Directed Diffusion and Orientational Harmonics: Long range boundary completion through short range interactions

Ph.D. Thesis, Boston University 1994.

Visual illusions offer an invaluable tool for exploring the mechanism of visual perception, because illusory contours and surfaces appear predictably in response to certain configurations of visual inducers. Computer simulations that can replicate those same visual illusions therefore reveal the computational principles behind early visual processing. In this thesis I discuss the limitations of neural network theory to account for illusory contour formation, and present an alternative approach using analog field-like processes as suggested by Gestalt theory. The Directed Diffusion model accounts for collinear illusory contour formation by a spatial diffusion mechanism, while the Orientational Harmonic model proposes patterns of standing waves in the neural substrate to account for illusory contour formation through image vertices and intersections. The key ideas in this thesis appear in more concise and clear form in my MLRF and Directional Harmonic Theory papers respectively.

Harmonic Resonance in Visual Perception Suggests a Novel Form of Neural Communication

This was my very first paper which proposes an interesting new principle of neurocomputation to address some fundamental problems with the conventional concept of fixed receptive fields. Submitted to Perception and Psychophysics November 1994, Rejected June 1995.

Symmetry and Resonance in Visual Perception

Whether designing animals, insects, or plants, Nature draws upon symmetry and periodicity to play a fundamental role in defining the body plan. When implemented with the proper chemical mechanisms, these principles guide our bodies from single-celled embryos to bilaterally symmetric creatures with intricate periodic structures, such as the spine and rib cage. The properties of symmetry and periodicity also appear to be fundamental to visual perception. I propose that this is no accident but is a manifestation of harmonic resonance in visual perception. Submitted to Spatial Visionfor a special issue on symmetry Rejected July 1994.

Gestalt Isomorphism I: Emergence and Feedback in the Perception of Lightness, Brightness, and Illuminance

Gestalt Isomorphism II: The Interaction Between Brightness Perception and Three-Dimensional Form

This was my first attempt to publish my most radical ideas about isomorphism. The paper matured considerably during the three years that it was in review, and in its final form it was certainly deserving of publication, but for a particularly bone-headed set of reviewers who seem to consider all radical ideas to be beyond the bounds of science! The best ideas from this paper found their way into my Computational Implications I paper, and my Gestalt Isomorphism paper. Originally submitted to Perception in 1996 for a special issue on brightness perception. Rejected October 1996. Substantially revised and split into two parts, resubmitted October 1997, revise & resubmit January 1998. Resubmitted August 1998, rejected April 1999

Gestalt Isomorphism and the Quantification of Spatial Perception

A short paper on the dimensions of conscious experience and their implications for theories of spatial representation. This was my very first (and so far ONLY) publication in a peer-reviewed journal. Published in Gestalt Theory. This paper won the Wolfgang Metzger Award for significant contribution to Gestalt theory.

Gestalt Isomorphism and the Primacy of the Subjective Perceptual Experience

A very short peer review commentary in response to a paper by Pessoa et al. But this too was published!

Gestalt Isomorphism and the Primacy of the Subjective Conscious Experience: A Gestalt Bubble Model

[Download pdf file]

The world of visual consciousness appears in the form of solid volumes, bounded by colored surfaces, embedded in a spatial void. The retinal input on which it is based however is two-dimensional. (in the monocular case) Visual processing therefore computes a solution to the inverse-optics problem, i.e. it performs a transformation from the two-dimensional retinal input to the three-dimensional spatial percept. But the inverse-optics problem is underconstrained, and has no unique solution. I propose that perception resolves this fundamental ambiguity by way of a unique kind of emergent field-like computation as suggested by the Gestalt soap bubble analogy. Submitted to Behavioral & Brain Sciences September 1999, semi-rejected March 2000, resubmitted April 2000, never arrived, resubmitted May 2000. semi-reviewed October 27 2000. revise & resubmit February 12 2001. Author's response September 2001.BBS responds March 2002. Author's response June 2002. Accepted! September 2002. [Summary of Whole Review Process]

Harmonic Resonance Theory: an Alternative to the "Neuron Doctrine" Paradigm of Neurocomputation to Address Gestalt properties of perception

A Harmonic Resonance theory is presented as an alternative to the Neuron Doctrine, to account for the holistic global aspects of perception identified by Gestalt theory, including such properties as emergence, reification, and invariance in recognition. I propose that harmonic resonance is the long-sought and elusive computational mechanism behind Gestalt theory. Submitted to Psychological Review July 1999, rejected November 1999 Resubmitted to Behavioral & Brain Sciences September 1999 Ignored

Computational Implications of Gestalt Theory I: A Multi-Level Reciprocal Feedback (MLRF) to Model Emergence and Reification in Visual Processing

Computational Implications of Gestalt Theory II: A Directed Diffusion to Model Collinear Illusory Contour Formation

This is a two-part paper that discusses the computational implications of Gestalt theory. I propose a perceptual modeling approach, i.e. to model the subjective experience of vision rather than the corresponding neurophysiological state. I propose specific computational interactions to account for the Gestalt properties of perception, and to explain the role of feedback in vision with a specific computational model that replicates a number of visual illusory phenomena. Part II of the paper demonstrates how the more subtle second order properties of illusory contour formation can be modeled computationally with a dynamic feedback model, as an alternative to the hard-wired receptive field embodied in the neural network approach. Submitted to Perception & Psychophysics June 1999, rejected October 1999. Resubmitted to Cognitive Psychology November 1999. Due to bureaucratic oversight, each of the two papers were sent to a different reviewer as copies of the same paper. rejected May 2000. Re-wrote as the single paper below.

Computational Implications of Gestalt Theory: The Role of Feedback in Visual Processing

Adapted from part I of the paper above, to explain with an explicit computational model how to resolve the concept of a hierarchical visual representation with the idea of top-down feedback, as an explanation for various illusory phenomena. Submitted to Cognitive Psychology July 2000. rejected November 2000.

The World In Your Head: A Gestalt View of the Mechanism of Conscious Experience

Accepted for publication by Lawrence Erlbaum Associates

This book presents a modern reformulation of Gestalt theory based on the philosophy of indirect realism, i.e. that the world you see around you is not the world itself, but merely an internal replica of the external world generated by perceptual processes in your brain. This reveals the primary function of visual perception to be the generation of a fully spatial virtual-reality internal model of the external world.

The Dimensions of Conscious Experience: A Quantitative Phenomenology

A short paper on the structure of conscious experience and its implications for the nature of consciousness. Submitted to: Journal of Consciousness Studies June 2000, rejected July 2000, appealed July 2000. Back & forth July - December 2000 Rejected April 2001. AUTHOR'S RESPONSE!

The Function of Conscious Experience: An Analogical Paradigm for Perception and Behavior

[Download pdf file]

A short paper on the function of conscious experience as an analogical representation of the external world. Submitted to: Consciousness and Cognition July 2000. Rejected (but maybe not?) January 2001. Resubmitted February 2002. Rejected June 2002. AUTHOR'S RESPONSE!

Directional Harmonic Theory: A Computational Gestalt Model to Account for Illusory Contour and Vertex Formation.

[Download pdf file]

Neural network models have been proposed to account for the formation of collinear illusory contours as seen for example in the Kanizsa figure. There are however a number of illusory grouping effects which involve perceived vertices defined by the intersection of two, three, four, or more illusory contours that meet at the vertex. A neural network approach to this kind of perceptual grouping leads to a combinatorial explosion in the number of required receptive fields. A Directional Harmonic theory is presented to account for all of these diverse grouping effects by way of a single simple mechanism that involves harmonic resonances, or patterns of standing waves in the neural substrate. A single resonance mechanism replaces a whole array of different receptive fields in a computationally equivalent neural network model. I propose that harmonic resonance is the long-sought and elusive computational principle behind the holistic emergent aspects of Gestalt theory. Submitted to Perception August 2001. Revise & Resubmit November 2001. Resubmitted January 2002. Revise & Resubmit April 2002. Resubmitted July 2002. Accepted! September 2002.