McGovern Institute Search Committee
McGovern Institute for Brain
Research
Building E19-307
Massachusetts Institute of
Technology
77 Massachusetts Aveneue
Cambridge, MA 02139
Dear Search Committee,
I am writing to apply for the advertised tenure track assistant professorship at the McGovern Institute. If the goal of the McGovern Institute is to fill the gap in understanding from "molecules to mind," my research makes the connection at the mind end of the problem: I have proposed quantitative models of visual experience to specify the information content of experience, as a necessary prerequisite for identifying the neurophysiological mechanism responsible for encoding that information. This approach has led to some interesting and counter-intuitive findings that pose a significant challenge to contemporary neuroscience. In particular, the fact that visual experience appears in the form of volumetric spatial structures provides direct evidence for a volumetric imaging mechanism in the brain capable of generating moving colored images. There is a constructive, or generative aspect of perception, apparent in the Gestalt illusions, as well as in dreams and hallucinations,and in normal perception, which is very difficult to account for in conventional neurophysiological terms.
In fact, there is a profound spatial problem with contemporary concepts of neural representation: no plausible principle has ever been proposed by which the brain might express filled-in volumetric spatial structures or spatial patterns in mental imagery, hallucination, illusion, or in everyday perception. To date, the only mechanism proposed for expressing spatial patterns in the brain is the spatial receptive or projective field, whose patterns of excitatory and inhibitory synapses encode the spatial pattern represented by that neuron. But the spatial receptive field is no different from a template theory, a concept whose limitations are well known, because it leads to a combinatorial explosion in the number of different receptive field templates required to recognize even simple objects at different locations, orientations, and scales. The spatial problem is so serious in contemporary neuroscience that it has inclined many theorists toward a naïve realism whereby they deny that the rich spatial structure of visual experience is a product of the brain at all. They propose that we view the world directly out in the world where it lies, rather than a representation, or replica of that world in an internal representation. But naïve realism is untenable on strict logical grounds, and it offers no explanation for the experience of dreams, hallucinations, and visual illusions, in which spatial structures are experienced which have no objective reality in the external world. (Lehar 2003a) [for references see Proposal for Future Research in Lehar-Proposal.doc]
I have proposed a neurophysiological principle to resolve the spatial problem. I propose that the brain expresses spatial structures by way of harmonic resonance, or spatial standing waves in the brain. (Lehar 2003a, 2003c) Like the neural receptive field, the standing wave defines a spatial pattern of positive and negative regions. Like a Fourier representation, standing waves can, in combination, express virtually any pattern in one, two, and three dimensions. Unlike the neural receptive field, the standing wave is not rigid and fixed, anchored to the tissue of the brain, but elastic and flexible, like an elastic template that morphs to match any distortion in the sensory stimulus. Unlike the neural receptive field, the standing wave does not require spatial templates of every possible shape replicated at every location and every orientation at each location across the visual field, but rather, the standing wave emerges spontaneously in the neural substrate at a location and orientation that best match the stimulus pattern. I have devised a detailed standing wave model to account for a number of illusory perceptual phenomena which are very difficult to account for plausibly in neural network terms. ( Lehar 2003c)
My work is rather controversial, and thus my publication and employment history are not quite at the level that one might expect from an applicant for a professorship at MIT. But my work is either complete and total nonsense, or it is a very significant advance in a fresh new direction addressing questions which have remained unaddressed for many decades. Either the world we see around us is the world itself, and our direct awareness of it must remain forever a deep dark mystery, or the world we see is merely a volumetric "picture" constructed by perceptual processes in our brain. If the latter is the case, then that opens the profound spatial problem: How are volumetric moving colored images generated in the tissue of the brain? Once we acknowledge the reality of the spatial problem in neuroscience, then our first priority must be to identify the neurophysiological principles behind the imaging mechanism that we know to be in the brain. Is there room at MIT for research in a bold new direction? Is that not what MIT is all about?
If I am invited to give a presentation at MIT, it will generally follow the lines of this on-line presentation, which is sure to stimulate active debate.
http://cns-alumni.bu.edu/~slehar/epist/epist.html
I have also provided a summary of my research to date, and a prospectus for future research into both perceptual modeling and into the principles of harmonic resonance for representing spatial patterns in the brain.
As an energetic and creative "out-of-the-box" thinker with some fresh new ideas, and as a teacher with a deep love of teaching, my presence would be an interesting and entertaining addition to the community of unique characters at MIT.
Steve Lehar