced to Koffka’s more simple example (which he took from Köhler, 1920) of a series of inward/outward brackets (or a circle that is almost closed except for a sufficiently tiny gap) where the focus is simply on the gap that is either mentally “filled in” or not, depending on how large the gap is[21].

Figure 18 (Wertheimer, 1923)

Figure 19 (Wertheimer, 1923)

Figure 20 (Koffka, 1936)

3.1.2.7 Principle of Good Continuity and the Principle of Closure in Interaction

Figure 21 presents two cases where the principle of good continuity and the principle of closure are pitted against each other. The question here is whether the observer would see the rectangular lines with a curvilinear line crossing through them or whether s/he would perceive three separate, asymmetrical, curvi-rectangular closed areas. Yet, the interaction of these two principles is likely to be much more complex because we see, for example, an opposite trend in Figure 22, where once the areas are closed as in 22b, it is no longer the running through lines of (AB) and (CD) that are perceived but the closed areas. Koffka takes this example from yet another work by Köhler. Of course, the way in which the two closed spaces are connected to each other is also quite different from that in Figure 21.

Figure 21 (Wertheimer, 1923)

Figure 22ab (Koffka, 1936)

3.1.2.8 Degrees of Prägnanz (Singularity)

Wertheimer uses words containing the root "prägnant" in many places in his 1923 article[22], but it is in its fifth part where he attempts to better clarify what he means by this term. In general, he describes a percept of utmost Prägnanz as the one that is most stable. He oftentimes uses the term synonymously with “good form”, “simple form”, yet, both of these again share the common quality of perceptual stability. At one point, he remarks that while in a tachistoscopic presentation the defect of a slightly imperfect circle mostly goes unnoticed, this is not so once presented at normal looking times. In the latter, the observer immediately “feels” that something is different and quickly thereafter detects it. He explains this as being due to a sensing or “feeling” of an instability of what was before stable.

What is important here is "stability" rather than "simplicity" (see Luccio, 2019). According to Luccio (1998), “Prägnanz” has assumed two different meanings within Gestalt theory. One of its meanings is "Ausgezeichnetheit" ("singularity"), which serves to identify which parts of a constellation will assume figural rather than ground qualities. The other meaning is the very process of the emergence of a figural shape from among a complex, say camouflaged layout. What is imperative in Gestalt theory is that this dynamic process of ‘emergence’ happens not arbitrarily but according to a certain "lawfulness". In fact, Köhler does not prefer to use the term "Prägnanz" and instead uses the term "self-distribution[23]" (as borrowed from physics) probably to better emphasize its dynamical (and “lawful”) quality.

Koffka, on the other hand, simply lists it as yet another principle, hence puts it on a par with all the other principles of perceptual organization. Koffka's striking example for Prägnanz can be seen in Figures 23a, b, and c. When Fig. 23a is presented to the participants first, they mostly perceive it as a two-dimensional shape, that is, the more stable, hence immediate image is the two-dimensional image. Even though, particularly after seeing Fig. 23c, 23a could be perceived also three-dimensionally, its two-dimensional form appears to be more stable. The exact opposite holds for Fig. 23c, which more immediately imposes a three- than a two-dimensional percept. Koffka ingeniously explains this on the basis of Prägnanz. In Fig. 23a, the two-dimensional “kite” like form provides the simplest, most symmetrical and smoothest (due to continuous lines) solution, whereas its three-dimensional percept is much more irregular, discontinuous (many corners, drastic changes in direction due to the edges of the cube etc.). In Fig. 23c, on the other hand, the much more prevalent cube shape is the simpler, more symmetrical form compared to its two-dimensional version, partly because the simple vertical line in Fig.23a is now broken in Fig. 23c. Fig. 23b, on the other hand, is a kind of transition figure which is amenable to both dimensionalities. Yet, all figures have a dual structure which allows for a shifting back and forth, although one structure will be more dominant than its alternative, due to its stronger Prägnanz.

Figure 23 (Koffka, 1936)

3.1.3 The Role of Experience and Learning

In his 1923 article, Wertheimer also provides interesting examples on the role of experience and earlier learning. He points to Figure 24a and states that --in line with the "frequency" proposition of empiricist theory's which claims that whatever is associatively learned more frequently will be sensed/perceived/remembered faster-- an educated person would be expected to see at first a handwritten version of the letter “W” on top of an “M” (its segmented version shown in Figure 24b[24]). However, this was not the case. Instead, most participants reported to see two large brackets facing outwards and a decorated diamond-like shape in the middle (its segmented version shown in Figure 24c). Wertheimer states that the frequency of encountering the discrete pieces of Figure 24c is likely to be much lower compared to the handwritten letters W and M. Hence, an empiricistic perspective seems to fall short of explaining the more prevalent segmentation of Figure 24a into two brackets and a middle part as in Figure 24c because it overlooks the role of structure per se. When discussing this example, Wertheimer likewise emphasizes that often, rather than the frequency of past encounter it is the perceptual stability of the possible sub-forms which drives the grouping process. In this specific case, the segmentation of Figure 24a would be driven by the higher degree of Prägnanz of each of the two sections marked in Figure 24c (say, the simplicity hence perceptual stability caused by the unbroken continuous flow of the curvilinear lines of each bracket and the both vertical and horizontal symmetry of the middle ornament) compared to those in Figure 24b.

Figures 24abc (Wertheimer, 1923; shapes b and c were created by the author to show more clearly the segmentation expected from the empiricist perspective and the segmentation actually experienced, respectively)

However, Wertheimer does not underestimate the effect of experience, as the brackets and diamond shape is also tied to experience. His objection is towards the attempt to explain everything only through a mechanism of linear associationism and the claim that this simple and singular mechanism is entirely sufficient to explain all there is.

In his next example (Figure 25), for instance, he points to the role of enculturation. He remarks that those coming from a Latin alphabet written culture will more likely see an ornamented letter “V” at first glance, whereas those trained in Greek letters will instead see a “gamma” (γ) or possibly a “sigma” (σ) if prone to focus on the lower part[25] of the shape. Hence, in strong contrast to common belief, Gestalt theory never denied the role of experience or culture, just the opposite[26].

Figure 25 (Wertheimer, 1923)

In fact, Wertheimer's article Musik der Wedda from 1909, which he wrote while working with the philosopher, psychologist and musicologist Carl Stumpf, is an analysis of cultural factors on musical perception. Likewise his inspiring 1912 article Über das Denken der Naturvölker, Zahlen und Zahlgebilde (On the Thinking of Natural People[27] s, Numbers and Number Structures) examines how traditional cultures handle and perceive numbers and countable entities differently compared to Western modern cultures.

3.1.4 Overall Conclusions from Wertheimer’s Seminal 1923 Article

Throughout his 1923 article, published in their newly founded Psychologische Forschung, Wertheimer emphasizes two things. One is that the variety in ways of perceiving cannot be easily explained with an associationistic learning model which fully ignores the importance of structure, grouping, and Prägnanz dynamics. Secondly, and relatedly, he points out that it is not the case that our perception is driven by equally weighted, random, repetitive processes of learning. Instead, there are internal dynamics intrinsic to structure (with attractor points, to use today’s terminology) which affect the way we perceive things. And it is these mechanisms and dynamics (the gestalt quality of “transposibility”, figure-ground segregation, grouping, and Prägnanz dynamics), most of which work “von unten nach oben (from bottom to up)”, that need further explaining. As mentioned earlier, after 100 years we are still unable to understand these fundamental perceptual phenomena that no one in the scientific world challenges yet equally no one can explain so far. In a later section (2.2 Conceptual Tenets of Gestalt Theory), I will focus on the conceptual, philosophical groundings of this theory to show that Gestalt theory is something far beyond “a simple set of rules about perception”.

3.1.5 Koffka's Discussions on Figure-Ground Relationships and Gottschaldt's

"Embedded Shapes"

Koffka's 1936 Principles of Gestalt textbook is intriguing in many regards and particularly because of the many striking questions he poses whose answers we do not know even today, and worse, which we are almost made to forget to ask[28]. In the introduction of his book he remarks that he will present many findings, but that he will examine and interpret them not disjointedly as "curious paraphernalia" lined up ‘like wax statues in the Madame Tussauds Museum’, but on the contrary, with respect to their meanings within a systems perspective. The book, with its fifteen chapters and more than seven hundred pages, in itself seriously challenges the widespread judgment in mainstream psychology that Gestalt theory made important propositions but could not go much beyond perception. Unlike today's psychology textbooks, it is a book that proceeds question by question rather than “drowning” the reader in a plentitude of disconnected findings (it should not be forgotten that there were many publications and experimental findings even at that time). A lot of those questions are the kind of questions that any person even a child would ask and like to know. In his 1959 speech, Köhler beautifully states that “in human psychology, we simply must use terms which --if I may use this expressions-- "sound human".” (p. 10). Mainstream psychology textbooks, on the other hand, and it gets worse with every next decade, seem to make us obliterate our own understanding of ourselves and replace it with a cold, disconnected, linear, associationistic paradigm that falls short of finding decent, meaningful explanations on more challenging though very basic topics such as form perception, creative thinking and the like. For this reason, I recommend this book to anyone who wants a fresh and completely different perspective on cognition, learning, motivation, memory, social dynamics and personality. Moreoever, a bit unlike Wertheimer's and Köhler's books, Koffka’s is extremely "reader-friendly".

In the visual perception chapters of his book, Koffka makes a couple of interesting remarks about figure-ground perception (which is missing in Wertheimer's 1923 work since its focus is on grouping principles only; and Rubin’s 1915 work, in turn, lacked a theoretical grounding to explain his discovery). In Figure 26, we can see that Koffka generated a figure-ground stimulus similar to Rubin’s stimuli. He then asks why in that particular configuration people predominantly see the white[29] part as figure, and the black part as ground. He hypothesizes that of the two, it is the one with a higher degree of complexity (hence a lower degree of homogeneity) that will strike out as the figure. Of course, the curious reader may ask if the white shape might have stood out as a figure not because of its relatively higher complexity but simply because it has a vertical appearance just as objects happen to stand on surfaces in our organized surroundings. However, the black background on top may as well be perceived not as some shapeless upside-down “thing”, but as, for example, a slightly lifted, upright, old-style handset phone. Koffka's proposition is different in that he does not ponder on “object knowledge” but on general factors of complexity vs. simplicity, more vs. less articulation, and heterogeneity vs. homogeneity: the more complex, articulated and heterogeneous units are the ones that tend to be seen as figures, while the simpler, less detailed and more homogeneous regions tend to be seen as ground.

Figure 26 (Koffka, 1936)

In another part, he discusses figure-ground dynamics again in black-and-white figures, this time keeping complexity, articulation, and heterogeneity constant while manipulating the relative sizes of the areas (Figure 27 a and b). When the black and white parts have comparable surface areas (Figure 27a), participants are about equally likely to perceive a white or a black cross as figure. When, in turn, say the white parts are reduced (Figure 27b) these smaller areas are more likely to be seen as figure, and the larger (black) ones as ground.

Figure 27ab (Koffka, 1936)

In a later section, Koffka inspects the role of articulation in figure-ground perception. He mentions a classroom experiment where students were shown Figures 28 a, b, and c. If it was the degree of articulation (“density of energy”) that drove figure-ground segregation (the more articulated, the more figural), then in all three cases it should be the patterned slices that will be seen as a figural whole against the white ones pushed back as background. Yet, he notes that for Figure 28a, the tendency was more in the opposite direction. He remarks that the outer radial black line per striped slice connects with the circle as a whole, which might have caused those areas to be rather perceived as ground than figure. Hence, he developed Figures 28b and 28c to control for this artefact. He finds that even in Figure 28b, the white cross and patterned cross alternated back and forth with none of the two predominating as figure. However, in Figure 28c, it was the more complex, more articulate cubes which stuck out as figure against an octagonal (white) background. Even though one could force oneself to see the white parts as figure, they would be less stable compared to the patterned shapes. He furthermore points out that this should be examined more systematically with a rigorous series of experiments[30]. As a side note, what Koffka is focusing on here also coincides with the "common region" that Palmer proposed as a new grouping principle in his 1992 article. However, I doubt that this is a new grouping if one remembers that Wertheimer’s principle of closure was not simply about the well-known “fill-in the blanks” phenomenon but about the fact that whatever is enclosed is prone to be grouped.

Figure 28abc (Koffka, 1936)

Koffka presents symmetry as yet another possible factor to cause figural perception. Unlike Figure 26, in Figures 29a and 29b only one of the white and black areas is symmetrical. Koffka reports a study by one of Rubin’s students which showed that persons viewing these plates mostly saw the symmetrical black stripes in 29a, and the symmetrical white stripes in 29b as figure. Although it is possible to see the asymmetrical stripes as figure, that percept is much less stable than the symmetrical ones.

Figure 29ab (Koffka, 1936)

I want to finish this section with the interesting work on embedded figures by Gestaltist Kurt Gottschaldt. This again seems related to the figure-ground conceptualization, where this time figures are made invisible by “burying” them into the ground[31]. One of Gottschaldt’s embedded figures is presented in Figure 30b where the observer has to find Figure 30a within its complex --and more confusingly—seemingly two-dimensional layout. Today, we see a wide use of such test batteries (e.g., "Embedded Figures Test", Witkin et al., 1971). It has been shown, for example, that children with autism are better at finding embedded shapes than neurotypical children (Jolliffe & Baron-Cohen, 1994; Van der Hallen et al., 2018).

Figure 30ab (Gottschaldt, 1926)

3.2 Conceptual Propositions of Gestalt Theory

In his highly philosophical and arduous 1938 book The Place of Values in a World of Facts, Köhler mentions that psychology as a science has constantly been producing facts while evading concepts of "value" and "meaning". Both Köhler (1938) and Koffka (1936) point to the dead-end generated by a persistent tendency to produce “interesting findings” that remain disconnected, something that perhaps is even more relevant today. In his 1938 book, Köhler remarks: “When asked to choose between writing badly about the greatest questions and well about more modest topics, we prefer the second alternative.” (pp. 6-7). Again, both Köhler and Koffka criticize that the findings accumulating within the science of psychology were either evaluated from a most mechanistic and meaningless perspective ('cognition arises from the senses that are interconnected randomly via successive association’) which fully missed on its most crucial component (meaning and value), or were explained from a vitalistic[32] point of view that would rely on innateness to explain phenomena while inventing new terms of explanation. They see the former as reductionism, and the second as simplicity.

In his book, Köhler tells the interesting story of an exhibition in a natural science museum. In one of the shelves, all the substances that make up humans such as oxygen, hydrogen and nitrogen are represented by cubes of different sizes in proportion to their amounts in the human body. Each cube has a tag that marks its monetary value based on the market prices of the time. When all of those were added one by one, the total monetary value of a human turned out to be approximately 63 dollars. Starting with this example, he emphasizes in detail why the sum of the parts cannot accurately express the whole they form. Unlike von Ehrenfels, who defines the whole as something more than the sum of the parts, Köhler and the other founders prefer to use the term different (as qualitatively different) instead of more (possibly, in order to evade any presupposition of a unilineal dimension of quantity). And as such, the goal of psychology from a Gestalt point of view is to focus on the qualitative aspects of the emerging whole, rather than just looking at its parts just because the latter have the handy advantage of being easily manipulated and tested in isolation.

3.2.1 Parts and the Whole: Microscopic and Macroscopic View

The most important conceptual proposition of Gestalt theorists is the emphasis on the direct effect of a whole onto the functions, values and meanings of its parts. With this emphasis, they revolt against the view that we can examine the parts, determine their consecutive interactions with each other on a local scale and this way understand the whole, which was the dominant view then and remained as such till today. Every point presented in 2.1 Main Tenets of the Theory should be evaluated from this very perspective. If we want to understand the perception of a stimulus from a Gestalt perspective, we must first determine its holistic configuration and then look at the "field" effects of the whole onto the parts within that specific configuration.

The term "field" is a concept that emerged from the physics of that period and was put forward by the Gestaltists to mean something very close to its meaning in physics. Köhler was a philosophy, natural sciences and psychology student who attended various lectures by 1918 Nobel Prize-winning theoretical physicist Max Planck. This background is strongly reflected in both his 1920 book Die Physischen Gestalten in Ruhe und im Stationären Zustand: Eine Naturphilosophische Untersuchung [Physical Gestalten at Rest and in a Stationary State: A Naturalphilosophical Analysis] and his 1938 The Place of Value in a World of Facts book.

A simple example of a "field effect" is the following in Koffka's book:

Figure 31 (Koffka, 1936)

Due to the impact of the whole on its parts, the middle line appears to be of different lengths in two different "environments". It is clear that this cannot easily be explained by a simple, unidirectional, mechanistic learning model. Likewise, the famous Müller-Lyer illusion (Figure 32) has been mentioned even by Rubin way back in 1915. In this illusion it is the entire field that affects how the middle line of the arrowed stimulus is perceived in terms of length. When the overall field is expanded due to outward facing arrows, the middle line is perceived as longer compared to when the arrows are facing inward. Rubin then goes on to report an experiment which wiped out the illusion simply by presenting the two stimuli of the Müller-Lyer illusion as flashing back and forth such that all but the middle line would move. Therefore, according to Gestalt theory, this illusion occurs as a result of the area effect that the entire stimulus imposes on the horizontal line[33].

Figure 32 Müller Lyer Illusion

Another interesting example is an experiment by Révész on the Jastrow illusion with chickens, as quoted by Koffka in his 1936 book. In this experiment, chickens were trained to eat their food always from the smaller one of two containers. When presented the containers in a Jastrow illusion configuration (Figure 33a), chickens always ate from container B. As such, this study seriously challenged all earlier experience-based learning explanations since the chickens never had seen this layout before and were obviously very unlikely to remember, say, their experiences with curved railroad tracks[34]! Instead, it is the specific positions of the parts within the spatial whole, that exerts an intense effect on the perception of each.

Figure 33a (Wiki Commons) Figure 33b

It is worth pointing out a common misunderstanding regarding Gestalt theory's standing with respect to “holism”, as critically raised by Mary Henle (1992). Henle rightly claims that contrary to popular belief, the theory cannot and should not be called "holistic" because a full-fledged holism proposes complete “melting” of the parts within a whole, whereas Gestalt theory never proposes such a complete “perishing” of parts. In any Gestalt theory based experiment, the parts are carefully manipulated so as to understand how and for what reason their specific constellation as a whole affects the way they will be perceived (say, as part of a figure or ground, as salient or embedded, as a certain figure x or a certain figure y, etc.). In Chapter V (The Nature of the Physical World) of his 1938 book, Köhler remarks that “objects of scientific investigation themselves generally have both macroscopic and microscopic properties…. neither of which are less "real" ” (pp. 198-199). Metaphysically speaking, Köhler criticizes the experimental psychology of his times (and not much has changed in this regard) for treating only the microscopic as real, while seeing the larger system generated by the specific constellation of the parts as "imaginary”, because to them, it is nothing but the sum of the parts, not an entity in itself. He gives an example from music and emphasizes that in any melody, a given note has not much meaning by itself nor in a simple two-note succession, but that the meaning per note derives directly from the system to which the whole melody belongs (for example, the key or scale in Western tonality or makam in Eastern tonality). He fiercely criticizes Behaviorists, their contemporaries, who would later even try to explain language as a simple chaining of parts via successive pairing. For that matter, it was this major failure to explain syntax that contributed significantly to the fall of behaviorism, paving the way to a "cognitive revolution" in the late 1950s with prominent scholars such as the linguist and theorist Noam Chomsky. Language, on this occasion, also serves as a very good example of the proposition that the whole determines the meaning and function of its parts[35].

3.2.2 Context, Field, and Structure Against Randomness

In his 1938 book, Köhler emphasizes that the experienced (i.e., "phenomenal") world “is neither an indifferent mosaic nor an indifferent continuum” (pp. 84-85). In other words, according to Gestalt theory and in stark contrast to empiricistic views, things do not get associated with each other in any possible random constellation, simply by being repeatedly presented and paired. Instead, there are intrinsic, dynamic, and contextualized dependencies due to field structure, that is, due to the elements of a whole being mostly parts belonging to a whole rather than random pieces within it (cf. Wertheimer, 1934). And here, Köhler argues that since the external world has structure as we know from physics, one might expect an experiential world to be "mirroring" that structure. This leads us to the concept of "isomorphism" that Köhler put forward in an attempt to take this idea one step further.

3.2.3 Isomorphism

Isomorphism is a term that Köhler proposes to refer to a certain correspondence between the physical world and the perceptual world. The idea is that things in the physical world must have a physiological reflection compatible with their external reality. He says, for example, that there must be a correspondence in terms of the figure-ground relations and figural constellation between the elephant in physical space and its phenomenally perceived version. The proposed correspondence is not a literal one but an abstract, structural one, that is, a correspondence of relational structure, i.e., of configuration and their internal and context-situated dynamics. He talks about latest findings in brain physiology of the time and makes reference to Sherrington who remarks that "All parts of the nervous system are connected together and no part of it is probably ever capable of reaction without affecting and being affected by various other parts, and it is a system certainly never absolutely at rest." (cited in Allen and Schwartz, 1940). He likewise cites Adrian et al.’s 1928 study as it reports that ganglionic cell groups show clearly interrelated activity and that the same is true for the retina and cortical layers. In fact, he makes an interesting analogy that the nervous system of the brain will also have a figure-ground dynamic; he states that while on the one hand there are clear neural activations that show structure and shape, on the other hand, the whole system can be a 'ground', --in today's unfortunate terminology-- 'noise[36]'. Then, he is as foresighted as to emphasize that the studies of physiologists who examine the brain waves of a living being in vegetative state should be closely followed.

3.2.4 Geographical Environment, Phenomenal[37] Environment, and Meaning and Values of Living Beings

To explain what he means by phenomenal versus geographical environment, Koffka gives the following example. During a terrible snow storm, a man on horseback arrives at an inn. Seeing the man, the innkeeper is surprised and asks where he came from. The man points straight backwards, whereupon the innkeeper asks in astonishment if he knew that he came over the giant lake Constance. Koffka states that there are two realities here. One is the physical reality, or in his own terms, the "geographical environment", i.e., the frozen lake. The other is the person’s subjective, phenomenal reality, which is nothing more than a firm, open plane as the person is unaware that he was coming over a frozen lake. This is something completely overlooked by the behaviorist school, but even by the cognitivist school since it, too, is predominantly interested in a participant’s behavior in an experiment, and never as invested in their subjective experiences and beliefs in itself[38].

Another example Koffka gives is from Köhler's 1914-1920 studies in Tenerife on the problem-solving abilities of chimpanzees. In one setup, chimpanzees see a banana hanging from the ceiling at unreachable height with wooden boxes all around the floor. While one of the chimpanzees sits on one of them and looks at the banana, another one takes one of the boxes and moves it to where the banana is hanging, in order to step on it and reach out for the banana. Koffka remarks that the geographic environment is the same for both chimpanzees, but for one chimpanzee, those boxes are nothing more but things to sit on, while for the other, they are a tool to lift them up[39].

The phenomenal environment is, on the one hand, related to the present physical space-time, and on the other hand, to the organism’s history with that environment, as well as their present attention and motivation. In other words, a Gestalt perspective requires an understanding of the field effects of physical space-time together with their phenomenal “reverberations”, as well as an understanding of the motivation and history of the experiencing organism[40].

This aspect of the theory is clearly visible in their studies which became an –albeit unacknowledged-- part of the then emerging field of social psychology in the United States. Probably no one in mainstream psychology knows that Solomon Asch was a Gestaltist whose 1951 experiment on conformity was fully grounded in that theory. In this famous experiment, participants were presented with a very obvious perceptual layout where they had to tell which of the three lines on the right was equally long as the standard presented on the left (Figure 34). The critical manipulation was that a social field was created by asking the participant to say out their answer as the last person of a group of confederates who, in some trials, would give correct answers and in some, unanimously, the same wrong answer. Compared to a control condition where the participant was allowed to privately write down their answer, an interesting effect of conforming occurred where about a third of the participants conformed to group pressure more than once. Asch interpreted this as the influence of group environment as a field onto a person’s behavior to a point where they would be willing to deny their own phenomenal reality. If one reads the original article, one sees that unlike the way this research is “packaged” in mainstream psychology textbooks, Asch places a lot of value on those who never conform (despite their feelings of uneasiness, insecurity and isolation within the group), that is, instead of constantly highlighting the empty part of the glass, he chooses to highlight the full part. This certainly does not come as a surprise given his Polish-Jewish background. After all, even in Nazi Germany there were those –albeit fewer ones-- who resisted… In today’s social psychology, which in its mainstream narrative constantly prefers to showcase those who conform/obey/yield/give in, one wishes that the emphases will switch back to those who do not. This change, however, will require an inclusion of participants’ phenomenal environment, so as to reach an understanding that cannot be attained through a simplistic quantification of whether they conformed or not.

Figure 34 (Source: Wiki Commons)

Wertheimer's study, in which he examines the ways of thinking of natural peoples, is also a good example on this subject of the phenomenal (Wertheimer, 1912b). When he asked members of the community (pointing to the image of two trees on one side of the physical plane and another tree at a distance from those trees) how many trees they saw in total, they said "two trees and one tree", not “three trees”. On the other hand, when he pointed to an area where three trees were side by side, the answer was “three”. In another interesting example, two horses were counted as "two" and likewise two people as "two". But when a man and a horse were shown together, they would say "a rider" to point to a singularity rather than a segregated duality. Most importantly, Wertheimer does not treat this difference as a wrong way of thinking, on the contrary, he draws attention to how differently the same physical environment can be experienced depending on one’s acculturations. According to Wertheimer, none of those answers could be treated as “random” or "meaningless", on the contrary, each had its own logic and embedded meaning.

4.In Place of a Conclusion: What Gestalt Theory Is and What It is Not…

The most critical emphasis of Gestalt theory is that any stimulus environment should be considered in its entirety, since the entire configuration determines the function and meaning of its parts. This is the very reason why Köhler (1921), for instance, created different configurations of sticks spread around the floor to observe chimpanzees’ abilities to single them out as tools to reach out for a banana. When the sticks were rendered "invisible" in an unorganized ground configuration, it became impossible for them to notice them as possible tools, that is, to detect them as singular figures against ground. This again shows how important the particular type of configuration of the external stimulus environment is. Therefore, in any research, the first step should be to focus on the entire configurations instead of its parts comfortably manipulated in sterile isolation to record their “solitary” effects. We can find this ecological view, in James J. Gibson's theory of "direct perception[41]" . Gibson's biggest objection to the studies of perception of his time was that they missed out on that vast richness in physical space. Like the Gestaltists, Gibson almost in anger expresses that those who study science with a 'microscopic view' as Köhler put it, ignore how much meaningful information the ecological environment actually provides for living beings moving within that space (Gibson, 1992)[42].

The second critical phase of Gestalt research is to examine the phenomenal environment, to use Koffka's words. For this, not only behavioral data (e.g., reaction times, accuracy, number of words remembered, points of segmentation in temporally unfolding visual, auditory or audiovisual study etc.) should be collected and analyzed --as has continuously been done even then-- but also knowledge about the way the participants (e.g. child, adult, elderly, culture/social group A or B etc.) perceive and make sense of the experiment, its stimuli, its context, and their tasks. In his research with chimpanzees, for example, Köhler focuses on how chimpanzees perceive a given physical environment. Therefore, he does not only evaluate the different behaviors and attitudes of chimpanzees in relation to the manipulated physical layout, but he also tries to find clues about their phenomenal experience of that environment (for example, their height, their past experiences, how long they lived in captivity etc.)[43]. Likewise, Wertheimer would ask how an indigenous person counts a man and a horse in combination and then inquire why they did so. If Wertheimer had thought it sufficient to merely count those who answered "2" as correct and those who did not answer "2" as incorrect, he would conclude that these peoples had a poor number concept. However, he was not content with simply collecting the answers but also tried to understand why these people would say that the sum of a person and a horse be “one”. It is only then that one may see that a person and a horse correspond to a new "indivisible Gestalt” in the minds of the individuals of this community[44] .

Adding such enriched methods to today's experimental studies means that research reports will have to include qualitative analyses as well. Yet, today’s "speeded" science enforced through explicit “publish or perish” policies at universities and research institutes is a serious obstacle for the advancement of a “slower science”, particularly in the empirical, social and behavioral sciences[45]. As much as human kind has understood the dangers of “fast food”, “fast cities” etc., and worked meticulously on their slow versions, such a measure may be even more urgent in our sciences. I believe that a fast kind of science, particularly in the empirical, social and behavioral sciences, is leading to a number of problems.

Firstly, under a constant pressure to publish, scientists are forced to write their work in a haste, the introduction and discussion parts of empirical articles consequently become more and more sketchy, even faulty, leaving out crucial references or citing them by simply copying the way they have come to be referred to (and not too seldom incorrectly)[46]. Such shortened introductions further hamper a structured, systematic, and accumulating review of previous, related scientific work, particularly, if no meta-analytical publication has been done yet. We also know that in this "speed hysteria", journals impose length restrictions as well as restrictions on citing references "not older than 20 years". The latter is particularly dangerous as it not only contributes to an ever-growing amnesia about past knowledge, but also falsely assumes that there is a single scientific paradigm that will remain unchallenged. If this way of doing and publishing science continues, if any new research is not meticulously interwoven into earlier research (which requires a time-intensive literature survey)[47] , I believe that the empirical, social and behavioral sciences face the danger of moving in circles which is no different from hitting a dead end.

Secondly, in today’s science, particularly in the empirical, social and behavioral sciences, most damaging of all is the pressure to present findings in a simple, clear and closed-ended "storytelling" package, ready to consume rather than to stimulate questions (also cf. Firestein, 2012). This is why journals and reviewers alike are aversive to complex, partly inexplicable data, many times forcing researchers to not report those or make them fit through various magical tools of statistics. Thanks to latest pre-registration and open science practices, this issue of filed, unpublished or “revamped” findings is at least partly addressed (cf. Spellman, Gilbert, & Corker, 2017).

Last but not least, since over the past few years we have become aware of the fact that a considerable amount of published findings in the empirical, social and behavioral sciences do not even replicate, let alone be meaningfully tied to related critical research (see Zwaan et al., 2018; also see Witte & Zenker comment in Zwaan et al., 2018).

The purpose of this article was to “excavate” one of the possibly most interesting but least understood theories in psychology. Two aspects of this theory are particularly exciting, one is its potential scope that goes far beyond perception (cf. Mungan, 2021a) and possibly even psychology (Mungan, 2021b). The other is its refusal to position itself in any of two camps (e.g., mind vs. body, materialistic vs. vitalistic, nature vs. nurture, bottom-up vs. top-down processing, objectivity vs. subjectivity) turned rival even hostile within mainstream North American psychology to a point where they have become useless and defective even (e.g., publication blocks on findings that go against a given stance in any of those dichotomies). Gestalt theory rejects the vitalistic view as well as the mechanistic view. It rejects a pure holistic view, which claims that parts melt into a whole and are hence “unanalyzable”, as it rejects an atomistic view of things. But perhaps most importantly, it points out the meaninglessness of the battle on whether something is innate or learned. In contrast to common beliefs about Gestalt theory, it does acknowledge the role of learning in how we perceive and cognize our surroundings but also broods on ideas of “innateness” in a way that goes beyond a simplistic, short-cut evolutionary explanation (“this is so because it was selected for due to its survival value”) by trying to understand what kind of natural principles affect our perception and cognition of our environment. Unfortunately, this approach and the “nativist” approach are often confused. For example, Wagemans et al. (2012), in their meticulous article written for the centennial of Gestalt theory, easily describe Gestalt principles of organization as "innate". Yet, just reading Köhler's striking 1950 article would suffice to see how wrong this is. The confusion here is between a property that belongs exclusively to living organisms (“innate”) and physical processes that define the universe and hence “overarch” the animate world.

This article was written to reclaim a striking theory which looks at the world from a completely different perspective, and whose actual claims, particularly their conceptual claims are little known, either due to a lack of translation or a negligence (possibly because of “not having the time” for it in the haste of research) of reading the original works and instead relying on secondary literature. Hence, it would be wrong to see this paper as a paper about some “dusty theory”. To the contrary, it aims to put this theory back in place. Interestingly, the Gestalt concept never fully disappeared. It appears as if with the advent of increasingly sophisticated mathematical models along with new developments in neuroscience and computer science there is a resurgence in making reference to the term, albeit superficially. A google scholar search with the keywords “Gestalt theory”[48] renders more than 65 thousand records between the years 2010 to 2020, among these, for example, an article titles Gestalt Theory Rearranged: Back to Wertheimer published in Frontiers in 2017 (Guberman, 2017). This theory centers on many issues that are yet to be understood, and it seems that in the 21st century, a view that can examine events with more complex, nonlinear, dynamical systems will be needed.

In closing, I must say that particularly the primary source readings were astounding in terms of their relevance to questions which have remained unresolved within 150 years of psychological research. What was as surprising was to see its potential to be a theory that might overcome the disconnectedness and severe fragmentation across various fields within psychology. Then why did it die out after its very promising beginnings? Maybe it was born prematurely in a time where the scientific community was not ready for such a complex proposal or it failed or was made to fail when its founders were torn apart from their intellectual “habitus” in Germany when the Nazis came to power. The question then emerges as to whether today may be the time for this theory to be “re-embraced”. This potential is discussed in a second article which focuses on Gestalt theory’s proposals about memory (Mungan, 2021a), and a third article which focuses on the theory’s conceptualization of productive thinking (Mungan, 2021b).

Footnotes

[1]The title does not use the term "perception", which may be because in those times all the discourse was on sensation rather than perception.

[2]But see Michael Wertheimer (2014).

[3]In Wagemans et al. (2012), this optimal time was mistakenly quoted as 60 milliseconds. In the original 1912 article, Wertheimer uses the expression “ein Zehntelsekunde”, that is, a tenth of a second.

[4]The Gestalt founders use the term "space-time" extensively because they treat the perceived and experienced world as non-static moving space.

[5]This concurrence of the phenomenal and the physiological will be addressed under the subheading “isomorphism” in the conceptual sections of this article.

[6]  This term was first used by the philosopher Christian von Ehrenfels (1890) to refer to anything that retains its identity when transposed. For example, a melody is still recognized as the same melody when sung from different starting notes, as long as the pitch intervals, the melodic contour and its rhythm is retained. This term will be re-addressed in a later section of this article.

[7]I intentionally used the term empirist rather than empiricist, following Köhler's rationale in his 1950 article “Psychology and Evolution”. Köhler uses "empiricist” to mean the philosophical school and “empirist” to mean any experimentalist who knowingly or unknowingly assumes a simple, mechanistic, associationistic view without necessarily an in-depth philosophical reference.

[8]Although this article has been cited 2270 times according to Google Scholar, it has not yet been fully translated into English. The English text on the web page of University of York, Psychology Classics, provides Ellis's 1938 "sketchy" summary. The original German version is 50 pages long whereas Ellis' English summary is just 18 pages. Ellis also changed the title of the article from Investigations on the Principles of Gestalt to Laws of Organization in Perceptual Forms and hence drops the most critical word in the title: Gestalt. The prequel to this article is Wertheimer’s more theoretical 1922 article.

[9]A detailed description of this can be found in Henle (1978).

[10]I partly benefited from Wagemans (2012 and 2015) and Mandler (2007) while writing these historical sections. Erich Goldmeier's name was brought to my attention by emeritus professor Riccardo Luccio, an important Gestaltist at the University of Trieste, who collaborated extensively with Gaetano Kanizsa, another important Gestaltist who produced the famous Kanizsa triangles of illusory contours.

[11]I would like to thank my student M. Aziz Akkaya for bringing this article to my attention.

[12]The founders of the so-called Gestalt therapy method, Fritz and Laura Perls, openly acknowledged that they never read any of the original works of Gestalt theorists. Today, however, there is a continental Gestalt theoretical school of psychotherapy (cf. Stemberger (ed., 2022), Essentials of Gestalt Theoretical Psychotherapy) which is well-grounded in the theory.

[13]Translated by author. Note that the Ellis translation does not contain the second part. A full translation can be found in Wertheimer, M., edited by Lothar Spillmann (2012). On Perceived Motion and Figural Organization. MIT Press.

[14]As a reminder, like Köhler (e.g., 1938) I use the word “empiricistic” to refer to the actual philosophy of empiricism, and “empirist” for the stance of non-philosophers who are appropriating it (often even unaware of doing so).

[15]Wertheimer and the other Gestaltists frequently use the term “rules” or even “laws” when describing grouping mechanisms in perceptual organization. However, since Gestalt theory also emphasizes the strong dependency of perceptual organization on context and the personal phenomenal field of the observer, I believe it is more appropriate to use the term principles instead of rules or laws.

[16]"Raum- und Tongestalten"

[17]1921 is the date of the German translation whereas the original Danish version dates back to 1915.

[18]I emphasize this because particularly in North American textbooks, there is always the narrative that the Gestaltists simply did demonstrations and no meticulous experimentations. This is clearly an oversimplification and a full negligence particularly of their ca. 11-yr long work in Germany. Even reading Wertheimer’s impressive 1912 experimental report on the phi phenomenon proves the opposite.

[19]In that sense, the tactile system seems much closer to a one-to-one mapping of the external world in terms of direct spatial distance representation.

[20]This principle has been translated by Ellis (1938) as "similarity" rather than “sameness”, although Wertheimer explicitly uses the term “Gleichheit”, not “Ähnlichkeit”. Indeed, all the examples he provides are examples of sameness rather than similarity. Naturally, similar stimuli are also expected to be grouped against less similar ones, however, one might expect many more individual differences and context-dependent effects in the case of mere similarity compared to exact sameness.

[21]Also see http://www.gestalttheory.net/archive/closure.html (I thank Gerrhard Stemberger for this reference).

[22]His 1922 prequel to this paper, Untersuchungen zu der Lehre von der Gestalt. I. Prinzipielle Bemerkungen [Investigations on the Principles of Gestalt. I. Principal Remarks] also uses the term very frequently. In German this term refers to “pronounced“, “clear”, “strong”, “distinct”, “distinctive”, “salient”.

[23]"Self-distribution" in Köhler's (e.g., 1938) terms, "self-organization" in today's usage. But also see http://www.gestalttheory.net/cms/uploads/pdf/GTH-Archive/Köhler1951_1993Prägnanz.pdf (I thank Gerhard Stemberger for this reference).

[24]Today's people might also see the letter "H".

[25]Here, Wertheimer points to the possibility of differences in perception even among people from the same culture. This was re-noticed and presented decades later by David Navon (1977), who pioneered a large literature on global versus local perception.

[26]Showing that something is not universal does not go against Gestalt theory, this is unfortunately a big misunderstanding because of a lack of reading of the original, theoretical works of its founders.

[27]This expression refers to peoples from non-industrialized cultures.

[28]I will present examples of this particularly in my second article (Mungan, 2021a), where I will examine the propositions of the Gestalt theorists on memory processes.

[29]Each shape has its white/black counterbalanced version as well, to ensure that it is not the black or white color but the shape-based dynamics that determine figure or ground identity.

[30]Today, the issue of "complexity" is still one of the most challenging topics in cognitive science. There are various attempts of how complexity can be defined and systematically be manipulated, some of these come directly from information processing theory (Shannon, 1948; for music see Pearce, 2005). The issue of figure-ground perception is a still intensely studied topic (e.g. Grossberg, 2016; Peterson, 2015).

[31]Lydia Maniatis brought up a critical objection to this interpretation: “I don't think it's quite correct to describe this in terms of figures being buried in the ground because Gottschaldt was trying to show that what figures or forms will emerge in perception depends on the overall structure and not on our expectation or experience. It's not that the cube in the figure 30 becomes part of the ground but that its elements are integrated in a different figure.” (Personal communication). When re-inspecting Figure 30b, as much as it is more complex than simple, more heterogeneous than homogenous, (as such fulfilling, I would say, two typical characteristics of “groundness”, as listed by Koffka), it is hard to describe it as less articulated as one likewise would expect from forms with ground rather than figural characteristics. Hence one should probably refrain from calling 30b a ground and say that in the Gottschaldt stimuli, figures are made invisible by burying them into another figural constellation.

[32]Vitalism suggests that living beings are the bearers of non-physical "life elements" that qualitatively differentiate them from non-living beings, hence making them exempt from scientific inquiry.

[33]As emphasized earlier, while looking at the global structural field dynamics of a given stimulus, Gestalt theory also places importance to the way the observer interact with it. A recent study which beautifully incorporates both components is Mundy's (2014) study on the influence of a local-global priming on the Müller-Lyer effect.

[34]Two recent studies show that the Jastrow illusion is seen in chimpanzees (Tomonaga, 2015), but not in macaques and capuchin monkeys (Agrillo, Berran, Parrish, 2019). These differences do not pose a problem for Gestalt theory because its critical conjecture is that different forms of perception do not have to depend on past experience. It would be quite bizarre to propose that in this case, we should conclude that the experience and knowledge of a chicken or a chimpanzee is matched with that of a human, but not with that of a macaque or capuchin monkey.

[35]It is surprising that the Gestalt founders while providing a plentitude of examples from music seemed to have almost fully ignored language. Certainly, it was not up until the launching of the cognitive revolution in the late 1950s/early 1960s that psycholinguistics, and as such, experimental research on language processing emerged. Hence, we might surmise that both Wertheimer and Koffka died too early, and Köhler was likely too busy trying to convince the behaviorist-turned-cognitivist scientific community that they were still committed to the very same assumption that truth can be reached by a step-by-step understanding of isolated parts. On the other hand, given that linguistics had been blooming since the late 19th century, I do not believe that this is a sufficient explanation.

[36]This is strongly reminiscent of Raichle's discovery that the brain has an activation pattern even in the absence of concrete mental activity, a discovery that could have been made much earlier if neuroscience had not been fixated to a given way of seeing and explaining things.

[37]Actually, Koffka uses the term "behavioral environment", but what he describes is the phenomenality from within which one perceives, experiences, and acts.

[38]Yet, we can find early cues of valuing an organism’s meaning-making in Tolman (e.g., Tolman & Honzik, 1930; Tolman, 1932), as he recognized the inadequacy of the behavioral school trapped in the stimulus-response narrowness. Hence, it comes as no surprise that both Köhler and Koffka mention Tolman’s studies quite often. Yet, Tolman left (or had to leave) the “O” part of his S(timulus) – O(rganism) – R(esponse) equation underdeveloped whereas it was that very “O” part which Gestalt theory and their phenomenology would analyze, and not as an “information-processing machine” but as an entity that has values and creates meaning as a natural part of its environment.

[39]I discuss Köhler’s intriguing studies with chimpanzees in more detail in a third article on Gestalt that focuses on Gestalt concept of ‘productive thinking’ (cf. Mungan, 2021b, for Turkish version).

[40]Koffka uses the term "ego", while Köhler uses the term "self" (for a discussion, see Stemberger https://www.academia.edu/49007634/). Köhler also puts forward interesting ideas about the "place" of this "self" in the living being and somehow argues that here too, mind and body should not be separated, but on the contrary, be seen in a unitary way. We can find the contemporary equivalent of this view in Damasio (see Damasio, 2018) and Varela, Thompson, & Rosch (1991/2016).

[41]It is worth noting that Köhler was the first to use the term “direct perception” (e.g., p. 151 in Köhler, 1938) and we know that Gibson was a regular attendee of Köhler’s seminars).

[42]Gibson differs from the Gestaltists in that he thinks that the careful mathematical study of direct perception is entirely sufficient to understand ecological perception (i.e., perception within the natural environment) because of the plentitude of information and invariances that allow a direct grasp of what is “out there”. Unlike the Gestaltists, according to Gibson, the 'geographic’ and 'phenomenal environment’ –to use Koffka’s terminology-- overlap perfectly, and the historical background of the perceiving living beings will not make a fundamental difference. Of course, it should not be forgotten that Gibson only studied visual perception, while Gestalt theorists aimed to understand all kinds of cognition and behavior of living beings.

[43]As mentioned, Köhler's 1921 study will be presented in more detail in a later Gestalt article (cf. Mungan, 2021b).

[44]An interesting resemblance can be found between this “oneness” and the 15th century indigenous people of South America who would perceive the Spanish horseback colonizers (the so-called “conquistadors”) as one being, since they never had seen horses before (cf. Galeano, 1971).

[45]I borrow this expression from Frank Zenker.

[46]It would be interesting to do a study just on this, where critical studies have been misrepresented, even utterly misunderstood. In our third article (Mungan, 2021b), we give one such example.

[47]Endel Tulving is said to be one of the most meticulous persons in this respect.

[48]When only using "Gestalt" as a keyword, plenty of articles pop up about Gestalt therapy that are not actually built on the basic readings of the theory, but are rather based on the "apparent" concept of the theory. The keyword "gestalt theory", in contrast, mostly lists Gestalt theory based articles.