Perception and Philosophy of Science

  • Hüseyin Yilmaz
Part of the Boston Studies in the Philosophy of Science book series (BSPS, volume 13)


It is argued that an evolutionary theory of perception and knowledge, and a perceptual philosophy of science is, at present, a realistic and useful way of looking at the whole cognitive endeavor of living beings including man. This theory conceives perceptions, concepts and theories as ways of adapting to the useful regularities of environmental niches, and further extending these niches by exploration. This may lead, at times, to readaptation to less complex niches, etc., but in the case of man the scientific means of observation and exploration seem to lead to ever greater expansion of his environment until, at least mentally, his niche becomes the whole observable universe. Man’s problem of knowledge is thus immensely more complex and also unitary compared to organisms which fill special niches. The task is nevertheless made conceivable because of the few comprehensive, and overriding regularities encountered throughout the universe. In this essay the evolutionary epistemology is assumed, defended and restated and eventually is made the foundation of a partially deductive system from which, with the help of other information of a physical, environmental nature the properties of sensory, perceptual and logico-conceptual organizations are, in essence, inferrable. We hope that this reversal provides further insights into the nature of knowledge and proves to be useful in helping us to create new and better concepts and theories consistent with our own status as products of biological and psycho-social evolution.


Special Relativity Physical Theory Color Vision Lorentz Transformation Color Perception 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


Appendix 1. Nature of a Perceptual Theory

  1. 1.
    H. Yilmaz, ‘Color Vision and New Approach to General Perception’, in E. E. Bernard and M. R. Kare (eds), Biological Prototypes and Synthetic Systems 1, Plenum Press, New York, 1962, pp. 126–41.Google Scholar
  2. 3.
    An example of such exploration via the statistical fluctuations is the process by which a physical system reaches thermal equilibrium, H. B. Callen, Thermodynamics, John Wiley & Sons, Inc., New York, 1963, pp. 267–92.Google Scholar
  3. 4.
    The ancient ‘principle of optimal design’ seems to have been formulated clearly for the first time by N. Rashevski (Mathematical Biophysics, Vol. II, Dover, New York, 1960, p. 292) and applied by him and his students to many biological problems;Google Scholar
  4. 4a.
    R. Rosen, ‘Optimality Principles in Biology,’ Plenum Press, New York and London, 1967. Note that Rashevski implies not only the optimization of the physical configuration of design but also functional behavior of the organ or organism with respect to the environment and other organisms in the environment.Google Scholar

Appendix 2. The Psychophysical Lawl

  1. 1.
    The principle that percepts tend to be invariant against the variations of the environmental situations appears to have been conceived by many people. For example, as early as Seventeenth Century Descartes introduced a similar argument in the perception of shapes. However, a clearer exposition and the idea of combining it with the concept of homological modeling and thereby deriving the psychological power law from them seems to be a recent event; H. Yilmaz, ‘Perceptual Invariance and the Psychophysical Law’, Perception and Psychophysics 2 II (1967), 533–38.CrossRefGoogle Scholar
  2. 2.
    S. S. Stevens, ‘On the Psychophysical Law’, Psychol. Rev. 64 (1957), 153–81. This paper, for the first time, established the power law with solid experimental evidence and replaced the century-old logarithmic law of Fechner. Weber’s law, on the other hand, remains empirically justifiable.CrossRefGoogle Scholar
  3. 3.
    E. H. Land, ‘Color Vision and the Natural Image’, Proc. Nat. Acad. Sci. 45 (1959), 115–29. This and subsequent papers of Land made popular some old but little appreciated experiments in color vision which show that all is not well with classical color theory.CrossRefGoogle Scholar
  4. 4.
    R. Plutchik, ‘The Emotions’, Random House, New York, 1962. This work, apart from being a highly imaginative approach to the problem of emotions, is interesting also from the point of view of methodology. The author does not make use of the analogues of the transformation formulae in color vision. With their inclusion some of the problems in emotional adjustment would seem to be possible to treat in the framework of the theory. In very general terms the perception theory seems to imply that there exists a state of equilibrium of perception which serves as a reference and deviations from this equilibrium are perceived as certain sensations feelings and perceptions. If the space of parameters have isotropy with respect to the equilibrium, as would be expected in most instances, then there are complementarities in perceptions and sensations. If the state of equilibrium is a result of competing processes then there are transformations in the perception etc. This perceptual equilibrium, however, should not be thought of as chemical or physical kind of condition which takes place on its own accord but rather as an active organizational property of the sensory devices which is evolved and structured by the information content of the environment to be perceived.Google Scholar

Appendix 3. Perception of Light and Color

  1. 2.
    For example, the fact that the human eye is also a pattern recognizer implies that a large portion of the channel capacity of the eye has to be devoted to pattern resolutions. Since pattern acuity is very important for the organism, only very few of the channels can be devoted to color. As a matter of fact in the fovea, where pattern resolution is greatest, the color vision happens to be quite deficient; E. N. Wilmer, in Julian Huxley, A. C. Hardy, and E. B. Ford (eds.), Color Vision and its Evolution in the Vertebrates, pp. 306–25, Collier Books, New York, 1963. The necessity of invariance under a given set of illuminants also plays a restricting role upon the number of Fourier terms. For example if the set of natural alluminants are so smooth it can be represented by the first three of these components, then there would be no way of implementing transformations for the higher eigenfunctions. This means they cannot be made inveariant hence not useful.Google Scholar

Appendix 4. Perception of Voice and Musicl

  1. 1.
    The residue pitch was first discovered by J. F. Schouten in 1937 and fascinated scores of researchers and interested laymen under the name of the ’perception of missing fundamentar. Three of the clearest articles on the subject are: J. F. Schouten, R. J. Ritsma, and B. L. Cardozo, J.A.S.A. 34 (1962), 1418.Google Scholar
  2. 1a.
    R. J. Ritsma and B. L. Cardozo, Philips Technical Rev. 25 (1964), 37,Google Scholar
  3. 1b.
    and R. J. Ritsma, J.A.S.A. 34 (1962), 1224.Google Scholar
  4. 2.
    Originally a non-linear mechanism in the ear was thought to explain the residue phenomenon. However, this ran into difficulties for reasons given in R. J. Ritsma and B. L. Cardozo, ibid. A time domain mechanism proposed by E. de Boer ‘Residue in Hearing’, thesis, Amsterdam, 1956, explains some of the gross properties of the phenomenon as indicated in the above references but this too runs into difficulties because of the existence of other phenomena known under the name of the ‘second effect of pitch’.Google Scholar
  5. 3.
    H. Yilmaz, ‘On the Pitch of the Residue’, Technical Report No. 41, I. P. O., Eindhoven, 1964.Google Scholar
  6. 5.
    To obtain the existence region in terms of f we plot against f as follows: If for a given v, the number of harmonics implemented is then for this v, up to the frequency f = n̄v the residue is perceptible. Therefore in (4.10) substitute v = f/n̄and plot the implicit function against f. This plot will give the existence region for a large modulation depth. At lower modulation depths smaller numbers of harmonics are excited and the existence region gets correspondingly narrower. R. J. Ritsma, ‘Existence Region of the Tonal Residue I’, J.A.S.A. 34 (1962), 1224. A striking confirmation of the residue effect is recently found in frequency-inverted speech. If frequency is inverted at, say, 1200 Hz, speech sounds will be inverted but not the pitch. Similarly a whistled melody will be unrecognizably inverted whereas the same tune sung by human voice will be heard as the same melody.Google Scholar

Appendix 5. Theory of Space and Time

  1. 2.
    The wave-particle duality implicit in the photoelectric (photon) theory of Einstein was extended by Louis de Broglie in 1924 to material particles, especially to electron, and thereby modern quantum theory was initiated. The rigorous relation between the orthogonality of rays and wave-fronts and the relativity of simultaneity also shows that Potier’s 1874 hypothesis, despite the denial of first order effects of ether drag could hardly be considered as a forerunner of special relativity since it does not take into account the full reciprocity of reference frames. M. A. Potier, J. Phys. (Paris), 3 (1874), 201. See also R. G. Newburgh and T. E. Phipps, Jr. (submitted to A.J.P.). It could, however, be considered as a forerunner of the Michelson-Morley experiment, in first order. More important, Potier implicitly regarded the Fresnel drag as a time transformation instead of a velocity effect in the space part. This was real advance in the direction of modern theory of relativity.Google Scholar
  2. 4.
    The idea that space and time concepts should arise from the existence of material elements goes back to Ernst Mach. His ideas, however, were along the line of global requirements which eventually gave rise to the so-called ‘Mach’s principle’ much discussed in general relativity and cosmology. Our intention is strictly on the local side as will be evident a little further. On a level more elementary and geometrical the perceptual character of special relativity was argued by D. Bohm, Special Theory of Relativity, Benjamin, New York-Amsterdam, 1962 (see Appendix, p. 185) and the present writer, Introduction to the Theory of Relativity, Blaisdell, New York, 1965 (see Introduction, p. xiii).Google Scholar
  3. 8.
    M. Born, Atomic Physics, Hefner Publishing Co., 1936, p. 94.Google Scholar

Appendix 6. Statistical Theory of Fields

  1. 2.
    This conclusion about the kinematical background of fields and particles including the reinterpretation of special relativity implicit in Equations (6.3) and (6.5) was reached as early as 1963 in a report ‘On the Theory of S-Matrix’, on the basis of the Equation (5.8). The publication of the consequent Statistical Theory of Fields, however, was delayed several years, mainly because of the unconventional nature of the presentation containing perceptual ingredients. Eventually it was possible to find a presentation of the theory along the lines of familiar formulations of quantum field theory; H. Yilmaz, ‘Statistical Theory of Fields’ in H. Feschbach and U. Ingard (eds.), In Honor of Philip M. Morse, MIT Press, Cambridge, 1969, pp. 170–80.Google Scholar
  2. 3.
    Usually this point is not clearly discussed in the exposition of variational techniques. See, however, H. Yilmaz, Introduction to the Theory of Relativity, Blaisdell Publishing Co., New York, 1965, p. 48.Google Scholar
  3. 4.
    L. Pauling and E. B. Wilson, Introduction to Quantum Mechanics, McGraw-Hill Publishing Co., New York, 1935, p. 3996. We apply the principle here in the context of the quantum theory of fields.Google Scholar
  4. 5.
    J. M. Jauch, and R. Rohrlich, Theory of Photons and Electrons, Addison-Weslley Publ. Co., Inc., Reading, Mass. 1955, pp. 22–23.Google Scholar
  5. 7.
    R. Rohrlich, Classical Charged Particles, Addison-Weslley Publ. Co., Inc., Reading, Mass. 1965, p. 238. Professor Rohrlich seems to have recognized clearly the deficiency of the conventional action principles but he attributes the difficulty to the operator character of the fields solely. In our view, this is in fact part of the problem but it can be overcome if the kinematical preliminaries are understood and separated properly.Google Scholar
  6. 9.
    The reason simply is that ζ = 0 leads to equal probabilities for all φ|1k|0(x) which then violates the requirement (6.7). Without such a requirement, on the other hand, integrals of field theory are not well defined (L. Streit, in Paul Urban (ed), Quamtum Electrodynamics, Springer Verlag, Inc., New York, 1965, p. 14.) For similar reasons theories with indefinite metric are not satisfactory although they may ameliorate some of the divergences. It seems that a theory with indefinite metric would lead at most to an asymptotic behavior ~ Δ (k)/k 2 hence logarithmically divergent charge renormalization although the mass and the vertex renormalizations are finite. In such a theory there are further problems associated with causality and zero-point energy which is also divergent.Google Scholar
  7. 10.
    N. G. van Kampen, in E. G. D. Cohen (ed.), Fundamental Problems in Statistical Mechanics, North-Holland Publ. Co., Amsterdam, 1962, pp. 173–202.Google Scholar

Appendix 7. The Problem of the Unity of Physics

  1. 1.
    Nowadays the Sagnac effect is used as a laser gyro to detect inertial motion W. M. Macek and T. M. Davis, Jr., Appl. Phys. Letters 2 (1963), 67.CrossRefGoogle Scholar
  2. 2.
    Michelson-Morley experiment was recently repeated with great accuracy with laser techniques by A. Arna, T. S. Jasea and C. H. Townes. See the summary article by A. L. Schawlow, Scientific American 209 (1963), 2.CrossRefGoogle Scholar
  3. 3.
    P. Langevin, Comp. Rend. 173 (1921) 831.Google Scholar
  4. 5.
    See for example W. H. Louisell, Radiation and Noise in Quantum Electronics for similar arguments on distributions not belonging to conserved quantities, McGraw-Hill Inc., 1964, pp. 221–252.Google Scholar
  5. 6.
    In elementary particle physics a linear harmonic oscillator equation was first introduced by Born in relation to his reciprocity principle, M. Born, Rev. Mod. Phys. 21 (1949), 463. Its extension by Equations (7.15–18) and their subsequent solution leading to the factorised Veneziano amplitudes was presented by the author. Since the validity of reciprocity as a general law of nature is quite doubtful it is probably more desirable to base it on other grounds such as the one we have indicated here. The possible connection between a linear harmonic oscillator and the Veneziano amplitude was also suggested by Suskind who conceived it along the more conventional lines of quark-antiquark interactionsCrossRefGoogle Scholar
  6. 6a.
    (L. Suskind, Phys. Rev. Letters. 23 (1969), 545). Venziano, Fubini and Nambu (see note 9, Appendix 7) arrived at the form of V from a study of the problem of factorization of the Veneziano amplitudes, whereas our intent has been to determine K 0 and V from a reasonable theoretical basis and then to look for solutions that reproduce the amplitudes in a rigorous or in a sufficiently accurate fashion.CrossRefGoogle Scholar
  7. 7.
    H. Lipkin, Groups for Pedestrians, Wiley, 1965, pp. 57–68.Google Scholar
  8. 9.
    S. Fubini and G. Veneziano, Duality in Operator Formalism, MIT preprint, CTP #114, 1969; Y. Nambu, University of Chicago, preprint, 1969.Google Scholar

Appendix 8. Nature of a Physical Theory

  1. 1.
    R. Blanche, Axiomatics, The Free Press, New York, 1962, pp. 56–64.Google Scholar
  2. 4.
    At this point comparing again the faculties of perception and theories of physics we find no essential difference between them, in principle. Both can be described as a modeling activity and both are conditioned with requirements of ever greater efficiency, economy, significance and scope. They can both be understood as part of the overall adaptation that living beings try to make to the realities of the world. In this conception, the distinctions made on the basis of what is learned what is inherited should not matter. What is inherited should be considered as something learned in the past by the species. As far as we can see there is no real obstacle in conceiving man’s cognitive development as a single process of adaptations. However, we must emphasize that there are areas where further efforts are needed for clarification. For example it seems to us that language behavior and linguistics as a transitional area between perception and science, and between language and logic need careful consideration. Linguistics seems to occupy a central position between natural sciences and the sciences of man (humanities). It seems clear that a detailed study of linguistics, both as a perception and as a science, would contribute decisively toward a unitary conception of all science: See the penetrating analysis by R. Jakobson, ‘Linguistics in Its Relation to Other Sciences’, Actes du X e Congres International des Linguistes, Editions Bucarest, 1969, pp. 75–122. In this connection we would like to mention two articles by the present author on the subject of speech perception which, in our opinion, treats some of the more elementary aspects of human speech in a form closely consistent with these goals3 H. Yilmaz, ‘A Theory of Speech Perception, I-II’, Bulletin of Mathematical Biophysics 29 (1967), 793–825, and 30 (1968), 455–79. Further work in the direction of a statistical theory of speech perception is now in progress and will be reported separately at a future occasion. This theory explains on a purely deductive manner the results obtained byCrossRefGoogle Scholar
  3. 4a.
    Li House and Hughes in JASA 46 (1969), 1019 and has similarities to the statistical theory of fields presented in Appendix 6. A theory of Psycho-social evolution of the author (unpublished, 1963) analyses the essentially Lamarckian-Spencerian process of cultural transmission in the light of language processes and identifies the essentially selection retention nature of this evolution in view of the unbiased, active mental powers of man and his manipulative, tool-making characteristics. A science of man, leading to essentially all the distinctive human characteristic are thereby inferable at least in an outline form. In all these activities we are in close resonance with Michael Polanyi (to appear in this series of essays) andGoogle Scholar
  4. 4b.
    K. R. Popper (Logic of Scientific Discovery, Basic Books, N.Y., 1963;Google Scholar
  5. 4c.
    K. R. Popper Conjectures and Refutations, Basic Books, N.Y., 1963;Google Scholar
  6. 4d.
    K. R. Popper Of Clouds and Clocks, Arthur Holly Compton Memorial Lecture, Washington University, 1965) although there are also noticeable differences, especially in our greater willingness to found physical theory firmly on cognitive and perceptual propositions of a tractable nature. With regard to social status of science we find ourselves in essential agreement with the writing of D. T. Campbell (Objectivity and Social Locus of Scientific Knowledge, to appear in this series of essays) and others with some variations and extensions.Google Scholar

Appendix 9. A Theory of Psychosocial Evolution

  1. 1.
    J. S. Huxley, Evolution in Action, Harper, N.Y., 1953.Google Scholar
  2. 2.
    T. Dobzhansky, Mankind Evolving, Bantam Books, 1962.Google Scholar
  3. 3.
    H. Yilmaz, Categorization and Multistability, Unpublished, 1964.Google Scholar
  4. 4.
    J. S. Huxley, Categorization and Multistability, Unpublished, 1964. ibid. Google Scholar
  5. 5.
    We believe this kind of parallel was epitomized most eloquently by Karl R. Popper in his Conjectures and Refutations, Routledge and Kegan Paul, London, 1963, as the most and the only intelligent behavior in the face of the unknown. It is interesting to note that it is essentially the process of acquiring information by trial and error, and continual revisions in view of new information similarly gathered.Google Scholar

Copyright information

© D. Reidel Publishing Company, Dordrecht, Holland 1974

Authors and Affiliations

  • Hüseyin Yilmaz
    • 1
  1. 1.Perception Technology CorporationWinchesterUSA

Personalised recommendations