Articles from the past 4 years, with abstracts
Olshausen BA, Field DJ (2005) How close are we to understanding V1? Neural Computation, in press [PDF]A wide variety of papers have reviewed what is known about the function of primary visual cortex. In this article, rather than stating what is known, we attempt to estimate how much is still unknown about V1 function. In particular, we identify five problems with the current view of V1 that stem largely from experimental and theoretical biases, in addition to the contributions of non-linearities in the cortex that are not well understood. Our purpose is to open the door to new theories, a number of which we describe along with some proposals for testing them.
Olshausen BA, Field DJ (2004) Sparse Coding of Sensory Inputs Current Opinion in Neurobiology, 14: 481-487 [PDF]
Several theoretical, computational, and experimental studies suggest that neurons encode sensory information using a small number of active neurons at any given point in time. This strategy, referred to as 'sparse coding', could possibly confer several advantages. First, it allows for increased storage capacity in associative memories; second, it makes the structure in natural signals explicit; third, it represents complex data in a way that is easier to read out at subsequent levels of processing; and fourth, it saves energy. Recent physiological recordings from sensory neurons have indicated that sparse coding could be a ubiquitous strategy employed in several different modalities across different organisms.
Olshausen BA, Field DJ (2004) What is the other 85% of V1 doing? In: Problems in Systems Neuroscience. T.J. Sejnowski, L. van Hemmen, Ed$ [PDF]
This article will pose the following challenge: that despite four decades of research characterizing the response properties of V1 neurons, we still do not have a decent picture of how V1 really operates - i.e., how a population of its neurons represents natural scenes under realistic viewing conditions. We identify five problems with the current view that stem largely from biases in the design and execution of experiments, in addition to the contributions of non-linearities in the cortex that are not well understood. Our purpose is to open the window to new theories, a number of which we describe along with some proposals for testing them.
Field, D. and Hayes, A. (2004) "Contour Integration and the Lateral Connections of V1 Neurons"In The Visual Neurosciences, Eds. L M Chalupa and J S Werner, MIT Press.
Olshausen BA, Field DJ (2004) What is the other 85% of V1 doing? In: Problems in Systems Neuroscience. T.J. Sejnowski, L. van Hemmen, Eds. Oxford University Press. [PDF]
This chapter will pose the following challenge: that despite four decades of research characterizing the response properties of V1 neurons, we still do not have a decent picture of how V1 rea lly operatesi.e., how a population of its neurons represents natural scenes under realistic viewing conditions. We identify five problems with the current view that stem largely from biases in the design and execution of experiments, in addition to the contributions of non-linearities in the cortex that are not well understood. Our purpose is to open the window to new theories, a number of which we describe along with some proposals for testing them
Webster, Michael A., John S. Werner, and David J. Field. Adaptation and the Phenomenology of Perception. To appear in Fitting the Mind to the World: Adaptation and Aftereffects in High Level Vision: Advances in Visual Cognition Series, Volume 2, C. Clifford and G. Rhodes (Eds.) Oxford University Press. [DOC] [HTML]
To what extent do we have shared or unique perceptual experiences? We examine how the answer to this question is constrained by the processes of visual adaptation. Adaptation constantly recal ibrates visual coding so that our vision is normalized according to the stimuli that we are currently exposed to. These normalizations occur over very wide ranging time scales, from milliseconds to evolutionary spans. The resulting adjustments dramaticall y alter the appearance of the world before us, and in particular alter visual salience by highlighting how the current image deviates from the properties predicted by the current states of adaptation. To the extent that observers are exposed to and thus a dapted by a different environment, their vision will be normalized in different ways and their subjective visual experience will differ. These differences are illustrated by considering how adaptation influences properties which vary across different envi ronments. To the extent that observers are exposed and adapted to common properties in the environment, their vision will be adjusted toward common states, and in this respect they will have a common visual experience. This is illustrated by considering t he effects of adaptation on image properties that are common across environments. In either case, it is the similarities or differences in the stimuli and not the intrinsic similarities or differences in the observers which largely determine the rel ative states of adaptation. Thus at least some aspects of our private internal experience are controlled by external factors that are accessible to objective measuremen
Hess RF, Hayes A, Field DJ (2003) Contour integration and cortical processing. J Physiol Paris. Mar-May;97(2-3):105-19.
Our understanding of visual processing in general, and contour integration in particular, has undergone great ch ange over the last 10 years. There is now an accumulation of psychophysical and neurophysiological evidence that the outputs of cells with conjoint orientation preference and spatial position are integrated in the process of explication of rudimentary con tours. Recent neuroanatomical and neurophysiological results suggest that this process takes place at the cortical level V1. The code for contour integration may be a temporal one in that it may only manifest itself in the latter part of the spike train a s a result of feedback and lateral interactions. Here we review some of the properties of contour integration from a psychophysical perspective and we speculate on their underlying neurophysiological substrate.
Kingdom, F. A. A., Hayes, A. & Field, D. J. (2001) Sensitivity to contrast histogram differences in synthetic wavelet-textures. Vision Res. 41, 585-598 [PDF]
Recent research on texture synthesis suggests that characterisation of those properties of textures to which human observers are sensitive may be provided by the histograms of the coeffic ients of a wavelet decomposition. In this study we examined the properties of wavelet histograms that affect texture discrimination by measuring observer sensitivity to differences in the wavelet histograms of synthetic textures. The textures, generated v ia Gabor micropattern synthesis, were broadband, with amplitude spectra that are characteristic of natural images, i.e. 1:f. We measured texture-difference thresholds for three moments of the wavelet histograms variance, skew and kurtosis by manipul ating the contrast, phase, and density, of the Gabor elements used to construct the textures. Observers discriminated more efficiently between textures that had differences in kurtosis, than between textures that had differences in either variance or skew . Performance was compared to two model observers; one used the pixel-luminance histogram, the other used the histogram of the output of wavelet-filters. The results support the idea that the visual system is relatively sensitive to the kurtosis, or 4th m oment, of the wavelet histogram of textures. We argue that higher than 4th-order moments will, in practice, become increasingly difficult for the visual system to represent because the lack of a perfect match between the elements and the receptive fields effectively blurs the response histogram, thereby attenuating higher moment
Brady, N. and Field, D. J. (2000) Local contrast in natural images: Normalisation and coding efficiency. Perception, 29, 1-15.
The visual system employs a gain control mechanism in the cortical coding of contrast whereby the response of each cell is normalised by the integrated activity of neighbouring cells. While restricted in space, the normalisation pool is broadly tuned for spatial frequency and orientation, so that a cell's response is adapted by stimuli which fall outside its ' classical' receptive field. Various functions have been attributed to divisive gain control: in this paper we consider whether this output nonlinearity serves to increase the information carrying capacity of the neural code. 46 natural scenes were analyse d with the use of oriented, frequency-tuned filters whose bandwidths were chosen to match those of mammalian striate cortical cells. The images were logarithmically transformed so that the filters responded to a luminance ratio or contrast. In the first s tudy, the response of each filter was calibrated relative to its response to a grating stimulus, and local image contrast was expressed in terms of the familiar Michelson metric. We found that the distribution of contrasts in natural images is highly kurt otic, peaking at low values and having a long exponential tail. There is considerable variability in local contrast, both within and between images. In the second study we compared the distribution of response activity before and after implementing contra st normalisation, and noted two major changes. Response variability, both within and between scenes, is reduced by normalisation, and the entropy of the response distribution is increased after normalisation, indicating a more efficient transfer of inform ation.
Field, D.J. (2000) "Matched Filters, wavelets, and the statistics of natural scenes." Proceedings of the First International Workshop on Matched Filtering. St. Petersburg, Russia. Journal of Optical Technology
Olshausen, B.A. and Field, D.J (2000) "Vision and the Coding of Natural Images" American Scientist, 88, 238-24
Although the images we take in with our eyes are highly varied, most scenes share certain statistical properties. The eyes and brain take advantage of this similarity to encode images as they are sensed by the photoreceptors in the retina and passed on to higher levels of the visual system. In this way, the neural networks of the visual system function much like the image-compression algorithms devised for telecommunications. The authors explain this correspondence and describe their work to demonstrate the self-organizing principle behind this phenomenon.
Field, D. J., Hayes, A, and Hess, R.F. (2000) "The role of polarity and symmetry in the perceptual grouping of contour fragments", Spatial Vision, 13, 51-66.
We describe two experiments that investigate the roles of polarity and symmetry in the perceptual grouping of contour fragments. Observers viewed, for one second on each presentation, arrays of oriented, spatial-frequency band-pass, elements, in which a subset of the elements was aligned along a twisting curve. In each of five conditions we measured observers' ability to detect aligned combinations of even- and odd-symmetric elements, of the same and different polarities, against a background of 'noise' elements. As with previous experiments we found that the 'path' could be reliably detected, even when the elements of the path were oriented at angles of up to +/- 60 deg relative to each other. Detection of the path was still possible when the polarity of path elements alternated. However, the probability of detection of the path was raised significantly when the path elements were all of the same polarity. Perceptual grouping of even-symmetric elements was no different to perceptual grouping of odd-symmetric elements. The results provide evidence, that in achieving integration of contour fragments, the visual system uses a process that is to some degree phase selective. We use the results to describe how the visual system may resolve natural contours when they occur against backgrounds that vary over a wide range of intensities. The data presented here have been published in conference-abstract form (Hayes et al., 1993; Field et al., 1997).