Introduction It has been suggested that visual representations are optimised to transmit the maximum information about the images encountered in everyday life (Uttley, 1973; Linsker, 1988; Barlow, 1989). This simple assumption has proven sufficient to account for the characteristics of large monopolar cells in the fly (Srinivasan et al., 1982; Hateren, 1992; Laughlin, 1981), the temporal characteristics of retinal ganglion cells (Dong & Atick, 1995), human spatial frequency thresholds (Atick & Redlich, 1992; Van Hateren, 1993), and the psychophysics of orientation perception for short presentation times (Baddeley & Hancock, 1991). Maximisation of information is a powerful theoretical principle that leads to testable predictions about the firing patterns of neurons. However, to generate specific predictions we must make some assumptions about the nature of the neural code and the type of constraint that limits its information carrying capacity. To appl

Neurophysiological evidence is described, showing that some neurons in the macaque temporal cortical visual areas have responses that are invariant with respect to the position, size and view of faces and objects, and that these neurons show rapid processing and rapid learning. A theory is then described of how such invariant representations may be produced in a hierarchically organized set of visual cortical areas with convergent connectivity. The theory proposes that neurons in these visual areas use a modified Hebb synaptic modification rule with a short-term memory trace to capture whatever can be captured at each stage that is invariant about objects as the object changes in retinal position, size, rotation and view. Simulations are then described which explore the operation of the architecture. The simulations show that such a processing system can build invariant representations of objects.

. A computational model of hippocampal activity during spatial cognition and navigation tasks is presented. The spatial representation in our model of the rat hippocampus is built on-line during exploration via two processing streams. An allothetic vision-based representation is built by unsupervised Hebbian learning extracting spatio-temporal properties of the environment from visual input. An idiothetic representation is learned based on internal movement-related information provided by path integration. On the level of the hippocampus, allothetic and idiothetic representations are integrated to yield a stable representation of the environment by a population of localized overlapping CA3-CA1 place fields. The hippocampal spatial representation is used as a basis for goal-oriented spatial behavior. We focus on the neural pathway connecting the hippocampus to the nucleus accumbens. Place cells drive a population of locomotor action neurons in the nucleus accumbens. Reward-based learnin...

... In this article, we show how trace learning could solve the problem of in-depth rotation-invariant object recognition by developing representations of the transforms that features undergo when they are on the surfaces of 3D objects. Moreover, we show that having learned how features on 3D objects transform geometrically as the object is rotated in depth, the network can correctly recognize novel 3D variations within a generic view of an object composed of a new combination of previously learned features. These results are demonstrated in simulations of a hierarchical network model (VisNet) of the visual system that show that it can develop representations useful for the recognition of 3D objects by forming perspective-invariant representations to allow generalization within a generic view.

& Experiments investigating the mechanisms involved in visual processing often fail to separate low-level encoding mechanisms from higher-level behaviorally relevant ones. Using an alternating dual-task event-related potential (ERP) experimental paradigm (animals or vehicles categorization) where targets of one task are intermixed among distractors of the other, we show that visual categorization of a natural scene involves different mechanisms with different time courses: a perceptual, task-independent mechanism, followed by a taskrelated, category-independent process. Although average ERP responses reflect the visual category of the stimulus shortly after visual processing has begun (e.g. 75±80 msec), this difference is not correlated with the subject's behavior until 150 msec poststimulus. &

Diverse theories of animal navigation aim at explaining how to determine and maintain a course from one place to another in the environment, although each presents a particular perspective with its own terminologies. These vocabularies sometimes overlap, but unfortunately with different meanings. This paper attempts to precisely define the existing concepts and terminologies, so as to comprehensively describe the different theories and models within the same unifying framework. We present navigation strategies within a 4 level hierarchical framework based upon levels of complexity of required processing (Guidance, Place recognition-triggered Response, Topological navigation, Metric navigation). This classification is based upon what information is perceived, represented and processed. It contrasts with common distinctions based upon availability of certain sensors or cues and rather stresses the information structure and content of central processors. We then review computat...

Psychophysical findings accumulated over the past several decades indicate that perceptual tasks such as similarity judgment tend to be performed on a low-dimensional representation of the sensory data. Low dimensionality is especially important for learning, as the number of examples required for attaining a given level of performance grows exponentially with the dimensionality of the underlying representation space. In this chapter, we argue that, whereas many perceptual problems are tractable precisely because their intrinsic dimensionality is low, the raw dimensionality of the sensory data is normally high, and must be reduced by a nontrivial computational process, which, in itself, may involve learning. Following a survey of computational techniques for dimensionality reduction, we show that it is possible to learn a low-dimensional representation that captures the intrinsic low-dimensional nature of certain classes of visual objects, thereby facilitating further learning of tasks...

Abstract not found

. It has been proposed that the receptive fields of neurons in V1 are optimised to generate "sparse", Kurtotic, or "interesting" output probability distributions (Barlow & Tolhurst, 1992; Barlow, 1994; Field, 1994; Intrator & Cooper, 1991; Intrator, 1992). We investigate the empirical evidence for this further and argue that filters can produce "interesting" output distributions simply because natural images have variable local intensity variance. If the proposed filters have zero D.C., then the probability distribution of filter outputs (and hence the output Kurtosis) is well predicted simply from these effects of variable local variance. This suggests that finding filters with high output Kurtosis does not necessarily signal interesting image structure. It is then argued that finding filters that maximise output Kurtosis generates filters that are incompatible with observed physiology. In particular the optimal difference--of--Gaussian (DOG) filter should have the smallest possible s...

n Backward masking can potentially provide evidence of the time needed for visual processing, a fundamental constraint that must be incorporated into computational models of vision. Although backward masking has been extensively used psychophysically, there is little direct evidence for the effects of visual masking on neuronal responses. To investigate the effects of a backward masking paradigm on the responses of neurons in the temporal visual cortex, we have shown that the response of the neurons is interrupted by the mask. Under conditions when humans can just identify the stimulus, with stimulus onset asynchronies (SOA) of 20 msec, neurons in macaques respond to their best stimulus for approximately 30 msec. We now quantify the information that is available from the responses of single neurons under backward masking conditions when two to six faces were shown. We show that the information available is greatly decreased as the mask is brought closer to the stimulus. The decrease is more marked than the decrease in #ring rate because it is the selective part of the #ring that is especially attenuated by the mask, not the spontaneous #ring, and also because the neuronal response is more variable at short SOAs. However, even at the shortest SOA of 20 msec, the information available is on average 0.1 bits. This compares to 0.3 bits with only the 16-msec target stimulus shown and a typical value for such neurons of 0.4 to 0.5 bits with a 500msec stimulus. The results thus show that considerable information is available from neuronal responses even under backward masking conditions that allow the neurons to have their main response in 30 msec. This provides evidence for how rapid the processing of visual information is in a cortical area and provides a fundamental constra...