A neural surveyor in somatosensory cortex

SummaryPerhaps the most recognizable sensory map in all of neuroscience is the somatosensory homunculus. Though it seems straightforward, this simple representation belies the complex link between an activation in somatosensory Area 3b and the associated touch location on the body. Any isolated activation is spatially ambiguous without a neural decoder that can read its position within the entire map, though how this is computed by neural networks is unknown. We propose that somatosensory cortex implements multilateration, a common computation used by surveying and GPS systems to localize objects. Specifically, to decode touch location on the body, the somatosensory system estimates the relative distance between the afferent input and the body’s joints. We show that a simple feedforward neural network which captures the receptive field properties of somatosensory cortex implements a Bayes-optimal multilateral decoder via a combination of bell-shaped (Area 3b) and sigmoidal (Areas 1/2) tuning curves. Simulations demonstrated that this decoder produced a unique pattern of localization variability between two joints that was not produced by other known neural decoders. Finally, we identify this neural signature of multilateration in actual psychophysical experiments, suggesting that it is a candidate computational mechanism underlying tactile localization.

Download Full-text

A neural surveyor to map touch on the body

Proceedings of the National Academy of Sciences ◽

10.1073/pnas.2102233118 ◽

2022 ◽

Vol 119 (1) ◽

pp. e2102233118

Author(s):

Luke E. Miller ◽

Cécile Fabio ◽

Malika Azaroual ◽

Dollyane Muret ◽

Robert J. van Beers ◽

...

Keyword(s):

Somatosensory System ◽

Body Part ◽

Afferent Input ◽

The Body ◽

Positioning Systems ◽

Tactile Localization ◽

Global Positioning ◽

Somatosensory Neurons ◽

Somatotopic Map ◽

Sensory Map

Perhaps the most recognizable sensory map in all of neuroscience is the somatosensory homunculus. Although it seems straightforward, this simple representation belies the complex link between an activation in a somatotopic map and the associated touch location on the body. Any isolated activation is spatially ambiguous without a neural decoder that can read its position within the entire map, but how this is computed by neural networks is unknown. We propose that the somatosensory system implements multilateration, a common computation used by surveying and global positioning systems to localize objects. Specifically, to decode touch location on the body, multilateration estimates the relative distance between the afferent input and the boundaries of a body part (e.g., the joints of a limb). We show that a simple feedforward neural network, which captures several fundamental receptive field properties of cortical somatosensory neurons, can implement a Bayes-optimal multilateral computation. Simulations demonstrated that this decoder produced a pattern of localization variability between two boundaries that was unique to multilateration. Finally, we identify this computational signature of multilateration in actual psychophysical experiments, suggesting that it is a candidate computational mechanism underlying tactile localization.

Download Full-text

Reorganization of somatosensory area 3b representations in adult owl monkeys after digital syndactyly

Journal of Neurophysiology ◽

10.1152/jn.1991.66.3.1048 ◽

1991 ◽

Vol 66 (3) ◽

pp. 1048-1058 ◽

Cited By ~ 169

Author(s):

T. Allard ◽

S. A. Clark ◽

W. M. Jenkins ◽

M. M. Merzenich

Keyword(s):

Temporal Structure ◽

Receptive Fields ◽

Somatosensory System ◽

Critical Time ◽

Experimental Manipulation ◽

Afferent Input ◽

Single Digit ◽

Cortical Maps ◽

Afferent Inputs ◽

Area 3B

1. These experiments were designed to test the hypothesis that temporally correlated afferent input activity plays a lifelong role in the establishment and modification of receptive fields (RFs) and representational topographies in the primary somatosensory cortex of adult monkeys. They were based in part on the finding that adjacent digits of the hand are represented discontinuously in area 3b of the adult owl monkey. If cortical receptive fields and the details of cortical topographic representations are shaped by the weights of the temporal correlations among afferent inputs, then representational discontinuities between digits would be expected to arise because inputs from the skin surfaces of adjacent digits are largely independent in the critical time domain. 2. In the present experiments, the skin of adjacent digits 3 and 4 of the monkey hand was surgically connected to create an artificial syndactyly, or webbed-finger condition. Highly detailed microelectrode maps of the cortical representation of the syndactyl digits were obtained 3-7.5 mo later. This experimental manipulation greatly increased the amount of simultaneous or nearly simultaneous input from the normally separated, now fused, surfaces of adjacent fingers. 3. Cortical maps of the representations of finger surfaces were highly modified from the normal after a several-month-long period of digital fusion. Specifically, the normal discontinuity between the cortical representations of adjacent fingers was abolished. Within a wide cortical zone, RFs were defined that extended across the line of syndactyly onto the surgically joined skin of both fused digits. The representational topography of the fused digits was similar to any normal single digit and was characterized by a continuous progression of partially overlapping RFs. 4. Control observations revealed that these reorganizational changes cannot be accounted for by any changes in cutaneous innervation induced by the surgery. They must arise from representational changes in the central somatosensory system. 5. These findings reveal that cortical maps can be altered in detail in adult monkeys by modifying the distributed temporal structure of afferent inputs. They support the longstanding hypothesis that the temporal coincidence of inputs plays a role in the grouping of input subsets into specific cortical RFs and, consequently, in the shaping of selected effective cortical inputs and representational topographies throughout life.

Download Full-text

Organization of primary somatosensory cortex in the cat

Journal of Neurophysiology ◽

10.1152/jn.1980.43.6.1527 ◽

1980 ◽

Vol 43 (6) ◽

pp. 1527-1546 ◽

Cited By ~ 104

Author(s):

R. W. Dykes ◽

D. D. Rasmusson ◽

P. B. Hoeltzell

Keyword(s):

Somatosensory Cortex ◽

Abrupt Change ◽

The Body ◽

Primary Somatosensory Cortex ◽

Caudal Portion ◽

Area 3A ◽

Rostral Portion ◽

Cytoarchitectonic Areas ◽

Area 3B ◽

Made In

1. Multi-unit recordings were made from SI cortex of barbiturte-anesthetized cats. In four cats, multiple vertical penetrations were made at closely spaced intervals. In 12 cats, long surface-parallel penetrations were made in the rostrocaudal or the lateromedial directions with observations taken every 100 micron. 2. Evidence is presented suggesting that cytoarchitectonic area 3a receives input from deep receptors and area 3b receives input from cutaneous receptors. 3. Within area 3b there was an abrupt change in submodality such that the rostral portion of 3b was activated by slowly adapting (SA) afferents, while the caudal portion was activated by rapidly adapting (RA) afferents. 4. The change in modality from deep to cutaneous occurred at the 3a/3b border, but the change in submodality occurred within area 3b and there was no obvious anatomical correlate of the latter transition. 5. These data suggest that there are modality- and submodality-specific bands in register with the bands of cytoarchitecture that extend across the mediolateral dimension of primary somatosensory cortex (SI). 6. A particular receptor population (or populations) from all regions of the body delivers information to each functionally specific band--one map is found in area 3a and two are in area 3b. If this pattern holds for the rest of cat SI, then there must be additional maps of the body in cytoarchitectonic areas 1 and 2.

Download Full-text

Faculty Opinions recommendation of Optical imaging of a tactile illusion in area 3b of the primary somatosensory cortex.

Faculty Opinions – Post-Publication Peer Review of the Biomedical Literature ◽

10.3410/f.1016771.211559 ◽

2004 ◽

Author(s):

Corinna Darian-Smith

Keyword(s):

Optical Imaging ◽

Somatosensory Cortex ◽

Primary Somatosensory Cortex ◽

Tactile Illusion ◽

Area 3B

Download Full-text

Somatosensory Cortex Efficiently Processes Touch Located Beyond the Body

Current Biology ◽

10.1016/j.cub.2019.10.043 ◽

2019 ◽

Vol 29 (24) ◽

pp. 4276-4283.e5 ◽

Cited By ~ 6

Author(s):

Luke E. Miller ◽

Cécile Fabio ◽

Valeria Ravenda ◽

Salam Bahmad ◽

Eric Koun ◽

...

Keyword(s):

Somatosensory Cortex ◽

The Body

Download Full-text

Serial and parallel processing of tactual information in somatosensory cortex of rhesus monkeys

Journal of Neurophysiology ◽

10.1152/jn.1992.68.2.518 ◽

1992 ◽

Vol 68 (2) ◽

pp. 518-527 ◽

Cited By ~ 124

Author(s):

T. P. Pons ◽

P. E. Garraghty ◽

M. Mishkin

Keyword(s):

Substantial Reduction ◽

Receptive Fields ◽

Somatosensory System ◽

Encounter Rate ◽

Specific Information ◽

Tactile Information ◽

Cortical Areas ◽

Area 3A ◽

Area 3B ◽

Made In

1. Selective ablations of the hand representations in postcentral cortical areas 3a, 3b, 1, and 2 were made in different combinations to determine each area's contribution to the responsivity and modality properties of neurons in the hand representation in SII. 2. Ablations that left intact only the postcentral areas that process predominantly cutaneous inputs (i.e., areas 3b and 1) yielded SII recording sites responsive to cutaneous stimulation and none driven exclusively by high-intensity or "deep" stimulation. Conversely, ablations that left intact only the postcentral areas that process predominantly deep receptor inputs (i.e., areas 3a and 2) yielded mostly SII recording sites that responded exclusively to deep stimulation. 3. Ablations that left intact only area 3a or only area 2 yielded substantial and roughly equal reductions in the number of deep receptive fields in SII. By contrast, ablations that left intact only area 3b or only area 1 yielded unequal reductions in the number of cutaneous receptive fields in SII: a small reduction when area 3b alone was intact but a somewhat larger one when only area 1 was intact. 4. Finally, when the hand representation in area 3b was ablated, leaving areas 3a, 1, and 2 fully intact, there was again a substantial reduction in the encounter rate of cutaneous receptive fields. 5. The partial ablations often led to unresponsive sites in the SII hand representation. In SII representations other than of the hand no such unresponsive sites were found and there were no substantial changes in the ratio of cutaneous to deep receptive fields, indicating that the foregoing results were not due to long-lasting postsurgical depression or effects of anesthesia. 6. The findings indicate that modality-specific information is relayed from postcentral cortical areas to SII along parallel channels, with cutaneous inputs transmitted via areas 3b and 1, and deep inputs via areas 3a and 2. Further, area 3b provides the major source of cutaneous input to SII, directly and perhaps also via area 1. 7. The results are in line with accumulating anatomic and electrophysiologic evidence pointing to an evolutionary shift in the organization of the somatosensory system from the general mammalian plan, in which tactile information is processed in parallel in SI and SII, to a new organization in higher primates in which the processing of tactile information proceeds serially from SI to SII. The presumed functional advantages of this evolutionary shift are unknown.

Download Full-text

Human cortical potentials evoked by stimulation of the median nerve. II. Cytoarchitectonic areas generating long-latency activity

Journal of Neurophysiology ◽

10.1152/jn.1989.62.3.711 ◽

1989 ◽

Vol 62 (3) ◽

pp. 711-722 ◽

Cited By ~ 225

Author(s):

T. Allison ◽

G. McCarthy ◽

C. C. Wood ◽

P. D. Williamson ◽

D. D. Spencer

Keyword(s):

Spatial Distribution ◽

Median Nerve ◽

Somatosensory Cortex ◽

Sensorimotor Cortex ◽

Preceding Paper ◽

Short Latency ◽

Cortical Surface ◽

Long Latency ◽

Area 3B ◽

Stimulation Of

1. The anatomic generators of human median nerve somatosensory evoked potentials (SEPs) in the 40 to 250-ms latency range were investigated in 54 patients by means of cortical-surface and transcortical recordings obtained during neurosurgery. 2. Contralateral stimulation evoked three groups of SEPs recorded from the hand representation area of sensorimotor cortex: P45-N80-P180, recorded anterior to the central sulcus (CS) and maximal on the precentral gyrus; N45-P80-N180, recorded posterior to the CS and maximal on the postcentral gyrus; and P50-N90-P190, recorded near and on either side of the CS. 3. P45-N80-P180 inverted in polarity to N45-P80-N180 across the CS but was similar in polarity from the cortical surface and white matter in transcortical recordings. These spatial distributions were similar to those of the short-latency P20-N30 and N20-P30 potentials described in the preceding paper, suggesting that these long-latency potentials are generated in area 3b of somatosensory cortex. 4. P50-N90-P190 was largest over the anterior one-half of somatosensory cortex and did not show polarity inversion across the CS. This spatial distribution was similar to that of the short-latency P25-N35 potentials described in the preceding paper and, together with our and Goldring et al. 1970; Stohr and Goldring 1969 transcortical recordings, suggest that these long-latency potentials are generated in area 1 of somatosensory cortex. 5. SEPs of apparently local origin were recorded from several regions of sensorimotor cortex to stimulation of the ipsilateral median nerve. Surface and transcortical recordings suggest that the ipsilateral potentials are generated not in area 3b, but rather in other regions of sensorimotor cortex perhaps including areas 4, 1, 2, and 7. This spatial distribution suggests that the ipsilateral potentials are generated by transcallosal input from the contralateral hemisphere. 6. Recordings from the periSylvian region were characterized by P100 and N100, recorded above and below the Sylvian sulcus (SS) respectively. This distribution suggests a tangential generator located in the upper wall of the SS in the second somatosensory area (SII). In addition, N125 and P200, recorded near and on either side of the SS, suggest a radial generator in a portion of SII located in surface cortex above the SS. 7. In comparison with the short-latency SEPs described in the preceding paper, the long-latency potentials were more variable and were more affected by intraoperative conditions.

Download Full-text

Stretch of Quadriceps Inhibits the Soleus H Reflex During Locomotion in Decerebrate Cats

Journal of Neurophysiology ◽

10.1152/jn.1997.78.6.2975 ◽

1997 ◽

Vol 78 (6) ◽

pp. 2975-2984 ◽

Cited By ~ 27

Author(s):

John E. Misiaszek ◽

Keir G. Pearson

Keyword(s):

Stretch Reflex ◽

Stance Phase ◽

Background Level ◽

Afferent Input ◽

H Reflex ◽

The Body ◽

Afferent Feedback ◽

Electromyographic Activity ◽

Locomotor Rhythm ◽

Decerebrate Cats

Misiaszek, John E. and Keir G. Pearson. Stretch of quadriceps inhibits the soleus H reflex during locomotion in decerebrate cats. J. Neurophysiol. 78: 2975–2984, 1997. Previously, it has been demonstrated that afferent signals from the quadriceps muscles can suppress H reflexes in humans during passive movements of the leg. To establish whether afferent input from quadriceps contributes to the modulation of the soleus H reflex during locomotion, the soleus H reflex was conditioned with stretches of the quadriceps muscle during bouts of spontaneous treadmill locomotion in decerebrate cats. We hypothesized that 1) in the absence of locomotion such conditioning would lead to suppression of the soleus H reflex and 2) this would be retained during periods of locomotor activity. In the absence of locomotion, slow sinusoidal stretches (0.2 Hz, 8 mm) of quadriceps cyclically modulated the amplitude of the soleus H reflex. The H reflex amplitude was least during the lengthening of the quadriceps and greatest as quadriceps shortened. Further, low-amplitude vibrations (48–78 μm) applied to the patellar tendon suppressed the reflex, indicating that the muscle spindle primaries were the receptor eliciting the effect. During bouts of locomotion, ramp stretches of quadriceps were applied during the extensor phase of the locomotor rhythm. Soleus H reflexes sampled at two points during the stance phase were reduced compared with phase-matched controls. The background level of the soleus electromyographic activity was not influenced by the applied stretches to quadriceps, either during locomotion or in the absence of locomotion. This indicates that the excitability of the soleus motoneuron pool was not influenced by the stretching of quadriceps, and that the inhibition of the soleus H reflex is due to presynaptic inhibition. We conclude that group Ia afferent feedback from quadriceps contributes to the regulation of the soleus H reflex during the stance phase of locomotion in decerebrate cats. This afferent mediated source of regulation of the H reflex, or monosynaptic stretch reflex, would allow for rapid alterations in reflex gain according to the dynamic needs of the animal. During early stance, this source of regulation might suppress the soleus stretch reflex to allow adequate yielding at the ankle and facilitate the movement of the body over the foot.

Download Full-text