Peer Frantzen

Essay

The Association Experiment: The Forgotten Science of Postmodern Architecture: From Psychophysics to the Perceptron

Author
Peer Frantzen
Published
6 May 2026, Zenodo
DOI
10.5281/zenodo.20049562
Licence
CC BY 4.0
Keywords
association, postmodernism, Kandinsky, Jencks, perceptron, architectural meaning, psychophysics, cybernetics, structuralism, Hebbian learning

PDF on Zenodo · Academia.edu

Abstract

This essay traces the concept of association across two centuries of experimental inquiry — from nineteenth-century psychophysics to the perceptron — by following a single recurring gesture: the attempt to make visible the connections between things and meanings. Moving through Fechner, Galton, Ebbinghaus, Freud, Jung, the Surrealists, Kandinsky, Lévi-Strauss, and Rosenblatt, it argues that the association experiment is never neutral. Each era reshapes it according to institutional, political, and epistemological demands. Kandinsky uses it to find a universal visual language for a new society; Jencks uses the same instrument to demonstrate the irreducible plurality of meaning; Rosenblatt builds a machine that learns associations without a subject. The essay proposes that the juxtaposition of Kandinsky and Jencks constitutes a dialectic at the heart of the modernism-postmodernism debate — not a contrast between two aesthetics but between two accounts of where meaning is located in the associative stack. The synthesis that dialectic demands is complicated by the computational turn, which introduces a third term neither position was prepared for: a system that operates on the associative mechanism directly, without addressing meaning at any layer. The essay concludes by asking what architectural discourse has not yet asked — which layer its tools are built to address, and which layers they leave unaddressed.

Introduction

Fifty years apart, both Charles Jencks and Wassily Kandinsky conducted studies with questionaires. The instruments are nearly identical, but their aims diverge. Kandinsky seeks universal laws — a visual language beneath contingent associations. Jencks seeks the opposite: plurality, ambiguity, the irreducibility of meaning. The same method produces different knowledge. This essay traces that divergence by following the association experiment across two centuries — from nineteenth-century psychophysics to the perceptron. In each context, the experiment is reshaped by institutional, political, and epistemological demands. Kandinsky pursues a universal language for a new society; Jencks demonstrates plurality against modernist singularity; Rosenblatt makes association operational. The experiment itself — questionnaire, word test, neural network — is never neutral. Each era uses it to extract what it needs: universality, multiplicity, or action. What recurs is the tracing of connections between things and meanings; what changes is the purpose of that tracing — and ultimately the question of whether association requires a mind at all. For architecture, this is not abstract. The associative operations that design both relies on and produces have long been subject to systematic inquiry, though rarely understood as a continuous lineage. The web of associations is not given; it reflects what a society values — whether universality, ambiguity, or statistical prediction.

The Questionnaire and the Building

A spectator is handed a questionnaire. What does this building remind you of? The answers trickle in: trains, trash-cans, letter-boxes. A chain of associations emerges, leading finally to a value judgement — good or bad, beautiful or grotesque. Charles Jencks, the man who would later give postmodernism its name, watches as meaning is revealed not in the building itself but in the network of metaphors it evokes. Jencks’ experiment, described in “A Semantic Analysis of Stirling’s Olivetti Centre Wing” (1980), was an attempt to map the way the mind sees architecture. Borrowing techniques from experimental psychology, he sought to trace the ‘associative chains’ that guide perception, to show that aesthetic judgement is neither immediate nor objective but formed through a sedimentation of past observations. Meaning, he argued, arises from a sign’s relations in a semantic network, from its metaphoric associations absorbed from its cultural context. 1 Meaning is “conveyed through associations, metaphors or the whole treasure of past memory”2 and visual information is subject to “inevitable semantization.”3 The Olivetti Centre, by James Stirling, is used as a case study. The building combines traditional references with unconventional elements: two obliquely arranged classroom wings clad in vivid prefabricated polyester panels, curved plastic surfaces sweeping across walls and roofs, a glazed greenhouse-like structure acting as circulation spine connecting to the original Edwardian house. Three different codes coexist — modernist, futurist, historicist — and it is their collision that makes the building generative for Jencks’ purposes. Jencks’ premise is that architecture is classified in terms of metaphors, that similarity determines aesthetic judgement. Similarity with attributes assigned according to values which have become attached to past observations. Meaning is created through “its relation to all the other signs in a context or chain” and “through the other signs for which it has become a metaphor by association, or similarity.”4 A few decades earlier, in the nineteen-twenties, Wassily Kandinsky — lawyer, painter, revolutionary — conducted a comparable inquiry inside a Soviet institution in Moscow. He too handed questionnaires to subjects and asked them to report their associations. He too sought to trace the links between form and feeling. But where Jencks wanted to show that meaning is plural, contingent, irreducibly personal, Kandinsky wanted the opposite: to find the universal beneath the contingent, a visual language grounded not in culture but in the nervous system itself. Two figures, separated by fifty years and an ideological chasm, performing the same gesture for opposite reasons. What the association experiment means — what it is for, what it can prove — depends entirely on who is holding the questionnaire, and why.

The Laboratory and the Web

But the experiment belonged to a much older tradition. Associations had occupied philosophers, artists, architects, pedagogues, and psychologists long before anyone thought to measure them.

Archibald Alison theorised that “if we trace them back, we shall discover not only a connection between the individual thoughts of the train, but also a general relation among the whole.” 5 The concept drifted across disciplines, shifting as it moved — from philosophical afterthought, to natural law, to cognitive mechanism, to something that could, by the nineteenth century, be measured. In 1876, Gustav Fechner published a series of experiments on synaesthetic correspondences between vowels and colours. He organised his data carefully: vowel, sex of respondent, reported quality — glänzend (shiny), durchsichtig (transparent), unbestimmt (undetermined).6 The results showed clear tendencies rather than random variation. The vowel ‘a’ was most frequently associated with white and red; ‘i’ inclined toward yellow; ‘u’ toward black. From these correspondences Fechner drew a larger picture: people inhabit a web of associations in which every node is connected to others, and to touch one region is to set its neighbours vibrating. The thing is seen immediately with its associated meaning; to separate the thing from its significance would be an intellectual and effortful act — the default is fusion, not analysis. He used the example of an orange. Its aesthetic appeal lies not in its colour or its form taken in isolation, but in the cluster of memories and sensations it gathers: a taste, perhaps the image of a tree, a moment of sun in France or Italy or Spain, a romantic longing. The orange is a locus of conditions that evoke pleasure, not a property of the fruit itself. Three years later, inspired by Fechner’s methods, Francis Galton devised one of the earliest word association tests. He began by walking down Pall Mall in London, letting each object he saw evoke associated thoughts and making brief mental notes before moving on. Repeating the walk, he was struck by the persistence of the same associations — the same ideas returning to the same objects, as if the mind had worn grooves. The experience, he wrote, was like discovering for the first time “the complex system of drains and gas and water pipes, flues, bell-wires, and so forth, upon which our comfort depends, but which are usually hidden out of sight.”7 He refined the method: writing stimulus words on separate slips, setting them aside for several days, then sliding each partway under a book so that a single unpredictable word would meet his eye. In his other hand he held a small chronograph, starting it the instant the word appeared and stopping as soon as two associated ideas had formed. He noted the associations and the elapsed time. The problem, which Galton only partially acknowledged, was that the associations on Pall Mall were his associations — shaped by class, education, and a very particular Victorian biography. The experiment was contaminated by the very self it was supposed to observe. Ebbinghaus recognised this contamination as the central methodological problem and invented a solution: the nonsense syllable. He constructed approximately 2,300 units — jös, häm, mök, nit, a vowel placed between two consonants — and combined them into series of varying lengths, typically twelve to sixteen syllables long.8 He read each series aloud at a regulated tempo of 150 beats per minute, keeping time with a metronome. The goal was to repeat a series until he could reproduce it faultlessly from memory at the same tempo. By eliminating prior interest, aesthetic quality, and established linguistic association, he believed he could isolate the pure mechanics of how the mind retains and reactivates connections — memory stripped of meaning, association without content. What Ebbinghaus demonstrated, in the end, was that associations can exist and exert influence even when they cannot be consciously recalled. The web is there whether or not you can see it. Fechner, Galton, and Ebbinghaus were not asking the same question. Fechner was interested in the aesthetic and symbolic dimensions of association — what a form evokes, what emotional resonance it carries. Galton was interested in the individual genesis of association — the rapid, involuntary, deeply personal pathways through which thought unfolds. Ebbinghaus wanted to strip all of that away and study the mechanism underneath. Three different aims, three different methods — but all three understood association as something that could, in principle, be measured. What had once been a philosophical concept had become, by the end of the nineteenth century, an experimental object.

Dreams, Automatism, and the Associative Unconscious

Ebbinghaus had tried to isolate memory from the contingency of association. But at the turn of the century, and especially after the rupture of the First World War, the contamination itself became the object of interest. For Freud and Jung, association was not something to suppress but something to trace — the anomalies in the web, the hesitations and disturbances, were where the meaning was hiding. Freud’s early ambition had been to build a quantitative, neurological model of the mind. But it was his observation that the compulsion to associate prevails in dreams that led to the work that made him famous. Dreams became the material substrate of the talking cure — broken into fragments, freely associated upon, their chains of connection traced back toward the hidden causes of neurosis. The method was not so different from Galton’s walk down Pall Mall: follow the associations, and see where they lead. The difference was that Freud did not want to map the associations. He wanted to find what they were concealing. Jung noticed Freud’s work on dreams around 1900 and recognised its relation to his own association experiments.9 His method used a formulary of one hundred stimulus words chosen to represent common life situations. Subjects responded with the first thought that came to mind. Failures in memory, prolonged reaction times, unwanted repetitions — these were the significant moments. The disturbing variable, which Ebbinghaus had invented nonsense syllables to eliminate, became the primary source of meaning. What had been methodological noise became clinical signal. The same gesture that Galton had used to map his own unconscious, and that Ebbinghaus had used to purify memory of its contents, was now being used to locate the precise points where the web of association had knotted around a wound. In 1917, the young poet André Breton, drafted into the war as a nurse in Nantes, encountered the early psychoanalytic writing of Pierre Janet alongside traumatised soldiers and the fragility of the mind. Shortly after, he was introduced to Philippe Soupault. On Soupault’s initiative they began an experiment in automatic writing. “Drops of water hold us prisoner, we are no more than perpetual animals” was its first sentence. Breton sent a copy of the resulting work, The Magnetic Fields, to Sigmund Freud in Vienna. Freud never responded.10 What Breton found in Freud was that the unconscious has logic. This logic allowed the Surrealists to make association into a tool — a deliberate estrangement that challenged meaning. The products of free association and automatic writing had nothing to do with aesthetic criteria, and everything to do with what Breton called an attack on “simplistic materialism.”11

“Psychic automatism in its pure state, by which one proposes to express — verbally, by means of the written word, or in any other manner — the actual functioning of thought. Dictated by thought, in the absence of any control exercised by reason, exempt from any aesthetic or moral concern.”

On October 11, 1924, the Surrealists opened a Bureau for Surrealist Research at 15 rue de Grenelle. A new periodical followed: La Révolution surréaliste, designed to resemble a scientific journal. Its last page asked readers to contribute their dreams, writings, drawings, and automatic objects — all in the service of collecting experimental data for an end that could not yet be perceived.12 The project most committed to irrationality organised itself as a research bureau, adopted the institutional form of the sciences it was attacking. The form mimicked the institution; the content dissolved it. Both INKhUK and the Bureau for Surrealist Research collected empirical responses to form. Both sought to systematise the associative life of the mind. The difference is that Kandinsky wanted to reduce that life to laws, while Breton wanted to preserve its irreducible strangeness.

The Universal and the Revolutionary

INKhUK was instituted shortly after the October Revolution, as part of the People’s Commissariat for Education.13 Its psychophysical department was assigned to “investigate the internal positive laws according to which artworks in each individual art field are created.”14 Kandinsky would be its director. His questionnaire probed associations with colour, form, and the correlation between sound and abstract shapes. He asked whether complex technical drawings elicited specific emotional responses. Whether triangles, squares, curved or pointed lines brought forth certain feelings. Whether feelings could be expressed by drawing. Some questions have a quality almost comic in their ambition — whether a triangle is “more intellectual than a square,” for instance.15 Kandinsky’s ambitions envisioned an encyclopedia of abstract qualities, a systematic framework that could establish a universal visual language based on psychophysical effects. He was not conducting research for its own sake. The questionnaire was a political instrument: to build a new language for a new society, one that reached beneath the contingencies of class, taste, and personal history to something universal and physiological.

Modernism and Postmodernism: The Antithesis

Jencks takes Kandinsky’s method and turns it against Kandinsky’s premise. Where Kandinsky used the questionnaire to find the universal beneath the contingent, Jencks uses it to prove that the contingent is all there is. The same instrument, the same gesture, the opposite conclusion. This is not merely a stylistic reversal. It is a philosophical one. Modernism’s formal language — abstraction, reduction, the elimination of ornament — rests on the claim that form can speak directly, beneath culture, to the nervous system. Postmodernism’s counter-claim is that there is no beneath. Every form arrives already laden with association, already caught in a web of metaphor and memory that no amount of reduction can dissolve. The white wall is not neutral. The right angle is not universal. They too remind you of something. Jencks referred, in passing, to Victorian associationism — to the architect J. C. Loudon, who “proposed a theory of associationism based on the notion of association of ideas” and argued that every house should convey the character of its owner.16 But Kandinsky’s project was its opposite: not the communication of individual identity through form, but the abolition of individual contingency in favour of universal response.

Cybernetic Experiments

In 1958, Frank Rosenblatt built a machine capable of recognising simple letters — one of the first functioning artificial neural networks. He connected a field of receptors to a network of nodes and wires, drawing on the mathematical model of the neuron that McCulloch and Pitts had developed fifteen years earlier. Where their model was boolean — neurons fire or do not fire — Rosenblatt’s introduced probability, and with it, learning. The perceptron consists of three integrated layers. The S-system (sensory) serves as input, sending impulses to the A-system — the association layer, which functions as the machine’s memory, activating when signals exceed a certain threshold. Initially randomised, the system learns through iteration by reinforcing connections consistently activated together. Finally, the R-system produces output. Information, as Rosenblatt put it, “is stored in connections or associations.”17 Rosenblatt described his paper as originating in “independent research in the field of physiological psychology.”18 The language is telling. The perceptron was not understood, even by its maker, as a departure from the human sciences, but as their extension into hardware. The association layer was named as such: A-Units, associative neurons, a vocabulary carried intact from Galton’s Pall Mall walk and Fechner’s laboratory into the wiring of a machine. The perceptron can be understood as a synthetic association experiment: instead of asking a human subject what a stimulus evokes, the system learns which response should follow a given stimulus. It builds a network of statistical associations between sensory inputs and conceptual labels. What began as a question of how the human mind connects a thing to its meaning becomes, in Rosenblatt’s perceptron, a question of whether meaning requires a mind at all. Nicholas Negroponte was among the first to imagine what this might mean for architecture. “This discussion is not about machines that necessarily can do architecture,” he wrote; “it is a preface to machines that can learn about architecture and perhaps even learn about learning about architecture.”19 The neural networks of his time were not advanced enough to answer his imagination. Most architectural proposals of the period focused on interactive and responsive systems — feedback, adaptation, dialogue — rather than machine recognition. The Rosenblatt genealogy would not become architecturally legible until much later, and by then it wore a different name.

Postmodernism and Cybernetics: Two Renovations

In the 1960s, Claude Lévi-Strauss wrote that structuralism ought “to rescue associational psychology from the discredit into which it had fallen.” A “renovated associationism” would show a logic not of three simple rules but one more like boolean algebra, a complex of oppositions, inclusions, and compatibilities.20 Structuralism, with all its heterogeneous voices, represented this renovation. It was not a break from associationism; it was its rehabilitation, conducted through the methods of linguistics. Jencks drew on this tradition. His concept of postmodernity is enveloped in structuralist discourse — signs, semantic networks, chains of metaphor. When he handed questionnaires to passersby outside the Olivetti Centre, he was working within a framework that Saussure, Barthes, and Eco had built on associationist foundations, even when they did not use that name. At almost the same moment, in the United States, Norbert Wiener noted that his cybernetic theories showed “many points of contact with Associationism on the one hand and with the theories of Pavlov on the other.”21 Warren McCulloch spoke of “associative neurons.”22 The same conceptual inheritance was being renovated simultaneously in two directions — one toward the irreducible human subject, the other toward a system that needed no subject at all. Jencks and the cyberneticians were contemporaries. Structuralism and cybernetics were contemporaries. They did not speak to each other. Architecture absorbed the first and, later, the second — without noticing that they rested on incompatible assumptions about what a mind is and what meaning requires.

Comparison: Layers of Meaning

The experiments of Jencks, Kandinsky, and Rosenblatt share a gesture. Each attempts to trace the connection between a thing and what it means — to make visible the link between stimulus and significance. Each asks, in its own vocabulary, the same question: how does a form acquire meaning? That the question persists across a century of institutional rupture suggests something. Not that the investigators are working in the same tradition, but that they may be circling the same phenomenon from positions their disciplines have made difficult to compare. The shared substrate is Hebbian: neurons that fire together wire together. Repeated co-activation strengthens connection. This is what Rosenblatt computes, what Kandinsky tries to reach beneath culture to find, and what Jencks tries to map at the cultural surface. The mechanism is the same. What each asks of it is completely different. For Rosenblatt, the question is recognition. Whether a machine can learn to distinguish a square from a circle. The meaning of the square is irrelevant — what matters is that the network, through iteration, reliably produces the correct output. The firing pattern is everything; what it feels like, what it evokes, what it reminds you of, does not enter the system. For Kandinsky, the square carries meaning that precedes culture — a psychophysical response grounded in the nervous system itself, beneath the contingencies of class, taste, and biography. The questionnaire is designed to strip away cultural accretion in order to find this bedrock. Association is what remains when the personal is eliminated. For Jencks, the personal is precisely the point. Association is not what lies beneath cultural meaning but what constitutes it. The sediment of past observations, the chain of metaphors, the idiosyncratic memory — these are not noise obscuring a universal signal. They are the signal. The building means what it reminds you of, and what it reminds you of is irreducibly yours. What changes across these three positions is not the mechanism but the layer each investigator chooses to address — and what they decide to do with the layers they ignore. Rosenblatt ignores meaning entirely and gains a working machine. Kandinsky ignores cultural contingency and gains a universal language that turns out to be historically specific. Jencks ignores the physiological substrate and gains plurality at the cost of any claim to generality.

Conclusion

Two figures, fifty years apart, hand a questionnaire to a stranger. The gesture is almost identical. The paper is similar. The question, in general terms — what does this remind you of? — is the same. But between Kandinsky in Moscow and Jencks outside the Olivetti Centre, everything else has changed: the institution, the ambition, the theory of what a mind is and what meaning requires. The method travels intact across ruptures — revolution, war, the collapse of modernism — because it is capacious enough to serve opposing purposes. Kandinsky uses it to abolish the contingent self. Jencks uses it to reinstate it. Rosenblatt removes the self entirely. First the self is observed through introspection, then theorised, then measured from the outside, until finally it is built. The self is first observed from the inside, then the outside, then modelled. What these experiments share is the question: how do people connect ideas, words, feelings, colours, and sounds with what is seen or heard? Each questionnaire attempts a form of cognitive cartography — responses organised to understand how associations form in chains and patterns. What they do not share is any agreement about what those chains are, who they belong to, or what authority their tracing carries. A trace made in a laboratory carries different weight than one made in a revolutionary commissariat, a Surrealist bureau, or handed to a passerby outside a building in postmodern England. Structuralism and cybernetics renovated associationism simultaneously, in opposite directions. Jencks and Rosenblatt were contemporaries who never spoke to each other. Architecture has since inherited both — the humanist language of metaphor, memory, and the meaning-making subject, and the computational tools of a system that requires no subject to function.

What This Means for Architecture

Jencks’s experiment is more than a methodological borrowing from experimental psychology. It is, in the most precise sense available, a definition of what postmodernism is. Not a style, not a period, not the return of ornament or historical reference — but a specific claim about where meaning lives. It lives in the associative chains of the perceiving subject, in the culturally sedimented sediment of metaphor and memory that every form inevitably activates. A building means what it reminds you of. There is no form that escapes this process; there is only form that acknowledges it and form that pretends otherwise. The pretence, in this account, is modernism. Kandinsky’s questionnaire is not incidental to the modernist project — it is its epistemological foundation. The reduction to essential geometry, the elimination of ornament, the aspiration toward a universal formal language: these rest on the claim that there is a layer beneath cultural association where form speaks directly to the nervous system. That the triangle is more intellectual than the square not because of what triangles have come to mean in a given culture, but because of what angular form does on itself. The juxtaposition of Kandinsky and Jencks is therefore not a contrast between two aesthetics. It is a dialectic between two accounts of the same phenomenon. Kandinsky reaches beneath the cultural surface toward the physiological substrate. Jencks insists the cultural surface is the only surface there is. Both are describing something true. Neither is describing everything. The synthesis this dialectic demands is not a compromise between universality and plurality. The Hebbian mechanism underlies both. Neurons that fire together wire together — but which neurons fire together first is a cultural and biographical fact. The bedrock Kandinsky was searching for is real. It is just not separable from the sediment Jencks was mapping. But, if the dialectic between modernism and postmodernism is at bottom a disagreement about which layer of association architecture addresses, then the discipline’s current turn toward computational and generative tools introduces a third term.

Bibliography

Alison, Archibald. Essay on the Nature and Principles of Taste. Edinburgh, 1790.

Breton, André. Manifestoes of Surrealism. Ann Arbor: University of Michigan Press, 1969.

Breton, André. La révolution surréaliste, n°2, 1925.

Colquhoun, Alan. “Postmodernism and Structuralism: A Retrospective Glance.” Assemblage, no. 5, 1988.

Ebbinghaus, Hermann. Über das Gedächtnis: Untersuchungen zur experimentellen Psychologie. Leipzig, 1885.

Fechner, Gustav Theodor. Vorschule der Aesthetik. Leipzig: Breitkopf und Härtel, 1876.

Flach, Sabine. Die WissensKünste der Avant-Garden. Berlin, 2013.

Galton, Francis. Inquiries into Human Faculty and Its Development. London: Macmillan, 1883.

Jencks, Charles. “A Semantic Analysis of Stirling’s Olivetti Centre Wing.” Architectural Association Quarterly, 1980.

Jencks, Charles. The Language of Post-Modern Architecture. New York: Rizzoli, 1977.

Jencks, Charles, and George Baird, eds. Meaning in Architecture. New York: Braziller, 1969.

Lévi-Strauss, Claude. Totemism. Translated by Rodney Needham. London: Merlin Press, 1964.

Lévi-Strauss, Claude. Structural Anthropology. Translated by Claire Jacobson and Brooke Grundfest Schoepf. New York: Basic Books, 1963.

McCulloch, Warren, and Walter Pitts. “A Logical Calculus of the Ideas Immanent in Nervous Activity.” Bulletin of Mathematical Biophysics 5, 1943.

Negroponte, Nicholas. “Toward a Theory of Architecture Machines.” Journal of Architectural Education 23, no. 2, 1969.

Pias, Claus, ed. Cybernetics: The Macy Conferences 1946–1953. Zürich: Diaphanes, 2016.

Pitts, Walter, and Warren McCulloch. “The Statistical Organization of Nervous Activity.” Journal of the American Statistical Association 4, 1947.

Plotnikov, Nikolaj. “Einleitung.” In Kunst als Sprache — Sprachen der Kunst, edited by D. Kassek and N. Plotnikov. Hamburg, 2009.

Podzemskaia, Nadia. “Die Kunsttheorie Wassily Kandinskys und die Anfänge der GAChN.” In Kunst als Sprache — Sprachen der Kunst. Hamburg, 2009.

Polizzotti, Mark. André Breton: Selections. Berkeley: University of California Press, 2003.

Rosenblatt, Frank. “The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain.” Psychological Review 65, no. 6, 1958.

Venturi, Robert, Denise Scott Brown, and Steven Izenour. Learning from Las Vegas. Cambridge: MIT Press, 1972.

Wiener, Norbert. Cybernetics: Or Control and Communication in the Animal and the Machine. New York: Wiley, 1948.

Wiener, Norbert. Cybernetics: Or Control and Communication in the Animal and the Machine. New York: Wiley, 1948.

Notes

  1. Jencks, Meaning in Architecture, 20. The concept of 'inevitable semantization' draws on semiotics: any visual form, once perceived, is drawn into a web of cultural meaning whether the designer intends it or not. ↑
  2. Jencks, Meaning in Architecture, 12. ↑
  3. Jencks, “A Semantic Analysis of Stirling’s Olivetti Centre Wing”, 238. ↑
  4. Jencks, “A Semantic Analysis of Stirling’s Olivetti Centre Wing”, 238. ↑
  5. Alison, Essay on Taste. Alison’s associationism was widely read by architects in the early nineteenth century and formed the basis for picturesque theory—the idea that landscape and architecture derive their beauty not from intrinsic formal properties but from the ideas and emotions they call up in the observer. ↑
  6. Fechner, Vorschule der Aesthetik, Erster Theil, 316. Fechner’s project was to put aesthetics on an empirical footing —what he called ‘aesthetics from below’, built from accumulated experimental data rather than deduced from philosophical principles. ↑
  7. Galton, Inquiries into Human Faculty and Its Development, 134. ↑
  8. Ebbinghaus, Über das Gedächtnis. The nonsense syllable was Ebbinghaus’s solution to a philosophical problem: if every word carries prior meaning, how can memory itself be isolated and measured? By using meaningless units, he believed he could study the pure mechanics of retention and forgetting. The method was widely adopted, though critics noted that the elimination of meaning was itself a kind of distortion—human memory, after all, works largely through meaning. ↑
  9. Fechner, Vorschule der Aesthetik, Erster Theil, 89. ↑
  10. Jung’s word association test was later standardised and widely used in both clinical and forensic contexts. The reaction-time anomalies Jung identified—hesitations, repetitions, failures of memory—became the basis for what he called the ‘complex’: a cluster of emotionally charged associations capable of disrupting conscious thought. The polygraph, or lie detector, is a distant descendant of the same logic. ↑
  11. Polizzotti, André Breton: Selections, 5. Freud’s silence was characteristic. He was generally sceptical of the Surrealists’ enthusiasm for psychoanalysis, writing to Breton in 1932 that though he received ‘every evidence of your esteem’, he found it difficult to grasp what Surrealism was. He regarded their interest in the unconscious as aesthetic rather than therapeutic. ↑
  12. Breton, Manifestoes of Surrealism, 26. ↑
  13. Breton, La révolution surréaliste, n°2, 31. The Bureau collected dreams, drawings, and automatic texts submitted by visitors and correspondents. The periodical La Révolution surréaliste was deliberately designed to resemble a scientific journal—columns, press clippings, data—while its content was dreams and hallucinations. The form mimicked the institution it was attacking. ↑
  14. Podzemskaia, “Die Kunsttheorie Wassily Kandinskys und die Anfänge der GAChN”, 179. ↑
  15. Plotnikov, “Einleitung”, 12. INKhUK and GAChN were not isolated experiments but part of a broader Soviet ambition to reorganise cultural production on scientific foundations. The psychophysical laboratory at GAChN ran parallel research into the physiological effects of colour and form, informed by the same Fechnerian tradition. ↑
  16. Flach, Die WissensKünste der Avant-Garden, 124. ↑
  17. Pitts and McCulloch, “The Statistical Organization of Nervous Activity”, 97. ↑
  18. Rosenblatt, The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain, 389. ↑
  19. Rosenblatt, The Perceptron, preface. The perceptron’s immediate reception was enthusiastic but short-lived. Minsky and Papert’s Perceptrons (1969) demonstrated formal limitations of single-layer networks, effectively freezing neural network research for over a decade. The approach was revived in the 1980s with the development of backpropagation, which Rosenblatt had gestured toward in his concept of ‘back-coupled perceptrons’. ↑
  20. Jencks, Language of Post-Modern Architecture, 75. ↑
  21. Lévi-Strauss, Totemism, 90. The full passage is worth noting: Lévi-Strauss argued that associationism had the ‘great merit of sketching the contours of this elementary logic, a direct expression of the structure of the mind (and behind the mind, probably of the brain)’—a claim that connects structuralism directly to the neurological ambitions of cybernetics, though Lévi-Strauss himself did not draw that connection. ↑
  22. Wiener, “Cybernetics”, 4. Wiener’s acknowledgement was not merely diplomatic. The feedback loops central to cybernetic theory—where output is fed back as input to regulate a system—had a structural parallel in the associationist idea that repeated co-activation strengthens connections. Donald Hebb would formalise this as a learning rule in 1949: neurons that fire together wire together. ↑