Responses of MST neurons to plaid stimuli

The estimation of motion information from retinal input is a fundamental function of the primate dorsal visual pathway. Previous work has shown that this function involves multiple cortical areas, with each area integrating information from its predecessors. Compared with neurons in the primary visual cortex (V1), neurons in the middle temporal (MT) area more faithfully represent the velocity of plaid stimuli, and the observation of this pattern selectivity has led to two-stage models in which MT neurons integrate the outputs of component-selective V1 neurons. Motion integration in these models is generally complemented by motion opponency, which refines velocity selectivity. Area MT projects to a third stage of motion processing, the medial superior temporal (MST) area, but surprisingly little is known about MST responses to plaid stimuli. Here we show that increased pattern selectivity in MST is associated with greater prevalence of the mechanisms implemented by two-stage MT models: Compared with MT neurons, MST neurons integrate motion components to a greater degree and exhibit evidence of stronger motion opponency. Moreover, when tested with more challenging unikinetic plaid stimuli, an appreciable percentage of MST neurons are pattern selective, while such selectivity is rare in MT. Surprisingly, increased motion integration is found in MST even for transparent plaid stimuli, which are not typically integrated perceptually. Thus the relationship between MST and MT is qualitatively similar to that between MT and V1, as repeated application of basic motion mechanisms leads to novel selectivities at each stage along the pathway.

Location and Functional Definition of Human Visual Motion Organization Using Functional Magnetic Resonance Imaging

In humans, functional imaging studies have found a homolog of the macaque motion complex, MT+, which is suggested to contain both the middle temporal (MT) and medial superior temporal (MST) areas in the ascending limb of the inferior temporal sulcus. In the macaque, the motion-sensitive MT and MST areas are adjacent in the superior temporal sulcus. Electrophysiology has identified several motion-selective regions in the superior temporal sulcus (STS) of the macaque. Two of the best-studied areas include the MT and MST areas. The MT area has strong projections to the adjacent MST area and is typically subdivided into the dorsal (MSTd) and lateral (MSTl) subregions. While MT encodes the basic elements of motion, MST has higher-order motion-processing abilities and has been implicated in the perception of both object motion and self motion. The macaque MST area has been shown to have considerably larger receptive fields than the MT area. The receptive fields of MT cells typically extend only a few degrees into the ipsilateral visual field, while MST neurons have receptive fields that extend well into the ipsilateral visual field. This study tentatively identifies these subregions as the human homologs of the macaque MT and MST areas, respectively (Fig. 1). Putative human MT and MST areas were typically located on the posterior/ventral and anterior/dorsal banks of a dorsal/posterior limb of the inferior temporal sulcus. These locations are similar to their relative positions in the macaque superior temporal sulcus.

Extrastriate Area MST and Parietal Area VIP Similarly Represent Forward Headings

Many studies have documented the involvement of medial superior temporal extrastriate area (MST) in the perception of heading based on optic flow information. Furthermore, both heading perception and the responses of MST neurons are relatively stable in the presence of eye movements that distort the retinal flow information on which perception is based. Area VIP in the posterior parietal cortex also contains a robust representation of optic flow cues for heading. However, the studies in the two areas were frequently conducted using different stimuli, making quantitative comparison difficult. To remedy this, we studied MST using a family of random dot heading stimuli that we have previously used in the study of VIP. These stimuli simulate observer translation through a three-dimensional cloud of points, and a range of forward headings was presented both with and without horizontal smooth pursuit eye movements. We found that MST neurons, like VIP neurons, respond robustly to these stimuli and partially compensate for the presence of pursuit. Quantitative comparison of the responses revealed no substantial difference between the heading responses of MST and VIP neurons or in their degree of pursuit tolerance.

Linear Responses to Stochastic Motion Signals in Area MST

The medial superior temporal (MST) area contains neurons with tuning for complex motion patterns, but very little is known about the generation of such responses. To explore how neuronal responses varied across complex motion pattern coherence, we recorded from single units while varying the strength of the global motion pattern in random dot stimuli. Stimuli were a family of optic flow patterns, consisting of radial motion, rotary motion, or combinations thereof (“spiral space”). We controlled the strength of the motion in the stimuli by varying the coherence—the proportion of dots carrying the signal. This allows motion strength to be varied independently of stimulus size, speed, or contrast. Most neurons’ responses were well described as a linear function of stimulus coherence. Although more than half the cells possessed significant nonlinearities, these typically accounted for little additional variance. Nonlinear coherence response functions could either be compressive (e.g., saturating) or expansive and occurred in both the preferred and null direction responses. The presence of nonlinearities was not related to neuronal response properties such as preferred spiral-space direction or tuning bandwidth; however, cells with compressive nonlinearities in both the preferred and null directions tended to have higher response amplitudes and were more sensitive to weak motion signals. These cells did not appear to form a distinct subpopulation within MST. Our results suggest that MST neurons predominantly linearly encode increasing pattern motion energy within their RFs.

MST Neurons Code for Visual Motion in Space Independent of Pursuit Eye Movements

When a person tracks a small moving object, the visual images in the background of the visual scene move across his/her retina. It, however, is possible to estimate the actual motion of the images despite the eye-movement-induced motion. To understand the neural mechanism that reconstructs a stable visual world independent of eye movements, we explored areas MT (middle temporal) and MST (medial superior temporal) in the monkey cortex, both of which are known to be essential for visual motion analysis. We recorded the responses of neurons to a moving textured image that appeared briefly on the screen while the monkeys were performing smooth pursuit or stationary fixation tasks. Although neurons in both areas exhibited significant responses to the motion of the textured image with directional selectivity, the responses of MST neurons were mostly correlated with the motion of the image on the screen independent of pursuit eye movement, whereas the responses of MT neurons were mostly correlated with the motion of the image on the retina. Thus these MST neurons were more likely than MT neurons to distinguish between external and self-induced motion. The results are consistent with the idea that MST neurons code for visual motion in the external world while compensating for the counter-rotation of retinal images due to pursuit eye movements.

Responses of Neurons in the Medial Superior Temporal Visual Area to Apparent Motion Stimuli in Macaque Monkeys

Monkeys fixated a stationary spot during presentation of dot textures that moved in apparent motion defined by the spatial and temporal separations, Δx and Δt, between successive flashes of each dot. For each neuron, we assessed the speed tuning for smooth motion (Δt = 2 or 4 ms) at speeds ≤128°/s and the effect of varying the value of Δt at speeds of 16 and 32°/s. Many medial superior temporal (MST) neurons, like middle temporal (MT) neurons, were tuned for the speed of smooth motion and showed decreases in firing rate as the value of Δt increased at a constant speed. A subset of MST neurons, however, showed monotonically increasing firing rates as a function of smooth stimulus speed and responses to apparent motion that paralleled a previously discovered illusion where estimates of target speed increase with the value of Δt. Opponent firing rate, defined as the difference between responses for motion in the preferred and opposite directions, peaked at values of Δt that were consistent with the behavioral illusion. Comparison with a new sample of MT neurons recorded with the same stimuli failed to reveal comparable effects. Attempts to map the population responses in MT and MST onto the behavioral illusion of increased speed succeeded by averaging the opponent response across MST neurons, but only by applying vector averaging to determine the preferred speed of the most active MT neurons. We suggest that a vector-averaging computation transforms MT's place code for target speed into the rate code of some MST neurons.

Relationship Between Extraretinal Component of Firing Rate and Eye Speed in Area MST of Macaque Monkeys

We have isolated extraretinal and retinal components of firing during smooth pursuit eye movements in the medial-superior-temporal area (MST) in the extrastriate visual cortex. Awake macaque monkeys tracked spots in total darkness to eliminate image motion inputs from the background. For 300 ms during sustained tracking at different speeds, the target was stabilized on the moving eye, practically eliminating image motion inputs from the tracking target. The extraretinal component of firing rate during image stabilization was direction selective and related to eye speed but sometimes showed a different preferred speed from the retinal component of the same neuron's responses. The highly variable firing rate of individual MST neurons allowed an ideal observer to predict target speed correctly on 25% of trials. Pooling the data from 71 MST neurons improved the correct response rate to 50%. Behavioral experiments imposed brief perturbations of target velocity to assess the gain of visual-motor transmission for pursuit. The average response to perturbations increased as a function of target speed. However, the size of the responses to individual perturbations allowed an ideal observer to predict target speed correctly on only 35% of the trials. The imprecision of MST responses argues that the output of MST may be a poor candidate to drive eye velocity and so may instead regulate another component of pursuit. The good agreement between the eye velocity precision of the behavioral responses to perturbations of target motion and the firing of MST neurons raises regulation of the visual-motor gain of pursuit as one candidate component.

Discharge Properties of MST Neurons That Project to the Frontal Pursuit Area in Macaque Monkeys

We have used antidromic activation to determine the functional discharge properties of neurons that project to the frontal pursuit area (FPA) from the medial-superior temporal visual area (MST). In awake rhesus monkeys, MST neurons were considered to be activated antidromically if they emitted action potentials at fixed, short latencies after stimulation in the FPA and if the activation passed the collision test. Antidromically activated neurons ( n = 37) and a sample of the overall population of MST neurons ( n = 110) then were studied during pursuit eye movements across a dark background and during laminar motion of a large random-dot texture and optic flow expansion and contraction during fixation. Antidromically activated neurons showed direction tuning during pursuit (25/37), during laminar image motion (21/37), or both (16/37). Of 27 neurons tested with optic flow stimuli, 14 showed tuning for optic flow expansion ( n = 10) or contraction ( n = 4). There were no statistically significant differences in the response properties of the antidromically activated and control samples. Preferred directions for pursuit and laminar image motion did not show any statistically significant biases, and the preferred directions for eye versus image motion in each sample tended to be equally divided between aligned and opposed. There were small differences between the control and antidromically activated populations in preferred speeds for laminar motion and optic flow; these might have reached statistical significance with larger samples of antidromically activated neurons. We conclude that the population of MST neurons projecting to the FPA is highly diverse and quite similar to the general population of neurons in MST.