Relating retinotopic and object-selective responses in human lateral occipital cortex

What is the relationship between retinotopy and object selectivity in human lateral occipital (LO) cortex? We used functional magnetic resonance imaging (fMRI) to examine sensitivity to retinal position and category in LO, an object-selective region positioned posterior to MT along the lateral cortical surface. Six subjects participated in phase-encoded retinotopic mapping experiments as well as block-design experiments in which objects from six different categories were presented at six distinct positions in the visual field. We found substantial position modulation in LO using standard nonobject retinotopic mapping stimuli; this modulation extended beyond the boundaries of visual field maps LO-1 and LO-2. Further, LO showed a pronounced lower visual field bias: more LO voxels represented the lower contralateral visual field, and the mean LO response was higher to objects presented below fixation than above fixation. However, eccentricity effects produced by retinotopic mapping stimuli and objects differed. Whereas LO voxels preferred a range of eccentricities lying mostly outside the fovea in the retinotopic mapping experiment, LO responses were strongest to foveally presented objects. Finally, we found a stronger effect of position than category on both the mean LO response, as well as the distributed response across voxels. Overall these results demonstrate that retinal position exhibits strong effects on neural response in LO and indicates that these position effects may be explained by retinotopic organization.     

Neural coding of global form in the human visual cortex

Extensive psychophysical and computational work proposes that the perception of coherent and meaningful structures in natural images relies on neural processes that convert information about local edges in primary visual cortex to complex object features represented in the temporal cortex. However, the neural basis of these mid-level vision mechanisms in the human brain remains largely unknown. Here, we examine functional MRI (fMRI) selectivity for global forms in the human visual pathways using sensitive multivariate analysis methods that take advantage of information across brain activation patterns. We use Glass patterns, parametrically varying the perceived global form (concentric, radial, translational) while ensuring that the local statistics remain similar. Our findings show a continuum of integration processes that convert selectivity for local signals (orientation, position) in early visual areas to selectivity for global form structure in higher occipitotemporal areas. Interestingly, higher occipitotemporal areas discern differences in global form structure rather than low-level stimulus properties with higher accuracy than early visual areas while relying on information from smaller but more selective neural populations (smaller voxel pattern size), consistent with global pooling mechanisms of local orientation signals. These findings suggest that the human visual system uses a code of increasing efficiency across stages of analysis that is critical for the successful detection and recognition of objects in complex environments.     

Predictive coding for motion stimuli in human early visual cortex

The current study investigates if early visual cortical areas, V1, V2 and V3, use predictive coding to process motion information. Previous studies have reported biased visual motion responses at locations where novel visual information was presented (i.e., the motion trailing edge), which is plausibly linked to the predictability of visual input. Using high-field functional magnetic resonance imaging (fMRI), we measured brain activation during predictable versus unpreceded motion-induced contrast changes during several motion stimuli. We found that unpreceded moving dots appearing at the trailing edge gave rise to enhanced BOLD responses, whereas predictable moving dots at the leading edge resulted in suppressed BOLD responses. Furthermore, we excluded biases in directional sensitivity, shifts in cortical stimulus representation, visuo-spatial attention and classical receptive field effects as viable alternative explanations. The results clearly indicate the presence of predictive coding mechanisms in early visual cortex for visual motion processing, underlying the construction of stable percepts out of highly dynamic visual input.     

Differential coding of pain intensity in the human primary and secondary somatosensory cortex

The primary (SI) and secondary (SII) somatosensory cortices have been shown to participate in human pain processing. However, in humans it is unclear how SI and SII contribute to the encoding of nociceptive stimulus intensity. Using magnetoencephalography (MEG) we recorded responses in SI and SII in eight healthy humans to four different intensities of selectively nociceptive laser stimuli delivered to the dorsum of the right hand. Subjects' pain ratings correlated highly with the applied stimulus intensity. Activation of contralateral SI and bilateral SII showed a significant positive correlation with stimulus intensity. However, the type of dependence on stimulus intensity was different for SI and SII. The relation between SI activity and stimulus intensity resembled an exponential function and matched closely the subjects' pain ratings. In contrast, SII activity showed an S-shaped function with a sharp increase in amplitude only at a stimulus intensity well above pain threshold. The activation pattern of SI suggests participation of SI in the discriminative perception of pain intensity. In contrast, the all-or-none-like activation pattern of SII points against a significant contribution of SII to the sensory-discriminative aspects of pain perception. Instead, SII may subserve recognition of the noxious nature and attention toward painful stimuli.     

Space coding by premotor cortex

Summary Many neurons in inferior area 6, a cortical premotor area, respond to visual stimuli presented in the space around the animal. We were interested to learn whether the receptive fields of these neurons are coded in retinotopic or in body-centered coordinates. To this purpose we recorded single neurons from inferior area 6 (F4 sector) in a monkey trained to fixate a light and detect its dimming. During fixation visual stimuli were moved towards the monkey both within and outside the neurons's receptive field. The fixation point was then moved and the neuron retested with the monkey's gaze deviated to the new location. The results showed that most inferior area 6 visual neurons code the stimulus position in spatial and not in retinal coordinates. It is proposed that these visual neurons are involved in generating the stable body-centered frame of reference necesary for programming visually guided movements.     

Early experience shapes vocal neural coding and perception in songbirds

Songbirds, like humans, are highly accomplished vocal learners. The many parallels between speech and birdsong and conserved features of mammalian and avian auditory systems have led to the emergence of the songbird as a model system for studying the perceptual mechanisms of vocal communication. Laboratory research on songbirds allows the careful control of early life experience and high-resolution analysis of brain function during vocal learning, production, and perception. Here, I review what songbird studies have revealed about the role of early experience in the development of vocal behavior, auditory perception, and the processing of learned vocalizations by auditory neurons. The findings of these studies suggest general principles for how exposure to vocalizations during development and into adulthood influences the perception of learned vocal signals.     

Representation of shapes, edges, and surfaces across multiple cues in the human visual cortex

The lateral occipital complex (LOC) responds preferentially to objects compared with random stimuli or textures independent of the visual cue. However, it is unknown whether the LOC (or other cortical regions) are involved in the processing of edges or global surfaces without shape information. Here, we examined processing of 1) global shape, 2) disconnected edges without a global shape, and 3) global surfaces without edges versus random stimuli across motion and stereo cues. The LOC responded more strongly to global shapes than to edges, surfaces, or random stimuli, for both motion and stereo cues. However, its responses to local edges or global surfaces were not different from random stimuli. This suggests that the LOC processes shapes, not edges or surfaces. LOC also responded more strongly to objects than to holes with the same shape, suggesting sensitivity to border ownership. V7 responded more strongly to edges than to surfaces or random stimuli for both motion and stereo cues, whereas V3a and V4 preferred motion edges. Finally, a region in the caudal intraparietal sulcus (cIPS) responded more strongly to both stereo versus motion and to stereo surfaces versus random stereo (but not to motion surfaces vs. random motion). Thus we found evidence for cue-specific responses to surfaces in the cIPS, both cue-specific and cue-independent responses to edges in intermediate visual areas, and shape-selective responses across multiple cues in the LOC. Overall, these data suggest that integration of visual information across multiple cues is mainly achieved at the level of shape and underscore LOC's role in shape computations.     

Neural activity in prefrontal cortex during copying geometrical shapes

We trained two monkeys to draw copies of geometrical shapes (e.g. squares, triangles) using a joystick, and found that several variables describing the arm trajectories were encoded in the activity of individual prefrontal neurons (Averbeck et al. 2003). Copy trajectories were drawn as sequences of segments, identified by the serial order in which they were drawn and the shape that they together produced. Here we use linear discriminant analysis to test how well the segments of copied shapes could be decoded from the neural activity patterns of small ensembles (3-22 neurons) of simultaneously recorded cells in prefrontal cortex. Using this analysis, the proper segment (drawn by the monkey) was correctly decoded from the ensemble activity pattern during the drawing of that segment in 60-80% of the cases when the largest ensembles were considered. The information transmitted by these ensembles, as well as by single neurons, was also calculated. We found that the information transmitted by the ensembles increased on average with the number of neurons they contained. Each neuron conveyed information about multiple segments within the drawing trajectory, suggesting that neurons were 'broadly tuned' across segments and that the neural code of segment was distributed.     

Odor-concentration coding in the guinea-pig piriform cortex

By optical imaging of intrinsic signals, we demonstrated a possible code for odor concentration in the anterior piriform cortex of the guinea-pig. Odor-induced cortical activation, which primarily originated in layer II, appeared in a narrow band beneath the rhinal sulcus over the lateral olfactory tract, corresponding to the dorsal part of the anterior piriform cortex. Lower concentrations activated the rostral region of the band, whereas higher ones generated caudally spreading activation, and the site at which neural activation reached its maximum extent depended upon odor concentration. Different odors with low concentrations generated distinct but somewhat overlapping patterns in the rostral region of the band; the limited extent of cortical activity may be one focal domain for each odor. It was hard to judge, however, that odor-specific domains appeared in the anterior piriform cortex, because the strong stimuli induced largely overlapping patterns. Furthermore, the total area activated increased in proportion to concentrations raised to a power of 0.5-0.9. Importantly, these imaging results were confirmed with unit recordings which indicated a rostro-caudal gradient in odor-sensitivity among cortical neurons. Our results suggest that the dorsal part of the anterior piriform cortex may be associated with odor concentration. Therefore, in addition to recruitment of more olfactory sensory cells and glomeruli in response to stronger stimuli, a rostro-caudal gradient in axonal projections from mitral/tufted cells and/or in association fibers may play an important role in odor-concentration coding in the anterior piriform cortex.     

Neural activity in prefrontal cortex during copying geometrical shapes

In drawing a copy of a geometrical shape, a sequence of movements must be produced to represent the sides of the object in the proper spatial relationship. We investigated neural mechanisms of this process by training monkeys to draw (using a joystick) copies of geometrical shapes (triangles, squares, trapezoids and inverted triangles) presented on a video monitor while recording single cell activity in prefrontal cortex. The drawing trajectories monkeys produced were divided into a series of discrete segments, varying in direction and length. We performed a stepwise multiple linear regression analysis to identify those copy parameters significantly influencing cell activity. The copied shape (e.g., triangle, square) and the serial position of the segment within each trajectory were the most prevalent effects (in 46% and 43% of cells, respectively), followed by segment direction (32%) and length (16%). Effects of temporal factors (maximum segment speed and time to maximum segment speed) were less frequent. These results demonstrate that prefrontal neurons encode several spatial and sequence variables that define copy trajectories. We also found that specific groupings of significant effects tended to occur together in single neurons. Specifically, single neurons simultaneously processed the serial position of a segment within each trajectory along with the corresponding spatial (but not temporal) attributes of that segment (i.e., direction and length), as well as with the overall shape to which the segments belong. Finally, we discovered that relationships between neural activity and segment serial position were systematic in many instances, described by monotonically increasing and decreasing functions, as well as parabolic functions. These findings indicate that, within the copying task, the serial segment position is a key factor for neural activity in the periprincipalis area of the prefrontal cortex. Electronic Publication     

Early coding of reaching in the parietooccipital cortex

Neural activity was recorded in the parietooccipital cortex while monkeys performed different tasks aimed at investigating visuomotor interactions of retinal, eye, and arm-related signals on neural activity. The tasks were arm reaching 1) to foveated targets; 2) to extrafoveal targets, with constant eye position; 3) within an instructed-delayed paradigm, under both light and darkness; 4) saccadic eye movements toward, and static eye holding on peripheral targets; and 5) visual fixation and stimulation. The activity of many cells was modulated during arm reaction (68%) and movement time (58%), and during static holding of the arm in space (64%), when eye position was kept constant. Eye position influenced the activity of many cells during hand reaction (45%) and movement time (51%) and holding of hand static position (69%). Many cells (56%) were also modulated during preparation for hand movement, in the delayed reach task. Modulation was present also in the dark in 59% of cells during this epoch, 51% during reaction and movement time, and 48% during eye/hand holding on the target. Cells (50%) displaying light-dark differences of activity were considered as related to the sight and monitoring of hand motion and/or position in the visual field. Saccadic eye movements modulated a smaller percentage (25%) of cells than eye position (68%). Visual receptive fields were mapped in 44% of the cells studied. They were generally large and extended to the periphery of the tested (30 degrees ) visual field. Sixty-six percent of cells were motion sensitive. Therefore the activity of many neurons in this area reflects the combined influence of visual, eye, and arm movement-related signals. For most neurons, the orientation of the preferred directions computed across different epochs and tasks, therefore expression of all different eye- and hand-related activity types, clustered within a limited sector of space, the field of global tuning. These spatial fields might be an ideal frame to combine eye and hand signals, thanks to the congruence of their tuning properties. The relationships between cell activity and oculomotor and visuomanual behavior were task dependent. During saccades, most cells were recruited when the eye moved to a spatial location that was also target for hand movement, whereas during hand movement most cells fired depending on whether or not the animal had prior knowledge about the location of the visual targets.     

Sparse coding in striate and extrastriate visual cortex

Theoretical studies of mammalian cortex argue that efficient neural codes should be sparse. However, theoretical and experimental studies have used different definitions of the term "sparse" leading to three assumptions about the nature of sparse codes. First, codes that have high lifetime sparseness require few action potentials. Second, lifetime-sparse codes are also population-sparse. Third, neural codes are optimized to maximize lifetime sparseness. Here, we examine these assumptions in detail and test their validity in primate visual cortex. We show that lifetime and population sparseness are not necessarily correlated and that a code may have high lifetime sparseness regardless of how many action potentials it uses. We measure lifetime sparseness during presentation of natural images in three areas of macaque visual cortex, V1, V2, and V4. We find that lifetime sparseness does not increase across the visual hierarchy. This suggests that the neural code is not simply optimized to maximize lifetime sparseness. We also find that firing rates during a challenging visual task are higher than theoretical values based on metabolic limits and that responses in V1, V2, and V4 are well-described by exponential distributions. These findings are consistent with the hypothesis that neurons are optimized to maximize information transmission subject to metabolic constraints on mean firing rate.     

Circuit dynamics and coding strategies in rodent somatosensory cortex

Previous experimental studies of both cortical barrel and thalamic barreloid neuron responses in rodent somatosensory cortex have indicated an active role for barrel circuitry in processing thalamic signals. Previous modeling studies of the same system have suggested that a major function of the barrel circuit is to render the response magnitude of barrel neurons particularly sensitive to the temporal distribution of thalamic input. Specifically, thalamic inputs that are initially synchronous strongly engage recurrent excitatory connections in the barrel and generate a response that briefly withstands the strong damping effects of inhibitory circuitry. To test this experimentally, we recorded responses from 40 cortical barrel neurons and 63 thalamic barreloid neurons evoked by whisker deflections varying in velocity and amplitude. This stimulus evoked thalamic response profiles that varied in terms of both their magnitude and timing. The magnitude of the thalamic population response, measured as the average number of evoked spikes per stimulus, increased with both deflection velocity and amplitude. On the other hand, the degree of initial synchrony, measured from population peristimulus time histograms, was highly correlated with the velocity of whisker deflection, deflection amplitude having little or no effect on thalamic synchrony. Consistent with the predictions of the model, the cortical population response was determined largely by whisker velocity and was highly correlated with the degree of initial synchrony among thalamic neurons (R(2) = 0.91), as compared with the average number of evoked thalamic spikes (R(2) = 0.38). Individually, the response of nearly all cortical cells displayed a positive correlation with deflection velocity; this homogeneity is consistent with the dependence of the cortical response on local circuit interactions as proposed by the model. By contrast, the response of individual thalamic neurons varied widely. These findings validate the predictions of the modeling studies and, more importantly, demonstrate that the mechanism by which the cortex processes an afferent signal is inextricably linked with, and in fact determines, the saliency of neural codes embedded in the thalamic response.     

Attentional coding for three-dimensional objects and two-dimensional shapes Differential interference effects

 The role played by the attentional mechanisms that enable dominance of relevant objects over distractor objects was investigated by measuring changes in the kinematics of the reach-to-grasp movement. Subjects reached towards three-dimensional (3D) stimuli while attention was diverted towards distracting information consisting of either two-dimensional (2D) projected shapes or 3D objects. Movement kinematics were influenced to a greater degree when a secondary task was performed involving a 3D object rather than a 2D projected shape. When the distractor was 3D, both the reaching and the grasping components were altered but, when it was 2D only, the reaching component was modified. It is suggested that, when attention is directed towards a distractor, it is associated with interference in the kinematics of the action towards the target. Further, the nature and dimensions of the distractor selectively influence the reach or the grasp component of a prehension movement. Received: 17 November 1997 / Accepted: 12 May 1998     

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.     

Task-related coding of stimulus and response in cat motor cortex

Summary In a previous study in the cat, we have reported that motor cortex neurons discharging before the initiation of an aimed forearm response (lead cells) are better timed to movement of a display (stimulus) than to the response. The present study was done to distinguish the coding of stimulus and response features in the discharge patterns of such early activity in motor cortex. Single neurons were recorded in the arm area of motor cortex in three cats performing the same pair of responses (forearm flexion and extension) but to display movements in either of the two directions by changing display polarity. The modulation of lead cell activity was contingent on the occurrence of the learned motor response and timed to the stimulus in all conditions. The majority of lead cells (88%, n = 50) fell into one of two distinct classes. In one class of neurons, force-direction (56%, n = 32), activity was contingent on a single direction of forelimb response (flexion or extension) and was thus independent of the direction of the display stimulus. The only muscles whose patterns matched the activity of this class of response-related neurons were forelimb flexors and extensors. In these neurons, the onset of modulation was timed to one or the other of the two stimuli according to the stimulus direction which elicited the appropriate response. Thus, the display-related input to these neurons varied according to the response required. In the second class of neurons, stimulus-direction (32%, n = 18), modulation was associated with a specific stimulus direction rather than the response direction. The pattern of activity of these neurons was similar to the pattern of EMG signals of shoulder and neck muscles during the different task conditions. The contraction of proximal and axial muscles corresponded to a second response elicited by the stimulus, namely attempts at head rotation towards the moving display and was independent of the conditioned forelimb response in both time of onset and direction. To test the possibility that stimulus-direction neurons participated in the control of head rotation we trained two of the animals to also produce isometric changes in neck torque in the direction of the moving display without making the forelimb response. The activity of stimulus-direction neurons was similarly modulated during performance of the neck task. By contrast, force-direction neurons examined during the neck task were either unmodulated or discharged after the neck response. These data suggest that force-direction neurons participate in response initiation and that their activity is triggered by stimuli specific for the task. The reorganization of the inputs to motor cortex is likely to result from gating mechanisms associated with behavioral set. Such neural gates could provide for the efficient transfer of any member of an array of behaviorally relevant stimuli to restricted sectors of the somatotopically organized motor areas.     

Neural coding of temporal information in auditory thalamus and cortex

How the brain processes temporal information embedded in sounds is a core question in auditory research. This article synthesizes recent studies from our laboratory regarding neural representations of time-varying signals in auditory cortex and thalamus in awake marmoset monkeys. Findings from these studies show that 1) the primary auditory cortex (A1) uses a temporal representation to encode slowly varying acoustic signals and a firing rate–based representation to encode rapidly changing acoustic signals, 2) the dual temporal-rate representations in A1 represent a progressive transformation from the auditory thalamus, 3) firing rate–based representations in the form of a monotonic rate-code are also found to encode slow temporal repetitions in the range of acoustic flutter in A1 and more prevalently in the cortical fields rostral to A1 in the core region of marmoset auditory cortex, suggesting further temporal-to-rate transformations in higher cortical areas. These findings indicate that the auditory cortex forms internal representations of temporal characteristics of sounds that are no longer faithful replicas of their acoustic structures. We suggest that such transformations are necessary for the auditory cortex to perform a wide range of functions including sound segmentation, object processing and multi-sensory integration.     

Neural coding of temporal information in auditory thalamus and cortex

How the brain processes temporal information embedded in sounds is a core question in auditory research. This article synthesizes recent studies from our laboratory regarding neural representations of time-varying signals in auditory cortex and thalamus in awake marmoset monkeys. Findings from these studies show that 1) the primary auditory cortex (A1) uses a temporal representation to encode slowly varying acoustic signals and a firing rate-based representation to encode rapidly changing acoustic signals, 2) the dual temporal-rate representations in A1 represent a progressive transformation from the auditory thalamus, 3) firing rate-based representations in the form of monotonic rate-code are also found to encode slow temporal repetitions in the range of acoustic flutter in A1 and more prevalently in the cortical fields rostral to A1 in the core region of marmoset auditory cortex, suggesting further temporal-to-rate transformations in higher cortical areas. These findings indicate that the auditory cortex forms internal representations of temporal characteristics of sounds that are no longer faithful replicas of their acoustic structures. We suggest that such transformations are necessary for the auditory cortex to perform a wide range of functions including sound segmentation, object processing and multi-sensory integration.