Superior colliculus: some receptive field properties of bimodally responsive cells

Circuits for Action and Cognition: A View from the Superior Colliculus

The superior colliculus is one of the most well-studied structures in the brain, and with each new report, its proposed role in behavior seems to increase in complexity. Forty years of evidence show that the colliculus is critical for reorienting an organism toward objects of interest. In monkeys, this involves saccadic eye movements. Recent work in the monkey colliculus and in the homologous optic tectum of the bird extends our understanding of the role of the colliculus in higher mental functions, such as attention and decision making. In this review, we highlight some of these recent results, as well as those capitalizing on circuit-based methodologies using transgenic mice models, to understand the contribution of the colliculus to attention and decision making. The wealth of information we have about the colliculus, together with new tools, provides a unique opportunity to obtain a detailed accounting of the neurons, circuits, and computations that underlie complex behavior.

Human speech- and reading-related genes display partially overlapping expression patterns in the marmoset brain

Language is a characteristic feature of human communication. Several familial language impairments have been identified, and candidate genes for language impairments already isolated. Studies comparing expression patterns of these genes in human brain are necessary to further understanding of these genes. However, it is difficult to examine gene expression in human brain. In this study, we used a non-human primate (common marmoset; Callithrix jacchus) as a biological model of the human brain to investigate expression patterns of human speech- and reading-related genes. Expression patterns of speech disorder- (FoxP2, FoxP1, CNTNAP2, and CMIP) and dyslexia- (ROBO1, DCDC2, and KIAA0319) related genes were analyzed. We found the genes displayed overlapping expression patterns in the ocular, auditory, and motor systems. Our results enhance understanding of the molecular mechanisms underlying language impairments.

Alignment of sound localization cues in the nucleus of the brachium of the inferior colliculus

Accurate sound localization is based on three acoustic cues (interaural time and intensity difference and spectral cues from directional filtering by the pinna). In natural listening conditions, every spatial position of a sound source provides a unique combination of these three cues in "natural alignment." Although neurons in the central nucleus (ICC) of the inferior colliculus (IC) are sensitive to multiple cues, they do not favor their natural spatial alignment. We tested for sensitivity to cue alignment in the nucleus of the brachium of the IC (BIN) in unanesthetized marmoset monkeys. The BIN receives its predominant auditory input from ICC and projects to the topographic auditory space map in the superior colliculus. Sound localization cues measured in each monkey were used to synthesize broadband stimuli with aligned and misaligned cues; spike responses to these stimuli were recorded in the BIN. We computed mutual information (MI) between the set of spike rates and the stimuli containing either aligned or misaligned cues. The results can be summarized as follows: 1) BIN neurons encode more information about auditory space when cues are aligned compared with misaligned. 2) Significantly more units prefer aligned cues in the BIN than in ICC. 3) An additive model based on summing the responses to stimuli with the localization cues varying individually accurately predicts the alignment preference with all cues varying. Overall, the results suggest that the BIN is the first site in the ascending mammalian auditory system that is tuned to natural combinations of sound localization cues.

GABAergic and glutamatergic identities of developing midbrain Pitx2 neurons

Pitx2, a paired-like homeodomain transcription factor, is expressed in post-mitotic neurons within highly restricted domains of the embryonic mouse brain. Previous reports identified critical roles for PITX2 in histogenesis of the hypothalamus and midbrain, but the cellular identities of PITX2-positive neurons in these regions were not fully explored. This study characterizes Pitx2 expression with respect to midbrain transcription factor and neurotransmitter phenotypes in mid-to-late mouse gestation. In the dorsal midbrain, we identified Pitx2-positive neurons in the stratum griseum intermedium (SGI) as GABAergic and observed a requirement for PITX2 in GABAergic differentiation. We also identified two Pitx2-positive neuronal populations in the ventral midbrain, the red nucleus, and a ventromedial population, both of which contain glutamatergic precursors. Our data suggest that PITX2 is present in regionally restricted subpopulations of midbrain neurons and may have unique functions that promote GABAergic and glutamatergic differentiation.

Organization and origin of the connection from the inferior to the superior colliculi in the rat

The inferior colliculus (IC) is the main ascending auditory relay station prior to the superior colliculus (SC). The morphology and origin of the connection from inferior to superior colliculus (I-SC) was analyzed both by anterograde and retrograde tracing. Irrespective of the subregion of the IC in which they originate, the terminal fields of these connections formed two main tiers in the SC. While the dorsal one primarily involved the stratum opticum and the stratum griseum intermediale, the ventral one innervated the deep strata, although some fibers did connect these tiers. While the dorsal tier occupied almost the whole extension of the SC, the ventral one was mostly confined to its caudomedial quadrant. The fiber density in these tiers decreased gradually in a rostral gradient and the terminal fields became denser as the anterograde tracer at the injection site was distributed more externally in the cortex of the IC. Retrograde tracing confirmed this result, although it did not reveal any topographic ordering for the I-SC pathway. Most presynaptic boutons of the I-SC terminal field were located either inside or close to the patches of acetylcholinesterase activity. Together with previous anatomical and physiological studies, our results indicate that the I-SC connection relays behaviorally relevant information for sensory-motor processing. Our observation that this pathway terminates in regions of the superior colliculus, where neurons involved in fear-like responses are located, reinforce previous suggestions of a role for the IC in generating motor stereotypes that occur during audiogenic seizures.

Multisensory contributions to the perception of motion

The ability to process motion is crucial for coherent perception and action. While the majority of studies have focused on the unimodal factors that influence motion perception (see, for example, the other chapters in this Special Issue), some researchers have also investigated the extent to which information presented in one sensory modality can affect the perception of motion for stimuli presented in another modality. Although early studies often gave rise to mixed results, the development of increasingly sophisticated psychophysical paradigms are now enabling researchers to determine the spatiotemporal constraints on multisensory interactions in the perception of motion. Recent findings indicate that these interactions stand over-and-above the multisensory interactions documented previously for static stimuli, such as the oft-cited 'ventriloquism' effect. Neuroimaging and neuropsychological studies are also beginning to elucidate the network of neural structures responsible for the processing of motion information in the different sensory modalities, an important first step that will ultimately lead to the determination of the neural substrates underlying these multisensory contributions to motion perception.

Dynamics of audio-visual interactions in the guinea pig brain: an electrophysiological study

Audio-visual interactions and their specifications, evaluated by bioelectrical activities, in guinea pigs are presented in this study. The difference potential, as the evidence of an interaction, was calculated by subtracting the sum of averaged potentials recorded in visual and auditory events from the averaged potential recorded in an event where two stimuli combined in the same sweep. Dynamic investigations have shown an interaction when auditory stimulus is applied 24 ms before and 201 ms after visual stimulation. Latency between the difference potential and auditory stimulus was stable. Directional investigations have shown that the interaction is not observed when auditory and/or visual stimulation is used ipsilaterally, according to the recording side.

Rapid adaptation to auditory-visual spatial disparity

The so-called ventriloquism aftereffect is a remarkable example of rapid adaptative changes in spatial localization caused by visual stimuli. After exposure to a consistent spatial disparity of auditory and visual stimuli, localization of sound sources is systematically shifted to correct for the deviation of the sound from visual positions during the previous adaptation period. In the present study, this aftereffect was induced by presenting, within 17 min, 1800 repetitive noise or pure-tone bursts in combination with synchronized, and 20 degrees disparate flashing light spots, in total darkness. Post-adaptive sound localization, measured by a method of manual pointing, was significantly shifted 2.4 degrees (noise), 3.1 degrees (1 kHz tones), or 5.8 degrees (4 kHz tones) compared with the pre-adaptation condition. There was no transfer across frequencies; that is, shifts in localization were insignificant when the frequencies used for adaptation and the post-adaptation localization test were different. It is hypothesized that these aftereffects may rely on shifts in neural representations of auditory space with respect to those of visual space, induced by intersensory spatial disparity, and may thus reflect a phenomenon of neural short-term plasticity.

Spatial characteristics of visual-auditory summation in human saccades

Bimodal (auditory + visual) stimuli reduce saccade latencies in human observers to a degree that exceeds levels predictable by probabilistic summation between parallel, independent unimodal pathways. These interactions have been interpreted in terms of converging visual and auditory afferents within the oculomotor pathways, specifically within the superior colliculus (SC). The present work describes the spatial tuning of auditory-visual summation in human saccades, using diagnostics derived from stochastic models of information processing. Consistent with expectations based on the electrophysiology of the SC, the magnitude of facilitation varied with the degree of spatial correspondence, and the spatial tuning was quite coarse.

Spatial-tuning properties of auditory neurons in the optic tectum of the pigeon

We studied the auditory neurons in the optic tectum of the unanesthetized pigeon, using single-unit recordings and acoustic free-field stimulation. Most units showed spatial tuning, with best areas located in the contralateral hemifield. All units responded also to visual stimuli, the auditory best areas being in rough alignment with visual receptive fields.

Auditory motion induces directionally dependent receptive field shifts in inferior colliculus neurons

This research focused on the response of neurons in the inferior colliculus of the unanesthetized mustached bat, Pteronotus parnelli, to apparent auditory motion. We produced the apparent motion stimulus by broadcasting pure-tone bursts sequentially from an array of loudspeakers along horizontal, vertical, or oblique trajectories in the frontal hemifield. Motion direction had an effect on the response of 65% of the units sampled. In these cells, motion in opposite directions produced shifts in receptive field locations, differences in response magnitude, or a combination of the two effects. Receptive fields typically were shifted opposite the direction of motion (i.e., units showed a greater response to moving sounds entering the receptive field than exiting) and shifts were obtained to horizontal, vertical, and oblique motion orientations. Response latency also shifted as a function of motion direction, and stimulus locations eliciting greater spike counts also exhibited the shortest neural latency. Motion crossing the receptive field boundaries appeared to be both necessary and sufficient to produce receptive field shifts. Decreasing the silent interval between successive stimuli in the apparent motion sequence increased both the probability of obtaining a directional effect and the magnitude of receptive field shifts. We suggest that the observed directional effects might be explained by "spatial masking," where the response of auditory neurons after stimulation from particularly effective locations in space would be diminished. The shift in auditory receptive fields would be expected to shift the perceived location of a moving sound and may explain shifts in localization of moving sources observed in psychophysical studies. Shifts in perceived target location caused by auditory motion might be exploited by auditory predators such as Pteronotus in a predictive tracking strategy to capture moving insect prey.

Interlaminar connections of the superior colliculus in the tree shrew. III: The optic layer

These experiments were designed to test the idea that the optic layer in the tree shrew, Tupaia belangeri, is functionally distinct and provides a link between the visuosensory superficial and the premotor intermediate layers of the superior colliculus. First, cells in the optic layer were intracellularly labeled with biocytin in living brain slices. Compared to cells in the adjacent lower part of the superficial gray layer, which have apical dendrites that ascend toward the tectal surface, optic layer cells have dendritic fields that are restricted for the most part to the optic layer itself. The differences in dendritic-field location imply that superficial gray and optic layer cells have different patterns of input. The axons of optic layer cells terminate densely within the optic layer and, in addition, project in a horizontally restricted fashion to the overlying superficial gray and subjacent intermediate gray layers. This pattern also is different from the predominantly descending interlaminar projections of lower superficial gray layer cells. Next, cells in the intermediate gray layer were labeled in order to examine the relationships between optic layer cells and these subjacent neurons that project from the superior colliculus to oculomotor centers of the brain stem. Neurons in the upper part of the intermediate gray layer send apical dendrites into the optic layer and therefore can receive signals from the superficial gray layer either directly, from descending axons of lower superficial gray layer cells, or indirectly, through intervening optic layer cells. In contrast, lower intermediate gray layer cells have more radiate dendritic fields that are restricted to the intermediate gray layer. Thus, these lower cells must depend on descending projections from optic or upper intermediate gray layer cells for signals from the superficial gray layer. Together, these results support the idea that the optic layer is a distinct lamina that provides a link between the superficial and intermediate gray layers. They also are consistent with the traditional view that descending intracollicular projections play a role in the selection of visual targets for saccades.

Visual, somatosensory, auditory and nociceptive modality properties in the feline suprageniculate nucleus

Response properties of 252 single-units to visual, auditory, somatosensory and noxious stimulation were recorded by means of extracellular microelectrodes in the suprageniculate nucleus of anaesthetized, immobilized cats. Of the 141 units tested for modality properties the majority (n=113, 80.1%) was found unimodal in the sense that stimuli of exclusively one sensory modality were able to elicit an activation of the unit. Twenty-four (17.0%) cells were bimodal and four (2.8%) were trimodal (visual, somatosensory and auditory). The visual modality dominated the unimodal cells (n=74, 65.5%), while cells responsive to somatic stimulation (n=20, 17.6%), auditory stimulation (n=16, 14.1%) or noxious stimulation of the tooth pulp (n=3, 2.6%) were less frequently encountered. Visual sensitivity dominated the multisensory cells, too. The visually responsive units were characterized by having a sensitivity to stimuli moving in a rather large, uniform receptive field that covered the contralateral lower quadrant, and encompassed a flanking area of about 20 degree width in both the upper contralateral and lower ipsilateral visual fields. Many cells (n=52, 47%) were sensitive to the direction of the stimulation and reacted to stimuli moving at a high velocity (20-200 deg/s). Most cells responded differently to stimuli of a variety of sizes. Somatosensory units reacted to stimuli presented over a wide area on the contralateral side of the body, thus showing no sign of somatotopic organization. The auditory sensitivity fell within a wide range of acoustic stimuli in extremely large auditory receptive fields. The physiological properties of suprageniculate nucleus cells strongly resemble the sensory properties of cells found along the ventral bank of the anterior ectosylvian sulcus and the deeper layers of the superior colliculus. Our results provide further support for the notion of a separate tecto-suprageniculate-anterior ectosylvian sulcus/insular pathway that takes part in the processing of multimodal signals important for various types of sensory related behaviours.

The effect of eye position on auditory lateralization

The present study examines whether the direction of gaze can influence sound lateralization. For this purpose, dichotic stimuli with variable interaural level difference (ILD) were presented under different conditions of visual fixation. In experiment 1, subjects with their head fixed directed their gaze to a given target, simultaneously adjusting the ILD of continuous pure tone or noise stimuli so that their location was perceived in the median plane of the head. The auditory adjustments were significantly correlated with gaze direction. During eccentric fixation, the psychophysical adjustments to the median plane shifted slightly toward the direction of gaze. The magnitude of the shift was about 1-3 dB, over a range of fixation angles of 45 degrees to either side. The eye position effect, measured as a function of pure-tone frequency, was most pronounced at 2 kHz and showed a tendency to decrease at lower and higher frequencies. The effect still occurred, although weaker, even when the eyes were directed to eccentric positions in darkness and without a fixation target. In experiment 2, the adjustment method was replaced by a two-alternative forced-choice method. Subjects judged whether sound bursts, presented with variable ILDs, were perceived on the left or right of the median plane during fixation of targets in various directions. Corresponding to experiment 1, the psychometric functions shifted significantly with gaze direction. However, the shift was only about half as large as that found in experiment 1. The shift of the subjective auditory median plane in the direction of eccentric gaze, observed in both experiments, indicates that dichotic sound is localized slightly to the opposite side, i.e., to the left when the gaze is directed to the right and vice versa. The effect may be related to auditory neurons which exhibit spatially selective receptive fields that shift with eye position.

Ventriloquist effect reinstates responsiveness to auditory stimuli in the 'ignored' space in patients with hemispatial neglect

We examined 6 patients with robust visual neglect following right hemisphere damage. All of them had signs of auditory neglect as documented by the inferior identification of syllables delivered through a loudspeaker on the left side. When the same stimuli on the left were administered in the presence of a fictitious source of sound (a dummy loudspeaker) visible in the homolesional space, a significant increase in the identification score of sounds was obtained (the "ventriloquist" effect). The result is in keeping with a notion of a strong coupling between auditory and visual systems. The effect is attributed to the activation by the fictitious source of sound of the audio-visual map in the left hemisphere. We draw attention to the possibility that loss of awareness of auditory input may arise due to the disconnection of the visual input from the audio-visual template.

Organization of the colliculo-suprageniculate pathway in the cat: a wheat germ agglutinin-horseradish peroxidase study

A wheat germ-agglutinated horseradish peroxidase (WGA-HRP) tracing technique was used to label the cell bodies of neurons in the superior colliculus that send projections into the visually sensitive region of the suprageniculate nucleus (Sg) in the feline thalamus. After determination of the position of the Sg by detecting characteristic single-unit responses to moving visual stimuli, WGA-HRP was injected into the Sg in five pentobarbital-anesthetized cats. The animals were than sacrificed, and serial frozen sections of the midbrain were processed for the demonstration of peroxidase activity. A total of 2,736 WGA-HRP-stained neurons were located within the ipsilateral superior colliculus (SC), and a few labeled cells were consistently found bilaterally in the external nuclei of the inferior colliculus. In each cat, a small but significant fraction of the labeled cells were encountered contralateral to the injection. Medial SC neurons tended to project to the posterior Sg, and lateral SC neurons tended to project to more rostral Sg. However, labeled cells were distributed homogeneously along the rostrocaudal extent of the SC, indicating the absence of a well defined topographic relationship. Nor was the Sg injection site location related to the laminar distribution of SC projection neurons. In all cases, the majority of the labeled cells were found in layer IV (49.0%), with fewer cells in layers III (17.5%) layer V (20.0%), and layer VI (11.8%). No labeled cells were located in layer I, although a few were located in the deep part of layer II. Five types of SC projection cells were distinguished morphologically. Of the 258 labeled cells that could be characterized, 25% were stellate cells, 25% vertical cells, 20% granular cells, 17% triangular cells, and 12% horizontal cells. The average diameter of 226 cells ranged between 8 and 47 microns. We conclude that a mixed population of SC cells projects to the Sg; the morphological heterogeneity and the distribution of these cells suggests that several functionally different pathways may be involved in the colliculothalamic pathway and in the processing of visual input in the SC.

Interlaminar connections of the superior colliculus in the tree shrew. I. The superficial gray layer

One of the most persistent problems in the study of the superior colliculus is the relationship between its superficial and deep layers. The superficial tier of layers is considered to be visuosensory in function, whereas the deep tier is multisensory and has premotor functions. This fundamental distinction is the primary basis for the view that a visually triggered shift in the direction of gaze depends on the transfer of information from sensory cells in the superficial tier to premotor cells in the deep tier. The goal of the present experiments was to examine the interlaminar projections of the superficial gray layer in the tree shrew Tupaia belangeri. We used biocytin as the marker for tracing the pathways. The tree shrew was chosen because its large and distinctly laminated superior colliculus facilitates the task of examining connections between the layers. Biocytin was used because of its sensitivity and because it allowed us to place very small injections restricted entirely to the superficial gray layer. The results demonstrated that a prominent pathway originates in the superficial gray layer and terminates in stratum opticum. In comparison, the projection from the superficial gray layer to the layers beneath stratum opticum is extremely sparse. The pathway from the superficial gray layer to stratum opticum has a columnar distribution, extending about 100 microns rostrally and caudally from the center of the injection site. There were no signs of more remote intracollicular connections, nor of patches or bands of terminals. The biocytin injection sites also labeled pathways to nuclei as distant from the superior colliculus as the diencephalon, including the dorsal and ventral lateral geniculate bodies, and the pulvinar. The results suggest that stratum opticum may serve as a link between the superficial gray layer and the deeper layers.

Calbindin-like immunoreactivity in the central auditory system of the mustached bat, Pteronotus parnelli

With the aid of a polyclonal antibody specific for Calbindin D-28k, we studied the distribution of this calcium-binding protein in the central auditory system of the mustached bat, Pteronotus parnelli. Components of the cochlear nucleus (CN) that were calbindin-positive (cabp(+] included the root of the auditory nerve, multipolar and globular bushy cells in the anteroventral CN, multipolar and octopus cells in the posteroventral CN, and small and medium-size cells in the dorsal CN. Not stained were spherical bushy cells of the anteroventral CN and pyramidal/fusiform cells in the dorsal CN. In the superior olivary complex, labeled cells were found in the lateral and medial nuclei of the trapezoid body, the ventral and ventromedial periolivary nuclei, and the anterolateral periolivary nucleus. No cellular labeling was seen in the lateral superior olive. In the medial superior olive, only marginal cells were cabp(+). Labeled fibers could be seen surrounding the gosts of unlabeled cells in both the latter nuclei. Most cells in the intermediate nucleus and the columnar division of the ventral nucleus of the lateral lemniscus were cabp(+). However, the dorsal nucleus was cabp(-). A group of cabp(+) cells was also seen in the paralemniscal zone. The inferior colliculus had a relatively low density of cabp(+) cells. Labeled cells were more common in the caudal half of the central nucleus, and in the external nucleus and dorsal cortex. In the auditory thalamus, nearly every cell in the medial geniculate body was cabp(+), but those in the suprageniculate nucleus and in the posterior group did not stain. Small cells in the intermediate layer and giant cells in the deep layers of the superior colliculus were densely cabp(+). In the pons, cabp(+) cells and neuropil could be seen in the medial and lateral pontine nuclei (pontine gray). In conclusion, calbindin-like immunoreactivity was found in most of the brainstem auditory system, as well as in regions associated with acoustic orientation or control of vocalization. However, except for a minority of cells of the medial superior olive, it is conspicuously absent from the nuclei receiving binaural input below the level of the inferior colliculus.

Retinotopic organization within the lateral posterior complex of the cat

Electrophysiological mapping methods were employed to systematically study the retinotopic organization within the cat's lateral posterior complex (LP). Visual responses were recorded in all the major subdivisions of the LP as well as in several adjoining cell groups. Specifically, separate representations of the visual field were identified for pulvinar, zones LP1-c, LP1-r, LPi, and LPm. Partial representations of the visual field were also evident in the geniculate wing, subdivisions of the lateral posterior shell, the inferior division of the posterior nuclear group, the suprageniculate nucleus, and the central lateral nucleus. Sufficient mapping observations were made to define the internal organization of major visual representations. Additionally, there was a very close correspondence between the mapping observations when they were compared with the cytoarchitectural criteria for recognizing functional cell groups (Updyke: J. Comp. Neurol. 219:143-181, '83).

Auditory and visual neurons in the cat's superior colliculus selective for the direction of apparent motion stimuli

In the cat, cells of the superior colliculus (SC) and the tectofugal pathways of the visual system are highly selective for the direction of a moving visual stimulus. Deep layer units of SC in addition respond to auditory and somatosensory stimuli, but the proportion of such non-visual cells is usually found to be much lower than that of visual cells. We recorded the responses of 174 cells in the SC to sequentially presented, localized visual and/or auditory stimuli that produced the sensation of apparent motion to human observers. Controls using single LED flashes or tone pips or clicks at very long intervals that did not produce apparent motion were also used. We found both visual and auditory units that responded vigorously to the apparent motion stimuli and showed pronounced directional selectivity. However, in the auditory domain such units were rare and thus did not increase the proportion of auditory responses in SC substantially. Varying the interstimulus interval (ISI) of these stimuli, both visual and auditory, indicated that the mechanism of direction selectivity in these cells was suppression of the response in the 'non-preferred' direction rather than facilitation in the 'preferred' direction. With long ISI's of 200 ms or more, every single stimulus gave a discrete response peak of constant amplitude. For ISI's of 50 ms or less the discrete peaks merged to a continuous response. Maximal firing rate in the preferred direction remained the same as for longer ISI's, but was decreased for movement in the non-preferred directions. Very short ISI's (10 ms) produced little response in any direction.(ABSTRACT TRUNCATED AT 250 WORDS)

Distribution of frequency sensitivity in the superior colliculus of the guinea pig

A field potential could be recorded in the deep layers of the superior colliculus (SCd) in the guinea pig following presentation of pure tone bursts at frequencies between 1 and 22 kHz, bursts of white noise, and clicks presented in the free field. The potentials evoked by these stimuli at different intensities had two negative and one positive component associated with the onset of the stimulus. The amplitude of the negative components of the potential varied according to the location of the recording electrode along the dorso-ventral and rostro-caudal axes of the SC and with the frequency of pure tone stimuli. The largest amplitude potentials were recorded in the deep grey layer of the SCd midway along the rostro-caudal axis of the nucleus. The distribution of frequency sensitivity along the rostro-caudal axis of the SCd was determined by recording the amplitude of the negative components of the field potential evoked by 10 ms pure tone bursts at a constant intensity. The distribution of frequency sensitivity across the nucleus was not cochleotopic. The caudal pole of the nucleus had largest responses to frequencies above 15 kHz, with a peak sensitivity around 20 kHz. In contrast, the rostral pole of the nucleus was most sensitive to frequencies around 10 kHz and the sensitivity to frequencies around 20 kHz was relatively low. At a point midway along the rostro-caudal axis of the SCd, the nucleus was sensitive to a broad range of frequencies from 5 to 25 kHz. The patterns of frequency sensitivity recorded in the rostral, middle and caudal SCd are qualitatively similar to the frequency transfer characteristics of the auditory periphery for sounds located in the anterior, orthogonal and posterior regions respectively of contralateral space. The correspondence between these two sets of data suggests that the pattern of frequency sensitivity along the SCd may provide a mechanism by which the nervous system can encode the spectral cues which are generated at the auditory periphery.

Perioral somatosensory but not visual inputs to the flank of the mouse superior colliculus

An electrophysiological and anatomical study identified the sensory inputs to the "flank" of the mouse's superior colliculus, a large, ventrolateral extension of layer IV (stratum griseum intermediate) that has no overlying visual layers (II and III). Electrophysiological recordings with subsequent histological localization showed that the flank receives predominantly somatosensory projections from the perioral region but not visual input. In its caudal parts, the flank also has limb and trunk somatosensory inputs and auditory inputs. The perioral somatosensory projections to the flank are ordered somatotopically. The flank is considered part of the superior colliculus since the perioral inputs are adjacent to inputs from mystacial vibrissae in the more medial parts of the superior colliculus, hence forming a single continuous map. The finding that no visual responses occur above the flank (due to the absence of superficial layers) or within it is in accord with the concept of an intermodality "spatial register". Since flank neurons have somatosensory receptive fields in non-visible parts of the body and they lack visual responses, the flank may be involved more in tactile-dependent rather than in vision-dependent orienting behaviors. Thus, the superior colliculus may, in parallel, carry out sensorimotor transformations related to (1) shift of gaze and (2) tactile-dependent behaviors not involving vision.

The interrelationship between superior colliculus and substantia nigra pars reticulata in programming movements of cats

Recent electrophysiological and behavioural studies have suggested that the deeper layers of the superior colliculus (dl-SC) are involved in the execution of targeting movements which are elicited but not continuously guided by external stimuli. In the first part of the present study the role of the GABAergic transmission in the dl-SC in the execution of these targeting movements was investigated. Therefore cats, trained to walk from one side of a narrow bar towards the target at the other side of the bar under stroboscopic illumination (2 flashes/s), were injected bilaterally in the dl-SC with picrotoxin (100 ng/0.5 microliter) or muscimol (75 ng/1 microliter). In the second part of this study the role of the GABAergic activity in the substantia nigra pars reticulata (SNR), projecting to the dl-SC through the GABAergic nigrotectal pathway, in the execution of the above-mentioned targeting movements was investigated. Therefore cats, trained to walk from one side of a narrow bar towards the target at the other side of the bar, either under full light or under stroboscopic illumination (2 flashes/s), were bilaterally injected in the SNR with picrotoxin (500 ng/0.5 microliter) or muscimol (200 ng/1 microliter). Under stroboscopic illumination (2 flashes/s) solvent-treated cats either continuously grasped the bar and/or continuously touched the bar with their whiskers, i.e. they executed movements which were continuously guided by external stimuli, while walking towards the target at the other side of the bar.(ABSTRACT TRUNCATED AT 250 WORDS)

Complementary and non-matching afferent compartments in the cat's superior colliculus: innervation of the acetylcholinesterase-poor domain of the intermediate gray layer

Three tectal afferent-fiber systems were experimentally labeled in the cat to learn how their distributions within the superior colliculus were related to the prominent compartments of high acetylcholinesterase activity found in the intermediate gray layer. Presumptive somatic sensory afferents were labeled by injections of horseradish peroxidase-wheatgerm agglutinin conjugate placed at the bulbospinal junction and in the ventral anterior ectosylvian cortex corresponding to somatic sensory area SIV. Vision-related afferents were labeled by injections of the same tracer substance into the lateral suprasylvian visual area. In each animal, a single type of injection was made and a detailed study was carried out to compare the patterns of anterograde labeling and acetylcholinesterase staining in serially adjoining sections through the superior colliculus. Fibers labeled by the three types of injection were distributed in clusters that resembled the acetylcholinesterase-positive patches in the intermediate gray layer. In no case, however, were the afferent-fiber clusters in register with the histochemically defined patches. Instead, the innervations derived from the bulbospinal junction, anterior estosylvian sulcus and lateral suprasylvian visual area all formed patchworks within the acetylcholinesterase-poor domain of the intermediate gray layer. In some instances, the afferent-fiber clusters and enzyme-positive patches appeared to have complementary distributions. In other instances, the afferent-fiber clusters seemed to be arranged in the acetylcholinesterase-poor parts of the intermediate layer in a fashion independent of, but not significantly overlapping, the acetylcholinesterase-positive patches. Not all of the space between the acetylcholinesterase-positive patches was taken up by any one of the afferent-fiber systems labeled. The complementary and non-matching distribution of these afferent systems in relation to the acetylcholinesterase-rich patches of the intermediate gray layer stands in contrast to the spatial registration of two other tectal afferent systems with the zones of high acetylcholinesterase activity. Both nigrotectal and frontotectal afferents converge on the acetylcholinesterase-positive patches. We conclude that afferent systems projecting to the intermediate gray layer can be divided into at least two groups: those innervating the acetylcholinesterase-rich compartments and those avoiding them.(ABSTRACT TRUNCATED AT 400 WORDS)

Integration of visual and auditory information in bimodal neurones in the guinea-pig superior colliculus

We have investigated the responses of neurones in the guinea-pig superior colliculus to combinations of visual and auditory stimuli. When these stimuli were presented separately, some of these neurones responded only to one modality, others to both and a few neurones reliably to neither. To bimodal stimulation, many of these neurones exhibited some form of cross-modality interaction, the degree and nature of which depended on the relative timing and location of the two stimuli. Facilitatory and inhibitory interactions were observed and, occasionally, both effects were found in the same neurone at different inter-stimulus intervals. Neurones whose responses to visual stimuli were enhanced by an auditory stimulus were found in the superficial layers. Although visual-enhanced and visual-depressed auditory neurones were found throughout the deep layers, the majority of them were recorded in the stratum griseum profundum. Neurones that responded to both visual and auditory stimuli presented separately and gave enhanced or depressed responses to bimodal stimulation were found throughout the deep layers, but were concentrated in the stratum griseum intermediale and extended into the stratum opticum.

Convergence of afferents from frontal cortex and substantia nigra onto acetylcholinesterase-rich patches of the cat's superior colliculus

The patterns of distribution of frontotectal and nigrotectal fibers were studied with the anterograde horseradish peroxidase method in the cat. Direct serial-section comparisons were made between the afferent-fiber patterns and the compartmentalized arrangements of acetylcholinesterase staining within the intermediate and deep collicular layers. Many of the patches of high acetylcholinesterase activity in the intermediate gray layer proved to be zones in which labeled frontotectal and nigrotectal fibers converged. These acetylcholinesterase-rich patches may thus represent sites at which functional influences from the basal ganglia and frontal cortex are coordinated. In the deeper tiers of the intermediate gray layer and layers ventral to it, there were also zones of heightened and diminished acetylcholinesterase staining. Much of this histochemical patterning was reflected in the arrangement of fibers labeled by large rostromedial frontal injections, but these deeper tiers were not strongly labeled after more lateral frontal injections or after injections placed in the substantia nigra. The deeper parts of the acetylcholinesterase-positive gridwork in the superior colliculus are thus distinct from its upper tier of acetylcholinesterase-positive patches. We conclude that the compartmentalized patterning of dense acetylcholinesterase staining in the intermediate and deep collicular layers represents a mosaic architecture to which collicular afferent circuitry is tightly related. This gridwork may serve to set up functional domains within which different aspects of collicular processing are accommodated.

Free-field acoustic stimulation: a reliable, inexpensive system for positioning loudspeaker

We describe a single mechanical system for locating a sound source anywhere on the surface of an imaginary sphere centred on an animal subject. The system has been used in neurophysiological studies of localization in guinea-pig and example neural data are shown.

Cortical projections to the superior colliculus in the macaque monkey: a retrograde study using horseradish peroxidase

The topical and laminar distribution of corticotectal cells, as well as their size and morphology, were studied in the macaque monkey with the horseradish peroxidase (HRP) technique. After HRP injections restricted primarily to the superficial layers of the colliculus, labelled cells were found in visual cortex (areas 17, 18, and 19) and both in the frontal eye field (area 8) and the adjacent part of premotor cortex (area 6). The clustering of labelled cells in visual cortex indicated that each of the anatomically and functionally distinct visual areas has its own set of collicular projections. When intermediate and deeper layers of the colliculus were injected, labelled cells were found also in posterior parietal cortex (area 7) where they were concentrated mainly on the posterior bank of the intraparietal fissure, in inferotemporal cortex (areas 20 and 21), in auditory cortex (area 22), in the somatosensory representation SII (anterior bank of sylvian fissure, area 2), in upper insular cortex (area 14), in motor cortex (area 4), in premotor cortex (area 6), and in prefrontal cortex (area 9). In the motor and premotor cortex, labelled cells formed a continuous band which appeared to stretch across finger-hand-arm-shoulder-neck representation. Similarly, the cluster of labelled cells in area 2 may correspond to the finger-hand representation of SII. The cortical regions not containing labelled cells were the somatosensory representation SI (areas 3, 1 and 2) and the infraorbital cortex. Labelled cells were restricted to layer V of all cortical areas except in the primary visual cortex, where labelled cells were found in both layer V and layer VI. The size spectrum of corticotectal cells ranged from 14.8 micron (average diameter) in area 17 to 27.8 micron in area 6, comprising cells as small as 8 micron and as large as 45 micron. Labelled cells in posterior parietal (area 7), in auditory (area 22), and in motor cortex (area 4) were small and distributed over only a narrow range of sizes. Those in premotor cortex (area 6) were often large and had a wide range in size distribution. The differences in size and morphology of corticotectal neurons suggest that they do not form a uniform class of neurons.

Relationships between the nigrotectal pathway and the cells of origin of the predorsal bundle

The goal of this study was to define the anatomical relationships between the terminal field of the nigrotectal pathway and the tectal neurons which project to contralateral brainstem gaze centers by way of the predorsal bundle. The distribution and morphology of the cells of origin for the predorsal bundle were determined by using a modification of the retrograde horseradish peroxidase technique which homogeneously filled their somas and dendrites. The terminal distribution of the nigrotectal tract was determined using both anterograde horseradish peroxidase and autoradiographic procedures. The results indicate that, in the grey squirrel (Sciurus carolinensis), the predorsal bundle cells are a heterogeneous population whose dendritic fields form a well-defined band confined to the inner half of stratum griseum intermediate. This inner sublamina also can be identified in Nissl and myelin stains. The same sublamina is the major target of the nigrotectal tract. The striking anatomical correspondence between the distribution of nigrotectal terminals and the cells projecting in the predorsal bundle supports a proposal, based on recent physiological investigations, that the nigrotectal tract plays an important role in the initiation of the saccade-related activity of the deep tectal cells (Chevalier et al., '81; Hikosaka and Wurtz, '83a-d).

Auditory receptive fields in primate superior colliculus shift with changes in eye position

The process by which sensory signals are transformed into commands for the control of movement is poorly understood. A potential site for such a transformation is the superior colliculus (SC), which receives auditory, visual and somatosensory inputs and contains neurones that discharge before saccadic eye movements. Along the primary sensory pathways, signals coding the spatial location of auditory, visual and somatosensory targets are based on distinctly different coordinate systems, and it is not known whether each type of sensory input uses a separate motor pathway or if they are converted into a common coordinate system in order to share a single pre-motor circuit. Sensory neurones in the SC have spatially restricted receptive fields (RFs) and are organized into maps across the collicular surface. Acute experiments have shown a rough correspondence between the spatial positions of RFs of neurones encountered along a single dorsal-ventral penetration of the colliculus, regardless of the modality of the effective stimulus, suggesting that auditory, visual and somatosensory maps might be in register. However, in these conditions the head-centred auditory system and the retinotopic visual system are aligned because the eyes are in the primary orbital position. Moreover, other data have suggested that the primate SC is organized in motor, not sensory, coordinates, although in the cat, eye position was found to have no effect on auditory receptive fields. We therefore sought here to determine what happens to the registration of the auditory and visual maps in the alert, behaving animal. Monkeys, with heads fixed, were trained to make delayed saccadic eye movements to auditory or visual targets from one of three initial fixation points while the activity of single neurones was recorded extracellularly. We found that the auditory receptive fields shifted with changes in eye position, allowing the auditory and visual maps to remain in register.

Direct projections to the pretectum and the midbrain reticular formation from auditory relay nuclei in the lower brainstem of the cat

Direct projections to the pretectum and the midbrain reticular formation from auditory relay nuclei in the lower brainstem were examined by the retrograde and anterograde tracer methods in the cat. After horseradish peroxidase (HRP) injection into the pretectomesencephalic reticular region (Pt-MRF), which includes caudoventral regions of the pretectum and rostrodorsal regions of the midbrain reticular formation, labeled neurons were seen in the dorsal nucleus of the lateral lemniscus (DLL), the pericentral (PC) and external (EN) nuclei of the inferior colliculus (IC), the rostral process of IC (RP) and the nucleus of the brachium of IC (NB); no labeled neurons were found in the main laminated portion of the central nucleus of IC. Subsequently, tritiated leucine was injected into DLL, EN, RP or NB for autoradiographic fiber tracing. After injection into DLL or EN, terminal labeling was confined to the ventral portions of the anterior pretectal nucleus. After injection into RP or NB, heavy terminal labeling was observed in the midbrain reticular formation, extending dorsally into the anterior pretectal nucleus. Thus, 3 sectors are distinguishable in Pt-MRF in terms of termination of fibers from the midbrain auditory relay nuclei; the dorsomedial, intermediate or ventrolateral Pt-MRF sector receives fibers arising from DLL, RP or NB, respectively. Fibers from EN terminate only in the dorsal portion (pretectal regions) of the intermediate sector.

Neurons in the superior colliculus of echo-locating bats respond to ultrasonic signals

Using conventional electrophysiological techniques, we demonstrate that neurons in the superior colliculus of the big brown bat (Eptesicus fuscus) respond to ultrasonic signals. Most response properties of these neurons are very similar to neurons of the inferior colliculus in the same bat.

Cells responsive to free-field auditory stimuli in guinea-pig superior colliculus: distribution and response properties

We have investigated the responses of superior colliculus neurones in the anaesthetized guinea-pig to free-field auditory stimulation. The auditory cells were located throughout the deeper laminae and also in the lower part of the stratum opticum. Auditory cells were not found in the rostral pole of the superior colliculus. The auditory responses consisted of a few spikes at stimulus onset with a latency from stimulus arrival at the ear of 7-27 ms. Frequency response areas were measured for forty-five neurones; many of these areas were broad or multipeaked although some were well defined and 'V' shaped. White noise was a more effective stimulus than tones. The majority of cells in our sample responded best to sounds from a restricted horizontal location. Two major response types were found: (1) neurones responding to the same localized area of space despite changes in sound level and (2) neurones responding only to a localized area of space near threshold, but to an extensive area for louder sounds. As the site of the recording electrode was moved from the rostral to the caudal part of the superior colliculus, the location of the auditory receptive fields shifted from the anterior to the posterior field of the animal, thus indicating the presence of a map of auditory space. The visual projection to the guinea-pig superior colliculus was determined and found to be similar to that in other vertebrates. Comparison of visual and auditory space maps in the guinea-pig superior colliculus reveals that receptive fields are coincident over a wide range, but severe discrepancies were evident between the visual and auditory receptive field positions represented at single locations in rostral and caudal colliculus.

Subdivisions of the inferior colliculus in the barn owl (Tyto alba)

The inferior colliculus in the barn owl contains three subdivisions: the central (ICC), external (ICX), and superficial (ICS) nuclei. The nuclei are distinguished on the basis of their cyto- and myeloarchitecture, connectivity, and physiological properties. The ICC may be further divided into dorsal (ICCd) and ventral (ICCv) parts. Auditory fibers ascending in the lateral lemniscus enter the ICCd and ICCv, but not the ICX or ICS. The ICX receives its auditory input from the ICC. The ICC and ICX in owls are similar in position, anatomy, connectivity, and physiology to the ICC and ICX in mammals, suggesting that these structures are homologous. Units in the ICC are organized tonotopically, whereas units in the ICX are organized according to the locations of their spatial receptive fields. This implies that a transformation from a tonotopic to a spatiotopic organization takes place in the ICX of the owl.

Subcortical auditory and somatosensory afferents to hamster superior colliculus

To determine whether the pattern of superior colliculus (SC) afferents seen in cat is generalizable to other mammalian species, HRP was injected into the SC of twenty hamsters (Mesocricetus auratus). After reaction with TMB, subcortical structures were examined for labeled perikarya. Although most subcortical afferents to SC were similar in cat and hamster, there were differences in auditory and somatosensory SC innervation. In hamster only two putatively auditory structures showed labeled cells, the external nucleus of inferior colliculus and nucleus of the brachium of the inferior colliculus, whereas in cat additional cells are reported in the dorsal cochlear nucleus, trapezoid and superior olivary nuclei, and nucleus of the lateral lemniscus. Again, in hamster the major trigeminal somatosensory input to SC is from spinal trigeminal nucleus caudalis whereas in cat trigeminal input is reported to be from the principalis and oralis portions of the trigeminal nucleus. Thus the hamster possesses a very restricted auditory input to and a different pattern of somatosensory innervation of the SC relative to the cat.

Auditory compensation of the effects of visual deprivation in the cat's superior colliculus

Neurones in the superior colliculus of normal and visually deprived cats were analyzed for their responses to visual, auditory and somatosensory stimuli. The percentage of auditory-responsive cells throughout all layers had increased from 11% to 42% after binocular deprivation. Some auditory responses were found even in superficial layers. The number of somatosensory responses, though not systematically tested, was also higher in the visually deprived animals. Visually responsive units did not significantly decrease in number, thus resulting in an increased proportion of multisensory neurones. The vigour of auditory responses had increased after visual deprivation, while the vigour of visual responses had decreased significantly. In addition to the auditory effects of visual deprivation found, our study confirms previous findings on the visual effects of visual deprivation in the superior colliculus. Since only qualitative changes of visual responses, but no suppression of visual by non-visual activity was found, the neuronal mechanisms responsible for these changes may be different from competition as present in the visual cortex.

Morphology of neurons in the torus semicircularis of the northern leopard frog, Rana pipiens pipiens

The neuronal morphology of the torus semicircularis of the northern leopard frog, Rana pipiens pipiens, was examined in Golgi-impregnated material. Neurons in each of the five subdivisions of the torus semicircularis (Potter, '65a) have distinct morphologies which are characteristic of the subdivision. Laminar nucleus neurons are mostly multipolar with spherical or ovoidal somata and smooth dendrites oriented primarily parallel and perpendicular to the cell laminae. Principal nucleus neurons have variable soma shapes with short dendrites (less than 100 micrometers) radiating in all directions. In the magnocellular nucleus, there are three major cell types: neurons characterized by small, spherical-shaped somata, with short, thin, radiating dendrites and many varicosities; bi- or tripolar neurons with ovoidal somata, and long (100-200 micrometers) and smooth dendrites orienting primarily dorsoventrally and mediolaterally; and multipolar neurons with triangular-shaped somata and very long (200-350 micrometers) dendrites, which are either smooth or highly spiny. Neurons in the commissural nucleus are mostly multipolar cells with ovoidal somata and beaded dendrites projecting mostly dorsally and ventrally. The subependymal midline nucleus contains mostly uni- or bipolar neurons with small ovoidal somata and straight, spiny dendrites. In addition to revealing the morphological features of neurons in the torus, the counterstained material shows further cytoarchitectural organization of the principal nucleus, i.e., the presence of a circular lamellar organization. The functional significance of these anatomical features is discussed.

Superior colliculus involvement in the locomotor effects of ambient noise and the stimulants

Partial destruction of the superior colliculus (45%) significantly decreased the normal facilitatory effect of ambient white noise on locomotor activity levels in young rats. As recovery from surgery occurred and as test experience increased, the loss observed immediately following surgery was reduced. Presumably because of the age of the rats examined, destruction of the superior colliculus failed to potentiate the stimulant effects of d-amphetamine or methylphenidate on locomotion. These data suggest that the superior colliculus is involved in changes in general activity that result from manipulation of auditory stimuli in the environment in addition to the documented involvement of the superior colliculus in alterations of general responsivity resulting from manipulations of visual stimuli in the environment. Moreover, the superior colliculus is implicated in maintaining both excitatory and inhibitory changes in response to the environment of the organism.

Evidence of sub-collicular auditory projections to the medial geniculate nucleus in the cat: an autoradiographic and horseradish peroxidase study

Connection of a posteromedial region of the ventral nucleus of the lateral lemniscus were examined in the cat using the autoradiographic tracing method. This sub-collicular region previously had been shown, using retrograde transport of horseradish peroxidase, to send axons to the superior colliculus. The autoradiographic findings revealed that many axons from the posteromedial region of the ventral nucleus of the lateral lemniscus that entered the superior colliculus continued into the midbrain reticular formation. Moreover, other axons traced rostral to the interior colliculus into the thalamus ended in the medial geniculate nucleus, bilaterally. Experiments in which horseradish peroxidase was placed in the medial geniculate nucleus retrogradely labeled the large neurons in the posteromedial region supporting the autoradiographic observations. Other sub-collicular regions also contained labeled cells in these cases, including the main body of the ventral nucleus of the lateral lemniscus and scattered cell groups around the superior olivary complex.

Illusory changes in a sound source and outflow theory

Two experiments, with a total of 220 Ss, were conducted to investigate the possibility of illusory movement of a sound source. In Experiment 1 intensity level (from 10 to 60 db SPL) and frequency of an auditory sound (from 1000 to 5000 Hz) were varied while the Ss head was stabilized with a head rest; in Experiment 2, intensity and frequency were also varied while the S's head was free to move. When head position was fixed, error signal variables such as intensity had no effect on the frequency of illusory direction changes of a sound source. However, when the head position was not fixed, increasing the intensity of the sound source significantly decreased the number of reported illusory direction changes. In addition, in both experiments, illusory changes of intensity were experienced by Ss, and increasing the intensity of the sound source also decreased this illusion. The results were interpreted in the context of an error signal and noise model of autokinetic effects, and the implications for the perception of auditory localization and movement of a sound source were discussed.

Bilateral and multimodal sensory interactions of single cells in the pigeon's midbrain

Standard microelectrode recording techniques were employed to monitor single unit activity in the pigeon's nucleus intercollicularis and medial substantia grisea et fibrosa periventricularis in response to visual, tactile and auditory stimuli. Approximately 40% of the units were driven exclusively by visual stimuli, 8% by tactile stimuli, 47% by both visual and tactile stimuli and a very small percentage by auditory stimuli. Visual receptive fields were generally excitatory in the contralateral eye and suppressive in the ipsilateral eye. Most units were movement selective and some demonstrated direction sensitivity, summation and habituation. Units were generally insensitive to stimulus shape or contrast reversal. Somatosensory receptive fields were located on both sides of the body and were either excitatory or suppressive or both. Ipsilateral visual and somatosensory bimodal inputs were most often of the same sign while ipsilateral visual and contralateral somatosensory bimodal inputs tended to be of opposite sign. Visual and somatosensory receptive field locations of bimodal units tended to be in register.

Afferent sources of a lateral midbrain tegmental zone associated with the pinnae in the cat as mapped by retrograde transport of horseradish peroxidase

A paralemniscal zone in the lateral midbrain tegmentum of the cat has been identified in a possible pathway from the superior colliculus to the facial nucleus that may control pinna movements (Henkel and Edward, '78). Other brainstem afferent projections to this paralemniscal zone have been mapped in the present study using the retrograde horseradish peroxidase tracing method and are discussed in three groups. First, potential sources of auditory afferents were limited mainly to the external nucleus of the inferior colliculus, the nucleus sagulum, and the dorsomedial periolivary cell group. Labeled cells in other superior olivary regions and the dorsal cochlear nucleus were apparently related to uptake of horseradish peroxidase from the axons of the lateral lemniscus. Second, afferents from several premotor regions involved in aspects of gaze control were identified. These were mainly from the nucleus prepositus hypoglossi and adjacent pontine reticular formation, but also included projections from the medial vestibular an abducens nuclei and possibly subthalamic regions such as the zona incerta and fields of Forel. Third, a relatively large group of midbrain afferents was closely related to the origin of the collicular projection to the paralemniscal zone. This group consisted of labeled cells in the periaqueductal gray matter, nucleus cuneiformis, and pretectum. The relatively sparse labeling in auditory regions that projected to the paralemniscal zone seems to indicate that the sensorimotor integration necessary to guide pinna movements does not take place primarily in the lateral midbrain tegmentum. The interaction of gaze-related sources with the pinna-related pathway is also discussed.

Graves' orbitopathy and the thyrotropin-releasing hormone (TRH) test

The value of the thyrotropin-releasing hormone (TRH) test may be insufficiently emphasized in the diagnosis of patients with euthyroid Graves' disease who have unexplained proptosis or vertical diplopia. We saw three patients who had these orbital symptoms and normal routine serum thyroid studies. The orbital computed tomograms (CT) found an orbital myopathy in all, and the diagnosis of Graves' orbitopathy was made by an abnormal TRH test. Not all euthyroid Graves' patients will show a positive result, but in those who do, the test is diagnostic. The clinical summaries of our three patients and the applications of the thyrotropin-releasing hormone test in ophthalmic practice are reviewed.