My recollections of Hubel and Wiesel and a brief review of functional circuitry in the visual pathway

The first paper of Hubel and Wiesel in The Journal of Physiology in 1959 marked the beginning of an exciting chapter in the history of visual neuroscience. Through a collaboration that lasted 25 years, Hubel and Wiesel described the main response properties of visual cortical neurons, the functional architecture of visual cortex and the role of visual experience in shaping cortical architecture. The work of Hubel and Wiesel transformed the field not only through scientific discovery but also by touching the life and scientific careers of many students. Here, I describe my personal experience as a postdoctoral student with Torsten Wiesel and how this experience influenced my own work.     

Recounting the impact of Hubel and Wiesel

David Hubel and Torsten Wiesel provided a quantum step in our understanding of the visual system. In this commemoration of the 50th year of their initial publication, I would like to examine two aspects of the impact of their work. First, from the viewpoint of those interested in the relation of brain to behaviour, I recount why their initial experiments produced such an immediate impact. Hubel and Wiesel's work appeared against a background of substantial behavioural knowledge about visual perception, a growing desire to know the underlying brain mechanisms for this perception, and an abysmal lack of physiological information about the neurons in visual cortex that might underlie these mechanisms. Their initial results showed both the transformations that occur from one level of processing to the next and how a sequence of these transformations might lead to at least the elements of pattern perception. Their experiments immediately provided a structure for conceptualizing how cortical neurons could be organized to produce perception. A second impact of Hubel and Wiesel's work has been the multiple paths of research they blazed. I comment here on just one of these paths, the analysis of visual cortex in the monkey, particularly in the awake monkey. This direction has led to an explosion in the number of investigations of cortical areas beyond striate cortex and has addressed more complex behavioural questions, but it has evolved from the approach to neuronal processing pioneered by Hubel and Wiesel.     

Hierarchical models of object recognition in cortex.

Visual processing in cortex is classically modeled as a hierarchy of increasingly sophisticated representations, naturally extending the model of simple to complex cells of Hubel and Wiesel. Surprisingly, little quantitative modeling has been done to explore the biological feasibility of this class of models to explain aspects of higher-level visual processing such as object recognition. We describe a new hierarchical model consistent with physiological data from inferotemporal cortex that accounts for this complex visual task and makes testable predictions. The model is based on a MAX-like operation applied to inputs to certain cortical neurons that may have a general role in cortical function.     

Neurons in parafoveal areas V1 and V2 encode vertical and horizontal disparities

Stereoscopic vision mainly relies on binocular horizontal disparity (HD), and its cortical encoding is well established in the foveal representation of the visual field. The role of vertical disparity (VD) is more controversial. Thus far, in the monkey, very few studies have investigated the HD sensitivity beyond 5 degrees of retinal eccentricity and no evidence of a real encoding of VD exists in the parafoveal representation of areas V1 and V2. Using dynamic random dot stereograms, we have tested both HD and VD selectivities in the parafoveal representation of V1 (calcarine V1) and V2 (eccentricities > 10 degrees ) in a behaving monkey. HD and VD selectivities have been characterized using fitting with Gabor function. A large proportion of the tested cells were both HD and VD selective (47%) and, to a lesser extent, HD selective only (8%) or VD selective only (23%). We found a real encoding of VD, with the same diversity in the tuning profiles as described for HD, that cannot be assimilated to a simple perturbation of the HD matching process. Moreover, the VD encoding had a finer scale than the HD one, which is coherent with the smaller range of naturally occurring VD. For the HD encoding, both the percentage of selective cells and the tuning parameters were close to those reported in foveal V1. These results show that, at parafoveal eccentricities in V1 and V2, disparity detectors are tuned to both horizontal and vertical dimensions of the positional disparity existing between matched features in both retinas.     

Role of neurotrophins in the development and plasticity of the visual system: experiments on dark rearing.

An extensive series of studies, beginning with the pioneering experiments of Wiesel and Hubel, have shown that correct visual experience is crucial for the development of the visual system. Several years ago, we put forward the hypothesis that neurotrophic factors of the neurotrophin family (NGF, BDNF, NT-3, NT-4) have a role in mediating the effects of visual experience in the developing visual system. This theory is based on the following experimental results: (a) exogenous supply of neurotrophins during the critical period prevents the effects of monocular deprivation; and (b) transplant of cells releasing NGF allows a normal development of the functional properties of visual cortical neurons in dark-reared rats.     

Retinotopic maps and foveal suppression in the visual cortex of amblyopic adults

Amblyopia is a developmental visual disorder associated with loss of monocular acuity and sensitivity as well as profound alterations in binocular integration. Abnormal connections in visual cortex are known to underlie this loss, but the extent to which these abnormalities are regionally or retinotopically specific has not been fully determined. This functional magnetic resonance imaging (fMRI) study compared the retinotopic maps in visual cortex produced by each individual eye in 19 adults (7 esotropic strabismics, 6 anisometropes and 6 controls). In our standard viewing condition, the non-tested eye viewed a dichoptic homogeneous mid-level grey stimulus, thereby permitting some degree of binocular interaction. Regions-of-interest analysis was performed for extrafoveal V1, extrafoveal V2 and the foveal representation at the occipital pole. In general, the blood oxygenation level-dependent (BOLD) signal was reduced for the amblyopic eye. At the occipital pole, population receptive fields were shifted to represent more parafoveal locations for the amblyopic eye, compared with the fellow eye, in some subjects. Interestingly, occluding the fellow eye caused an expanded foveal representation for the amblyopic eye in one early–onset strabismic subject with binocular suppression, indicating real-time cortical remapping. In addition, a few subjects actually showed increased activity in parietal and temporal cortex when viewing with the amblyopic eye. We conclude that, even in a heterogeneous population, abnormal early visual experience commonly leads to regionally specific cortical adaptations.     

The foundations of development and deprivation in the visual system

The pioneering work of Torsten Wiesel and David Hubel on the development and deprivation of the visual system will be summarised, together with some comments on their influence, and some personal reminiscences by the author.     

The contribution of spike threshold to the dichotomy of cortical simple and complex cells

The existence of two classes of cells, simple and complex, discovered by Hubel and Wiesel in 1962, is one of the fundamental features of cat primary visual cortex. A quantitative measure used to distinguish simple and complex cells is the ratio between modulated and unmodulated components of spike responses to drifting gratings, an index that forms a bimodal distribution. We have found that the modulation ratio, when derived from the subthreshold membrane potential instead of from spike rate, is unimodally distributed, but highly skewed. The distribution of the modulation ratio as derived from spike rate can, in turn, be predicted quantitatively by the nonlinear properties of spike threshold applied to the skewed distribution of the subthreshold modulation ratio. Threshold also increases the spatial segregation of ON and OFF regions of the receptive field, a defining attribute of simple cells. The distinction between simple and complex cells is therefore enhanced by threshold, much like the selectivity for stimulus features such as orientation and direction. In this case, however, a continuous distribution in the spatial organization of synaptic inputs is transformed into two distinct classes of cells.     

In praise of subjective truths

'To determine orientation we occasionally used a PDP-12 computer to produce a graph of average response vs orientation, generating the slit electronically on a television screen. This method took much longer, and the usual minute-to-minute variations in responsiveness of the cell tended to make the curves broader and noisier. We concluded that for both speed and for precision it is hard to beat judgments based on the human ear. Certainly [our curves] could not have been obtained with computer averaging methods before the authors reached the age of mandatory retirement.' ( Hubel & Wiesel, 1974 )     

Three-dimensional orientation tuning in macaque area V4

Tuning for the orientation of elongated, linear image elements (edges, bars, gratings), first discovered by Hubel and Wiesel, is considered a key feature of visual processing in the brain. It has been studied extensively in two dimensions (2D) using frontoparallel stimuli, but in real life most lines, edges and contours are slanted with respect to the viewer. Here we report that neurons in macaque area V4, an intermediate stage in the ventral (object-related) pathway of visual cortex, were tuned for 3D orientation--that is,for specific slants as well as for 2D orientation. The tuning for 3D orientation was consistent across depth position (binocular disparity) and position within the 2D classical receptive field. The existence of 3D orientation signals in the ventral pathway suggests that the brain may use such information to interpret 3D shape.     

Faster scaling of visual neurons in cortical areas relative to subcortical structures in non-human primate brains

Cortical expansion, both in absolute terms and in relation to subcortical structures, is considered a major trend in mammalian brain evolution with important functional implications, given that cortical computations should add complexity and flexibility to information processing. Here, we investigate the numbers of neurons that compose 4 structures in the visual pathway across 11 non-human primate species to determine the scaling relationships that apply to these structures and among them. We find that primary visual cortex, area V1, as well as the superior colliculus (SC) and lateral geniculate nucleus scale in mass faster than they gain neurons. Areas V1 and MT gain neurons proportionately to the entire cerebral cortex, and represent fairly constant proportions of all cortical neurons (36 and 3 %, respectively), while V1 gains neurons much faster than both subcortical structures examined. Larger primate brains therefore have increased ratios of cortical to subcortical neurons involved in processing visual information, as observed in the auditory pathway, but have a constant proportion of cortical neurons dedicated to the primary visual representation, and a fairly constant ratio of about 45 times more neurons in primary visual than in primary auditory cortical areas.     

Parvalbumin immunoreactivity: a reliable marker for the effects of monocular deprivation in the rat visual cortex.

In mammals, monocular deprivation performed during the early stages of postnatal development (critical period) dramatically affects the functional organization of the visual cortex. Since the early work of Hubel and Wiesel, the effects of monocular deprivation are accounted for by the fibers driven by the two eyes competing for the control of cortical territories. In cat and monkey striking structural changes accompany the functional effects of monocular deprivation. Also, in the rat, monocular deprivation causes functional alteration at the level of visual cortex; no structural correlates of these effects, however, have so far been described. Parvalbumin is a calcium binding protein that in the neocortex colocalizes with a subpopulation of GABAergic neurons. Here we report that in the rat monocular deprivation results in a dramatic reduction of parvalbumin-like immunoreactivity in the visual cortex contralateral to the deprived eye. This effect is due to competitive phenomena and not to visual deprivation itself, it is restricted to the binocular portion of the visual cortex and neither binocular deprivation, nor dark rearing can induce it. We conclude that parvalbumin-like immunoreactivity is a useful immunohistochemical marker for the effects of monocular deprivation in the rat visual cortex.     

The relationship between voltage-sensitive dye imaging signals and spiking activity of neural populations in primate V1

Voltage-sensitive dye imaging (VSDI) is a powerful technique for measuring neural population responses from a large cortical region simultaneously with millisecond temporal resolution and columnar spatial resolution. However, the relationship between the average VSDI signal and the average spiking activity of neural populations is largely unknown. To better understand this relationship, we compared visual responses measured from V1 of behaving monkeys using VSDI and single-unit electrophysiology. We found large and systematic differences between position and orientation tuning properties obtained with these two techniques. We then determined that a simple computational model could explain these tuning differences. This model, together with our experimental results, allowed us to estimate the quantitative relationship between the average VSDI signal and local spiking activity. We found that this relationship is similar to the previously reported nonlinear relationship between average membrane potential and spike rate in single V1 neurons, suggesting that VSDI signals are dominated by subthreshold synaptic activity. This model, together with the VSDI measured maps for spatial position (retinotopy) and orientation, also allowed us to estimate the spatial integration area over which neural responses contribute to the VSDI signal at a given location. We found that the VSDI-integration area is consistent with a Gaussian envelope with a space constant of ～230 μm. Finally, we show how this model and estimated parameters can be used to predict the pattern of population responses at the level of spiking activity from VSDI responses.     

Spatial coding of position and orientation in primary visual cortex

We examined the spatial distribution of population activity in primary visual cortex (V1) of tree shrews with optical imaging and electrophysiology. A line stimulus, thinner than the average V1 receptive field, evoked a broad strip of neural activity of nearly constant size for all stimulus locations tested within the central 10 degrees of visual space. Stimuli in adjacent positions activated highly overlapping populations of neurons; nevertheless, small changes in stimulus position produced orderly changes in the location of the peak of the population response. Statistically significant shifts in the population response were found for stimulus displacements an order of magnitude smaller than receptive field width, down to the limit of optical imaging resolution. Based on the pattern of population activity, we conclude that the map of visual space in V1 is orderly at a fine scale and has uniform coverage of position and orientation without local relationships in the mapping of these features.     

Local precision of visuotopic organization in the middle temporal area (MT) of the macaque

Summary The representation of the visual field in the middle temporal area (MT) was examined by recording from single neurons in anesthetized, immobilized macaques. Measurements of receptive field size, variability of receptive field position (scatter) and magnification factor were obtained within the representation of the central 25°. Over at least short distances (less than 3 mm), the visual field representation in MT is surprisingly orderly. Receptive field size increases as a linear function of eccentricity and is about ten times larger than in V1 at all eccentricities. Scatter in receptive field position at any point in the visual field representation is equal to about one-third of the receptive field size at that location, the same relationship that has been found in V1. Magnification factor in MT is only about onefifth that reported in V1 within the central 5° but appears to decline somewhat less steeply than in V1 with increasing eccentricity. Because the smaller magnification factor in MT relative to V1 is complemented by larger receptive field size and scatter, the point-image size (the diameter of the region of cortex activated by a single point in the visual field) is roughly comparable in the two areas. On the basis of these results, as well as on our previous finding that 180° of axis of stimulus motion in MT are represented in about the same amount of tissue as 180° of stimulus orientation in V1, we suggest that a stimulus at one point in the visual field activates at least as many functional modules in MT as in V1.     

Functional alignment of feedback effects from visual cortex to thalamus

Following from the classical work of Hubel and Wiesel, it has been recognized that the orientation and the on- and off-zones of receptive fields of layer 4 simple cells in the visual cortex are linked to the spatial alignment and properties of the cells in the visual thalamus that relay the retinal input. Here we present evidence showing that the orientation and the on- and off-zones of receptive fields of layer 6 simple cells in cat visual cortex that provide feedback to the thalamus are similarly linked to the alignment and properties of the receptive fields of the thalamic cells they contact. However, the pattern of influence linked to on- and off-zones is phase-reversed. This has important functional implications.     

A quantitative analysis of cytochrome oxidase-rich patches in the primary visual cortex of Cebus monkeys: topographic distribution and effects of late monocular enucleation

Summary We have studied the tangential distribution of cytochrome oxidase (cytox)-rich patches in striate cortex of normal and monocularly enucleated Cebus apella monkeys. Patch spatial density and patch cross-sectional area were analysed in cytox-reacted tangential sections of flat-mounted preparations of V1. In the upper cortical layers of V1, and specially in the middle of layer III, the Cebus has well-delimited cytox-rich patches. Rows of patches are less conspicuous in Cebus than in Old World monkeys. The spatial density of patches is nearly constant throughout the binocular field representation in V1, with a mean value of 4 patches per mm². In the monocular portions of V1, however, patch spatial density diminishes. In most cases, mean patch cross-sectional area decreases slightly towards the representation of the periphery in V1. However, patches in the representation of the monocular crescent tend to be larger than those in the adjacent binocular representation. The small variation of cytox patch topography with eccentricity contrasts with the large variation of cortical point-image size in V1. In monocularly enucleated monkeys, patches are larger and darker above and below the ocular dominance stripes of the remaining eye than in the alternate stripes. After long-term enucleation, the patches corresponding to the remaining eye columns appeared larger than in normal controls. In contrast, there is no difference in size between the patches located in the deprived and undeprived monocular crescent representations, although both patch and interpatch regions are darker staining in the latter. These results suggest the existence of competitive interactions which modify the cortical intrinsic organization even in adult monkeys.     

Rapid and precise retinotopic mapping of the visual cortex obtained by voltage-sensitive dye imaging in the behaving monkey

Retinotopy is a fundamental organizing principle of the visual cortex. Over the years, a variety of techniques have been used to examine it. None of these techniques, however, provides a way to rapidly characterize retinotopy, at the submillimeter range, in alert, behaving subjects. Voltage-sensitive dye imaging (VSDI) can be used to monitor neuronal population activity at high spatial and temporal resolutions. Here we present a VSDI protocol for rapid and precise retinotopic mapping in the behaving monkey. Two monkeys performed a fixation task while thin visual stimuli swept periodically at a high speed in one of two possible directions through a small region of visual space. Because visual space is represented systematically across the cortical surface, each moving stimulus produced a traveling wave of activity in the cortex that could be precisely measured with VSDI. The time at which the peak of the traveling wave reached each location in the cortex linked this location with its retinotopic representation. We obtained detailed retinotopic maps from a region of about 1 cm(2) over the dorsal portion of areas V1 and V2. Retinotopy obtained during <4 min of imaging had a spatial precision of 0.11-0.19 mm, was consistent across experiments, and reliably predicted the locations of the response to small localized stimuli. The ability to rapidly obtain precise retinotopic maps in behaving monkeys opens the door for detailed analysis of the relationship between spatiotemporal dynamics of population responses in the visual cortex and perceptually guided behavior.