Hillary R. Rodman

Single-unit analysis of pattern-motion selective properties in the middle temporal visual area (MT)

SummaryThe middle temporal visual area (MT) in macaque extrastriate cortex is characterized by a high proportion of neurons selective for the direction of stimulus motion, and is thus thought to play an important role in motion perception. Previous studies identified a population of cells in MT that appeared capable of coding the motion of whole visual patterns independent of the motions of contours within them (Gizzi et al. 1983; Movshon et al. 1985). These “pattern-motion selective” neurons are unlike motion sensitive cells that have been observed at earlier stages of the visual system. Using very different criteria, we have also previously indentified an apparently functionally distinct group of MT neurons (Albright 1984). We predicted that these “Type II” neurons correspond to the pattern-motion neurons. In the present study, we have applied both sets of criteria to individual neurons in MT and found that these two differently defined sets of cells actually form the same population. These results support the idea that MT contributes to a specialized type of motion processing which reflects the integrity of normal perception.

Coding of visual stimulus velocity in area MT of the macaque

We have studied the interaction of the direction and speed selectivities of neurons in cortical visual area MT of the macaque monkey. For a given cell, preferred direction and the shape of the direction tuning curve for moving edges were similar at different stimulus speeds, and deviations from the optimal speed did not systematically alter direction tuning bandwidth. Similar speed tuning was obtained for responses to motion in the preferred and anti-preferred directions even when the response to anti-preferred motion was an inhibitory one. The results are discussed in terms of the unique contributions of area MT to visual motion analysis.

Pattern of retinal projections in the California ground squirrel (Spermophilus beecheyi): Anterograde tracing study using cholera toxin

The retinofugal pathways in the California ground squirrel, Spermophilus beecheyi, were mapped after intravitreal injections of cholera toxin B‐subunit. The results of the current study are consistent with work in other mammals and provide new details relevant to the organization and evolution of the visual system. All retinorecipient nuclei received bilateral input, with a contralateral predominance. The suprachiasmatic nucleus is heavily innervated, and sparse terminals were noted in other hypothalamic areas. In addition to the interstitial, medial, lateral, and dorsal terminal nuclei, a few fibers of the accessory optic tract may enter the ventral lateral geniculate and the nucleus of the optic tract, though this innervation may not derive from the same ganglion cells innervating the accessory optic nuclei. Retinal terminals are found in the intergeniculate leaflet and the “dorsal cap” of the ventral lateral geniculate. Retinal fibers pass rostrally from the dorsal cap toward the anterodorsal thalamus, confirming a projection described in the tree shrew and monkeys. Retinal termination patterns in the dorsal lateral geniculate reveal a hexilaminate organization of alternating ipsilateral and contralateral input. Variations in terminal morphology suggest that sublayers receive input from distinct ganglion cell types and that laminar comparisons can be made with primates. Finally, terminal patterns in the superior colliculus reveal a dense, highly ordered columnar organization supporting functional properties of tectal receptive fields. All the visual structures in the ground squirrel are large and well differentiated, making the sciurid visual system an accessible rodent model for comparing visual processing with that in other diurnal vertebrates. J. Comp. Neurol. 463:317–340, 2003.

Calbindin immunoreactivity in the geniculo-extrastriate system of the macaque: Implications for heterogeneity in the koniocellular pathway and recovery from cortical damage

Although most projection neurons in the primate dorsal lateral geniculate nucleus (dLGN) target striate cortex (V1), a small number project instead to extrastriate visual areas and have been suggested to play a role in the preserved vision (“blindsight”) that survives damage to V1. Moreover, the distribution of dLGN cells projecting to extrastriate bears a striking similarity to that of neurons that stain for calbindin D‐28K (Cal), a calcium‐binding protein involved in regulating neuronal excitability and considered a marker for the koniocellular or “K” pathway of geniculocortical processing. In these studies, we used double‐labeling techniques to examine whether Cal content characterizes all or a subset of neurons making up the geniculo‐extrastriate pathway in normal macaque monkeys. After injections of cholera toxin B‐subunit into the prelunate gyrus, the proportion of retrogradely labeled neurons in the dLGN that were also immunoreactive for Cal varied from less than 40% to over 80%, indicating that only a subset of the geniculo‐extrastriate projection falls within the K pathway as defined by Cal content. Analysis of the injected territories indicated that identity of the extrastriate cortical target may be systematically related to Cal content in the geniculo‐extrastriate projection. To see whether the Cal‐immunoreactive dLGN population might potentially play a role in preserved vision after V1 damage, we also examined the dLGN of a macaque that had sustained a lesion of V1 in infancy and survived until 4 years. In this animal, large, intensely Cal‐immunoreactive neurons were found scattered throughout the otherwise degenerated dLGN zones and made up over 95% of the identifiable remaining neurons. The results support an emerging view that the macaque koniocellular system is highly heterogeneous in nature and also suggest that Cal content may be a critical feature of the pathway by which visual information reaches extrastriate cortex in the absence of V1. J. Comp. Neurol. 431:168–181, 2001.

A transient geniculo-extrastriate pathway in macaques? Implications for 'blindsight'.

Both monkeys and cats receiving primary visual cortex lesions in infancy show better residual vision than animals sustaining similar damage in adulthood. In cats, the better recovery has been explained in part by stabilization of a transient pathway from the dorsal lateral geniculate nucleus (dLGN) to cortical visual area PMLS. To test the hypothesis that a similar transient pathway from the dLGN to dorsal extrastriate areas exists in primates and thus serves as a candidate for recruitment after early V1 damage, retrograde tracers were injected into areas MT, MST, and/or 7a of infant macaques. No evidence of a transient pathway from the dLGN to these areas was obtained, despite projections from the pulvinar and other extrastriate areas in all cases.

The effects of combined superior temporal polysensory area and frontal eye field lesions on eye movements in the macaque monkey

We previously found [42] that lesions of the superior temporal polysensory area (STP) cause temporary deficits in the production of eye movements. In order to both define regions participating in the ensuing recovery and to further explore the cortical control of eye movements, we examined the effects of addition of frontal eye field (FEF) lesions to STP lesions, on visual fixation, saccadic eye movements, and smooth pursuit eye movements. Three monkeys received bilateral STP lesions followed by a FEF lesion and as a control, an additional monkey received a bilateral inferior temporal cortex (IT) lesion followed by a FEF lesion. All animals had a profound impairment in foveating the central fixation point. This impairment was completely eliminated by turning on a dim light in the testing chamber. Large neglect-like impairments in making saccades were only seen after combined STP and FEF lesions. Impairments in making smooth pursuit eye movements after combined lesions of STP and FEF were larger than those seen after STP lesions but within the range of deficits that have been reported after FEF lesions alone. The impairment of visual fixation in darkness and the lack of impairment under conditions of dim illumination appear to reflect a specific role for the FEF in spatial orientation in the absence of visual landmarks. The FEF also appears to play a more critical role than STP in smooth pursuit. By contrast, STP and the FEF appear to work cooperatively with respect to the production of saccades. We suggest that cortical oculomotor control can flow either through the midbrain or through the FEF and that the FEF pathway is specifically involved in tasks with a discontiguity between the stimuli and the behavioral response while the midbrain pathways are preferentially involved in more stimulus-driven eye movements.

Development and Plasticity of Extrastriate Visual Cortex in Monkeys

The extrastriate visual cortex of the monkey has in recent decades come to include an ever-expanding portion of the neocortical domain as more and more traditional “association” or even motor territories are shown to have significant visual connections, visual responsiveness, or role in visual behavior (Felleman and van Essen, 1991; also see the chapter by Gross in this volume). In this chapter, discussion will be restricted (or broadened, depending upon one’s viewpoint) to consideration of cortical zones shown to have at least some visual sensory responsiveness and direct connectivity with other, unimodal visual areas. First, we will consider the normal anatomical, physiological, and metabolic development of extrastriate visual cortex, including the prenatal specification of visual areas. Next, we will discuss the plasticity of extrastriate visual cortex in adulthood by examining the ability of extrastriate areas to function in parallel with striate cortex and the consequences of damage to extrastriate cortex in adult monkeys. In addition, we will examine evidence for learning-or experience-dependent plasticity in the response properties of neurons in extrastriate cortex. In the subsequent section, we will address the special plasticity associated with damage to visual cortex in developing animals. We will then briefly compare the development and plasticity of extratriate cortex in monkeys with phenomena described for other mammalian groups. In the final section, we will summarize the data presented and comment on general principles of extrastriate and cerebral cortical development.

Methods for repeated recording in visual cortex of anesthetized and awake behaving infant monkeys

This report describes methods for making repeated recordings from visual cortex in infant macaque monkeys (age range 5 weeks to 7 months) both under nitrous oxide anesthesia and in an alert behaving state. These methods permitted successful collection of single-unit data from inferior temporal cortex and other areas in a longitudinal fashion from individual infants. For the anesthetized experiments, modifications of our standard techniques for surgical preparation and maintenance of the animal in an anesthetized, artificially ventilated state permitted repeated recording sessions while assuring the animals continued health and normal growth. Following implantation with an eye coil, monkeys as young as 5 weeks could be trained to perform a fixation task during alert recording sessions which were likewise performed on a repeated basis with maintained health and development. Further, comparison of results from anesthetized and awake recording sessions indicates that alert recording is preferable for collecting certain types of data from visual cortex in infant monkeys.

The lateral geniculate nucleus does not project to area TE in infant or adult macaques

We sought to determine if there are any direct projections from the dorsal lateral geniculate nucleus (dLGN) to visual cortical area TE in either adult or infant primates. To do so, we examined labelling in the thalamus of eight macaque monkeys which received injections of the retrograde tracer cholera toxin-B subunit within TE. Four of these cases were infants in which the injections revealed transient patterns of inputs to TE from various other brain regions. Although each monkey showed extensive label in the pulvinar nucleus and other subcortical structures, none showed unambiguous labelling in the dLGN. The absence of direct connections between the dLGN and area TE indicates that rudimentary color and form processing capacities in the absence of striate cortex must utilize other pathways even when damage to striate cortex takes place early in life.

Development of Brain Substrates for Pattern Recognition in Primates: Physiological and Connectional Studies of Inferior Temporal Cortex in Infant Monkeys

Inferior temporal (IT) cortex is known to be critical for visual pattern perception and recognition in adult monkeys, but not in infant ones. We found that cells in IT cortex ofinfant monkeys showed adult-like response properties, including form selectivity for complex patterns such as faces, as early as the second month of life. However, the neural responses were weaker and of longer latency, and moreover were virtually absent under anesthesia below about four months of age. Anatomical inputs from visual cortical regions were adult like early in infancy, whereas connectivity patterns with non-visual cortical regions appeared to have a more protracted period of maturation.