Dermatomal and somatotopic representations in gracile tract at lumbar and cervical levels, illustrating dorsal column “shuffle” for topography that occurs in the projection. Drawings to left show greatly increasing overlap in dermatomal projections. Drawings to right show results of electrophysiological experiments; they reveal somatotopic representation, with improvement of resolution in projection from lumbar to cervical levels. This is accompanied by a modality shuffle, described in text.

Diagram of degree of modality convergence of primary afferents upon cells of origin of spinocervical tract in decerebrate (left column) and spinal (right column) cat. Difference is attributed to action of descending systems originating in brain stem that maintain in a dynamic way the modality specificities shown to left. That action is revealed by wide modality convergences that appear when descending influences are removed in spinal cat (right). HFT, HFG, and HFD: three subclasses of hair follicle afferents.

A: camera lucida reconstruction of 2 spinocervical tract (SCT) neurons of dorsal horn, closely adjacent in rostrocaudal direction, stained by intracellular injection of horseradish peroxidase. Dendrites of the 2 cells are closely entwined. Peripheral receptive field of each neuron is shown directly below; that of neuron to left is completely overlapped by that of neuron to right. B: similar reconstructions of 2 SCT cells closely adjacent in mediolateral direction. Somata are only 135 μm apart, but their dendrites are completely separate and their peripheral receptive fields do not overlap. These arrangements create rostrocaudal columnar organization of dorsal horn neurons from which axons of SCT arise.

A: positions within dorsal horn in a 2.5‐mm length of L₇ spinal segment of cat at which extracellular recordings were made from 53 different neurons of origin of spinocervical tract (SCT) fibers. Observations made in grid of microelectrode penetrations at 250‐μm intervals. Band of cells centers on Rexed's lamina IV. B: horizontal distribution of SCT neurons from another experiment, showing estimated positions of 60 neurons identified by extracellular recording in an experiment similar to that of A. C: plan view of opposite side of cord in same animal as in B, in which 60 SCT neurons were identified after retrograde staining with horseradish peroxidase.

Left: figurine map showing receptive fields of 60 spinocervical tract (SCT) neurons of dorsal horn identified in plan plot of Fig. 4B. There is an organization into longitudinal columns of neurons; receptive fields of those in a single column overlap greatly. As one moves in the transverse direction the receptive fields of neurons encountered jump abruptly from one location to another on body surface. Right: somatotopic schema of representation of hindleg in dorsal horn SCT cells at lumbosacral level of cat's spinal cord. Representation is homotopic, except that there is none for glabrous skin of pad.

Left: figurine map showing receptive fields of 60 spinocervical tract (SCT) neurons of dorsal horn identified in plan plot of Fig. 4B. There is an organization into longitudinal columns of neurons; receptive fields of those in a single column overlap greatly. As one moves in the transverse direction the receptive fields of neurons encountered jump abruptly from one location to another on body surface. Right: somatotopic schema of representation of hindleg in dorsal horn SCT cells at lumbosacral level of cat's spinal cord. Representation is homotopic, except that there is none for glabrous skin of pad.

Schematic diagram of spinocervicothalamic (SC‐LC) and the spinomedullothalamic system (SM‐Z) in cat, described in text. C, T, L: cervical, thoracic, and lumbar regions, respectively, of spinal cord; VPL, ventrobasal complex of thalamus; POm, medial portion of posterior nuclear group of thalamus; CL, centrolateral nucleus of thalamus; C and G, relative positions of cuneate and gracile nuclei; vertical dashed line, midline; Cb, cerebellum.

Schematic representation of dorsal column nuclei and spinal trigeminal nucleus of cat, made from a reconstruction of serial sections. TRIG, spinal trigeminal nucleus; CUN, cuneate nucleus; EX CUN, external cuneate nucleus; G, gracile nucleus. Nuclei × and y are not shown.

Two selected transverse sections through medulla of cat at level of dorsal column nuclear complex. Left: projections of 4 dorsal roots, determined in antegrade tracer experiment by Keller and Hand 330. Right: cartoon of representation of body surfaces of cat within DCN and trigeminal nucleus, composed by Hand from Kruger et al. 348 and from Hand 244. Thus in efferent projection from these nuclei into medial lemniscus to ventrobasal complex of thalamus, representation of body inverts and that of head reverses and inverts, to compose distorted but normal image of representation at thalamic level.

Results of experiment in which microelectrode penetrations were made in horizontal plane, passing from medial to lateral through dorsal column nuclear (DCN) complex of a cat. Receptive fields and modality properties of single neurons and small clusters of neurons of DCN were determined at 50‐μm intervals. Symbols: black, activation by stimulation of deep tissues; diagonal lines, cutaneous; cross‐hatching, Pacinian; white, not identified. Letters beside each reconstruction indicate further modality subdivisions: S, slowly adapting; R, rapidly adapting; L, low velocity; P, Pacinian; T, “tap”. G, gracile nucleus; C, cuneate nucleus; RC, rostral part of cuneate nucleus; EC, external nucleus; ST, spinal trigeminal nucleus. Results show modular organization of DCN complex.

Figurine map showing general pattern of representation of body in single plane of ventrobasal complex of thalamus of monkey, Macaca mulatta, obtained in a gross electrode, evoked‐potential experiment under deep barbiturate anesthesia. Inset: diagram of thalamus in frontal plane studied; dots, points at which electrical activity was evoked by tactile stimulation of body surface; each dot correlates to an appropriately located figurine drawing. Vertical electrode penetrations made at 0.5‐mm intervals in mediolateral direction; recording points 0.5 mm apart in vertical dimension. VBex and VBarc, external and arcuate components of ventrobasal complex; M, mediodorsal nucleus; CM, centre médian; PF, parafascicular nucleus; HP, habenulopeduncular tract; GLD, lateral geniculate nucleus.

Results of a microelectrode single‐neuron mapping study of ventrobasal complex of a cynomolgus monkey, made under barbiturate anesthesia. Receptive fields of single neurons and of small neuron clusters were determined at about 100‐μm intervals as electrodes were passed in an oblique plane from anterolateral to posteromedial, across ventrobasal complex. They reveal lamellar and modular organization of ventrobasal complex described in text. VPLc and VPLm, the 2 components of ventrobasal complex. Tracks 6A and 7A were made 250 μm dorsal to tracks 6 and 7. Clear or stippled, regions in which neurons and neuron clusters were activated by cutaneous stimuli; hatching, regions activated by stimulation of deep tissues. Bars, locations at which single neurons are studied.

Results of a microelectrode single‐neuron study of ventrobasal complex of a cynomolgus monkey, made under deep barbiturate anesthesia. Conventions and labeling as for experiment of Fig. 12, except that here vertical penetrations were made from above in same parasagittal plane, indicated by the upper right inset. Lower right inset, results of similar experiment in which vertical penetrations were made from above, all in a single frontal plane.

Schematic diagram showing on horizontal and frontal planes the lamellar somatotopic representation of contralateral body in ventrobasal complex. Drawings indicate a nearly complete separation of zones for cutaneous and deep modalities. Other evidence, discussed in text, indicates that modules for different modalities are to a certain extent interdigitated in ventrobasal complex.

Outline drawings (above) of thalamic nuclear configurations in histological sections through posterior third of ventrobasal complex of macaque monkey, showing reconstructions of three microelectrode penetrations (P1, P2, P3) and results obtained in them. Open and solid circles, positions at which single neurons studied were activated by stimulation of deep tissues or of skin, respectively. Insets, locations on body of peripheral receptive fields, correlated with modality; here observations on single neurons and small clusters of neurons are combined. Results indicate modular organization of ventrobasal complex described in text, with occasional mixed sequences of modality, and that deep and cutaneous modules are partially interdigitated, not completely sequestered in separate parts of complex. CL, centrolateral nucleus; CM, centre médian; GMM, magnocellular medial geniculate; GMP, principal nucleus, medial geniculate; ID, laterodorsal nucleus; LP, lateroposterior nucleus; MD, mediodorsal nucleus; PF, parafascicular nucleus; PUL, pulvinar; SG, suprageniculate nucleus; VPL, external portion of ventrobasal complex.

Examples of peripheral receptive fields of a sample of neurons studied in posterior nuclear group of thalamus of lightly anesthetized cats. Five neurons of this sample were activated by light mechanical stimulation of skin, from very large bilateral receptive fields. Two were responsive only to noxious stimulation, while one (48 cc) responded to either form of stimulation in large but different receptive fields. Ipsilateral side of body is to right in drawings. Stippled: receptive fields of units driven by mechanical stimuli; diagonal stripes: receptive fields of units driven by noxious stimuli.

Figurine map showing general representation of body surface in postcentral gyrus, as determined with slow‐wave, evoked‐potential method. Each drawing shows body areas within which light mechanical stimulation evoked responses at cortical position of drawing. Black and shading, different intensities of responses. Dashed line at left, depth of central sulcus, so that posterior bank of the central sulcus is turned forward into plane of lateral surface of hemisphere. Similarly, medial wall of the hemisphere and the superior bank of cingular fissure are swung upward into same plane. Upper dashed line, depth of cingular sulcus.

Diagrams of cortices of rat, rabbit, cat, and rhesus monkey showing locations and general plans of organization of precentral motor (MI), supplementary motor (MII), postcentral sensory (SI), and second somatic (SII) areas. Relations to auditory and visual areas are shown, except for monkey auditory area, which lies hidden on lower bank of Sylvian fissure. For rabbit, cat, and monkey the medial walls of hemispheres are swung upward in drawings to occupy same planes as lateral surfaces.

Diagram of the extent of areas 3b, 1, and 2 as seen in an external view of lateral surface of cerebral hemisphere of Macaca mulatta; reconstructed from study of serial sections. Almost all of 3b and all of 3a are hidden within the central fissure. A, arcuate sulcus; CS, central sulcus; IP, intraparietal sulcus; L, lunate sulcus; P, principal sulcus; PCS, postcentral sulcus; PR, superior precental sulcus; S, Sylvian fissure; SC, anterior subcentral sulcus; ST, superior temporal sulcus.

Lower two rows show by common symbols the locations of retrogradely labeled rods of cells in ventrobasal complex (VPLc + VPM) of a cynomolgus monkey, after local cortical injections of horseradish peroxidase. Sites of injection of label into postcentral gyrus are shown in outline drawing of right hemisphere, next above. Upper row, peripheral receptive fields of clusters of neurons at injection sites, determined before injection.

Schematic diagram showing, on sagittal sections, general pattern of input‐output connections of ventrobasal complex and adjacent nuclei, in cynomolgus monkeys. This drawing presents the “core‐shell” hypothesis, which is based on idea that neurons of cutaneous and deep classes are completely segregated in thalamus and thus in their cortical projections. Alternative view that differs slightly is discussed in text: namely, that while there is a degree of this segregation, the modular sets of slowly and quickly adapting cutaneous neurons and of Pacinian neurons, and the deep modalities, are to a certain extent inter‐digitated in ventrobasal complex, as they are in postcentral gyrus.

Diagrams to illustrate cortical connections of sensory and motor cortices in monkeys. Dashed line for connection on left indicates a suggested but still uncertain projection from area 1 to area 3b; otherwise, area 3b is a purely feed‐forward unit. At right: open circles, and solid arrowheads, unidirectional projections; solid circles and lines with double arrowheads, reciprocal connections.

Three‐dimensional model of relation between proliferative layer around ventricular surface of neural tube (VZ) and developing cortical plate (CP), in monkeys.

How results of nerve regeneration experiments provide evidence for columnar organization of somatic cortex; estimate of size of smallest identifiable functional columnar element. Left: results of an experiment in which a recording microelectrode passed nearly parallel to surface of cortex, through a region of neurons of same cutaneous modality. Adjacent columns are related to sequentially adjoining and overlapping peripheral receptive fields and might thus pass unnoticed. Right: the results obtained in a similar experiment after section and resuture of median nerve, with misdirection of the reinnervating nerve bundles. Sudden jumps of receptive‐field locations at intervals of about 40–60 μm reveal cortical minicolumns.

Upper left: drawing of left hemisphere of brain of owl monkey, showing location of representations of hand in ares 3b and 1 of postcentral gyrus. Upper and lower right: illustrations of how detailed maps are constructed from results of microelectrode mapping experiments. Drawings are of hand area of 3b; here rostral is toward lower edge of drawings, lateral to right. Cutaneous receptive fields on various parts of hand, lower center, were mapped for clusters of neurons encountered in the 172 penetration sites shown in drawing at upper right, where symbols represent centers of receptive fields encountered. Stars, penetrations in which receptive fields were on dorsal surfaces of hand. Boundaries between areas of representation for different skin locations were made by halving distance between adjacent penetrations with different receptive fields. D₁‐D₅: digits 1–5; P, M, D: proximal, middle and distal phalanges, glabrous surface; P₁, P₂, P₃, P₄: 4 pads at bases of digits; P₁: insular pads in center of palm; P_TH, P_H: thenar and hypothenar eminences; H, dorsal surface; W; wrist; F, face.

Somatotopic organization of representations of body surface in contralateral areas 3b and 1 of cynomolgus macaque monkey. Map determined by method of cluster recording and multiple microelectrode penetrations illustrated in Fig. 28. Left: representations shown on dorsolateral view of left hemisphere. Right: representations as they appear expanded and unfolded, with anterior bank of postcentral gyrus turned forward into plane of drawing. Receptive‐field regions outlined within drawing; heavy lines enclose areas 3b and 1, dotted line, position of central sulcus, and vertical dashed line near top, region along medial wall where portions of representation in areas 3b and 1 are contained in cortex of medial wall of hemisphere. Representation of individual digits of hand and foot outlined and numbered D₁‐D₅; shaded areas, zones of representation of hairy skin of dorsal surfaces of digits. Summary map constructed from maps obtained in several different animals in which more than one region was mapped in areas 3b and 1; individual maps were combined, based on their overlap.

Detailed evidence that there are 3 representations of digits in somatic sensory cortex of macaque monkeys, 1 each in areas 3b, 1, and 2. A: 3 sets of receptive‐field drawings on outlines of hand show progression, reversal, and regression of fields from tips of fingers toward palm and again out to finger tips as recording sites are moved from anterior edge of 3b toward and across its junction with area 1, and then toward posterior edge of area 1. B: similar evidence for reversals of representations of fingers as recording sites are moved in a similar way across areas 1 and 2. These mirror‐image reversals at the 3b‐1 and 1–2 borders are best evidence for multiple representation hypothesis.

General pattern of representation of deep structures of body in area 2 of somatic sensory cortex of owl monkeys. Sequence from medial to lateral is similar to that of areas 3b and 1, anteriorly. Circles, locations of penetrations in which no neurons drivable by somatic stimuli were observed in these anesthetized animals; these locations were probably in area 5.

Results of receptive‐field study of neuron of area 3b of postcentral gyrus of lightly anesthetized Macaca mulatta showing gradations in intensity of response to stimuli delivered at different positions in field. Receptive field (right) was determined by light mechanical stimulation of skin. Brief electrical stimuli were delivered at points marked by numerals via small electrodes thrust in skin. Intensity of response falls from center toward each edge of field as latency increases (solid line) and probability of response drops (dashed lines). Electrical stimulation of skin at points 3 and 4 also evoked responses, even though they are outside field determined with physiological stimuli. This wide fringe is discussed in text.

Plot in log‐log coordinates of receptive‐field areas for different portions of body surface, in areas 3b and 1 of somatic sensory cortex of owl monkeys.

Degree of overlap of cutaneous receptive fields in areas 3b and 1 of somatic sensory cortex of owl monkeys, plotted against distance between 2 neurons of each pair in tangential dimension of cortex. For fitted straight line, r = 0.79.

Experiment showing interdigitation of modules of cortex for slowly and rapidly adapting neurons in area 3b of owl monkey; digits 1–5 are represented in order from lateral to media. A: peristimulus time histograms of activity evoked in neuron of SA type (upper), in one of RA type (middle), and by 600‐μm step indentation of skin (bottom). B: area 3b of owl monkey. C: sites of microelectrode penetrations within and around digit 4 representation in area 3b. Closed circles, penetrations with SA neurons; open circles, penetrations with RA neurons. Blood vessels outlined on surface of cortex. D: representation of 2 cutaneous modalities for digit 4, obtained from penetrations shown in C, showing regions for tip, and for distal, middle, and proximal phalanges. This supports idea of an intermittently recursive mapping for modalities within topographic pattern.

Results of microelectrode mapping experiment carried out on forelimb region of first somatic area of anesthetized cat. Above right: region studied shown on drawing of dorsal view of cat brain. Receptive fields and modality types determined in each penetration by recording from small clusters of neurons at depths of 500–1,000 μm below cortical surface. Open triangles, deep; open circles, rapidly adapting; closed circles, slowly adapting cutaneous modalities. Results show modality segregation between areas 3a and 3b, and interdigitated regions for RA and SA cutaneous modalities within 3b.

Schematic diagram illustrating concept of Whitsel et al. that postcentral map of hindleg of squirrel monkey is generated by sorting of afferents of different modality types to different spinal pathways that project in a differential but overlapping manner upon postcentral gyrus, where only a single map is thought to exist. Sorted, rapidly adapting, cutaneous afferents of dorsal columns project via central core of ventrobasal complex‐heavily to distal hindlimb region of cortex, lightly to areas for proximal hindlimb. Afferents from muscle and joints, together with SA cutaneous afferents, after sorting and relay project via dorsolateral columns to border zones of ventrobasal complex and from thence predominantly to areas 3b and 1‐heavily to regions for proximal, lightly to those for distal hindlimb.

Inset in figure at right: lateral surface of monkey brain, with Sylvian fissure opened and expanded to reveal cortex buried within it; it is the key to the two larger drawings. Left: drawing summarizing studies of cytoarchitecture and sources of thalamic projections to cortex within Sylvian fissure. Cortical areas: 1, 2–3, 7. Auditory field: Al, first; Pa, posterolateral; RL, rostrolateral; Pi, parainsular. Insular fields: Ig, granular; la, agranular; Id, dysgranular. RI, retroinsular field; T1‐3, temporal fields; SII, second somatic area; PK, prokoniocortical field. Abbreviations for thalamic nuclei, in boxes: LP, lateroposterior; MD, mediodorsal; SG, suprageniculate. Nuclei of pulvinar: Pla, anterior; Pli, inferior; Pll, lateral; Plm, medial Po, posterior. VB, ventrobasal complex; VMb, basal ventromedial. Right: composite summary of patterns of representation of body in second somatic area of monkey and in several surrounding areas that receive somatic afferent input.

A: construction of responses of a projection neuron of cuneate nucleus of a cat to brief mechanical stimuli delivered at various locations within its receptive field on contralateral forefoot (inset upper left); this test stimulus, T₈, is shown on first trace lower left, and impulse histogram evoked by it on second trace. Scales are latency of response, distance of stimulated positions from tip of toes, and impulses per 2‐ms bin in response histogram. On reciprocal interpretation, construction shows distribution of population of cuneate neurons activated by a single stimulus delivered to center of receptive field, and variations in response intensity across population. B: repetition of experiment, but in this case each test stimulus is preceded by a conditioning stimulus delivered within receptive field, which produced excitation followed by afferent inhibition, traces 3 and 4, lower left. On reciprocal interpretation afferent inhibition has narrowed and sharpened distribution of discharging population.

Reconstructions (dashed lines) of populations of neurons of somatic sensory cortex of cat responding to each of 3 shearing mechanical stimuli delivered alone to points 15 mm apart on contralateral foreleg. They show shift of response population with shift of stimulus location, with overlap. Solid line plots population response when 3 stimuli are delivered simultaneously; it is unimodal, with greater response intensity in its center, the result of “funneling.”

A: linear subjective magnitude estimation by human subjects of intensities of mechanical stimuli delivered to glabrous skin of their contralateral forefingers. Points are means for eight healthy young adults, after normalization of their individual scales. Stimuli were step indentations of 600‐ms duration of probe tip 2 mm in diameter machined to a one‐third spherical surface. The 15 stimulus classes were separated by steps of equal amplitude, a range of 2,815 μm, and stimuli from those classes were delivered in a random order. B: linear stimulus‐response relation for slowly adapting (SA) myelinated afferent fiber innervating glabrous skin of monkey's finger. Stimuli as for psychophysical experiment in A, randomly varied over indentation range of 1,600 μm. Number of impulses evoked by each stimulus plotted as function of measured skin indentation for that stimulus. C: linear stimulus‐response relation for SA neuron of area 3b of unanesthetized monkey. Stimuli described above delivered to glabrous skin of contralateral hand; stimulus range 1,624 μm, divided into 15 classes separated by equal steps; stimuli from these classes were delivered in random order. Mean number of impulses evoked by 8–10 stimuli of each class is plotted as function of class amplitude.

Types of synaptic linkages serving afferent inhibition in synaptic transfer regions of lemniscal system. Input axons (1st order, 2nd order, etc.) may terminate on primary projection neurons: 1, on interneurons; 2, with axonal contacts on terminals of neighboring fiber; or on interneurons, 3, which possess polarizing synaptic contacts on adjacent projection neurons. Type 2 cells subserve presynaptic inhibitory mechanisms; type 3 cells are interneurons in postsynaptic inhibitory pathways; others not shown are in descending excitatory pathways. Both presynaptic and postsynaptic inhibitory interneurons, as well as excitatory interneurons, may be activated by recurrent collaterals of projection neurons and via descending pathways from the forebrain.

Spatial extents of excitatory and inhibitory peripheral receptive fields of 8 thalamocortical neurons of ventrobasal complex of anesthetized cats. Abscissa: normalized distance on skin of contralateral forepaw for which unit distance was taken as distance proximal from center of receptive field at which excitation had fallen to 20% of its maximum, at field center. Ordinate: excitation or inhibition as percentage of maximum at field center. Inhibition measured as reduction in response to a supramaximal test stimulus preceded 20–30 ms earlier by conditioning stimulus. All stimuli were gentle, brief jets of air delivered to hairy skin. Three principal features of afferent inhibition are illustrated: coincidence of excitatory and inhibitory field centers; difference in slopes of excitatory and inhibitory receptive fields; and pronounced surround inhibition. On reciprocal interpretation, graphs plot spatial distribution of neurons in ventrobasal complex excitated and inhibited by local peripheral stimulus. Vertical bars = 2 SE of means.

Interaction of excitatory and inhibitory afferent drive on slowly adapting neuron of area 1 of postcentral gyrus of anesthetized macaque monkey, produced by stimuli delivered to its peripheral receptive field on contralateral forearm. Neuron was excited by mechanical stimuli delivered to field outlined on preaxial side of arm and inhibited by mechanical stimuli delivered anywhere within a large surrounding field; only dorsal halves of excitatory and inhibitory fields are shown on drawing at left. Graph at right plots impulse frequency vs. time during excitatory‐inhibitory interaction. Application of steady mechanical stimulus to excitatory field evoked high‐frequency onset transient discharge that declined toward a plateau until interrupted by application of mechanical stimulus to inhibitory receptive field. Upon removal of latter, excitatory sequence was repeated in response to continuing stimulus to excitatory field.

Duration of afferent inhibition for neurons of somatic area I, studied in anesthetized cats. Inhibition measured as reduction in response to air‐jet test stimulus delivered to functional center of peripheral receptive field, produced by a conditioning air‐jet stimulus delivered either proximal or distal in receptive field. Observations on each neuron normalized with respect to maximal inhibition obtained for that neuron; means for 25 neurons displayed. A: mean number of impulses expected in response to excitatory stimulus alone, in each of 5‐ms bins indicated; expectation of an excitatory response is much shorter than is that of inhibition, as shown in B. B: time course of inhibition computed as deviation of observed response when C and T were delivered together (CT) from predicted algebraic summation of response to each of stimuli delivered alone. Vertical bars, SEM; filled circles, overlapping responses to test and conditioning stimuli; open circles, temporally isolated conditioning (C) and test (T) responses. Inhibition appears to increase simultaneously with excitation, and to last much longer.

Sensitivity of neuron of postcentral gyrus of unanesthetized, neuromuscularly blocked Macaca mulatta to velocity and direction of moving cutaneous stimuli. Graph plots peak response at point of intersection of chords shown on drawing to the right, as functions of stimulus velocity and direction. Sensitivity to movements in proximal to distal direction increased with increasing velocities of stimulus movement.

A: average psychometric function for groups of human and monkey subjects required to detect the presence of a 40‐Hz mechanical sinusoid stimulus delivered to glabrous skin of finger tip. Vertical lines = 2 × SEM. B: frequency‐threshold functions for groups of human and monkey subjects asked to detect presence of mechanical sinusoids of different frequencies, delivered to glabrous skin of finger tip. Vertical lines = 2 × SEM.

Psychometric functions for frequency discriminations made by 2 Macaca mulatta monkeys (left) and 2 human subjects (right), made when amplitude of comparison frequency was constant (dotted line) and equal in subjective intensity to standard, and when amplitude of comparison stimulus was intentionally varied above and below that of standard in its subjective intensity (solid lines).

Psychometric functions for amplitude discriminations made by 2 Macaca mulatta monkeys and 3 human subjects, at each of three standard amplitudes at 30 Hz.

Replicas of impulse discharges of neuron in postcentral gyrus of unanesthetized, neuromuscularly blocked Macaca mulatta, responding to mechanical sinusoids of 40 Hz and at different amplitudes (figures in μm) delivered to glabrous skin of contralateral hand. Each horizontal line replicates neural discharges during single stimulus; each group of lines represents successive responses to stimuli of identical amplitude. Each trace begins with onset of mechanical sinusoid. There is clear replication of frequency of stimulus in a periodicity code, which contrasts with the 1‐impulseper‐cycle code for peripheral rapidly adapting fibers, shown in Figs. 15 and 18 of the chapter on touch by Darian‐Smith, in this Handbook.

Expectation density (left) and renewal density (right) histograms for responses of neuron of postcentral gyrus of unanesthetized, neuromuscularly blocked Macaca mulatta, linked to rapidly adapting afferents innervating glabrous skin of contralateral hand. Mechanical sinusoids (at 30 Hz) delivered at amplitudes indicated (in μm, at far right) for each pair of histograms. Histograms to left show strong replication of stimulus frequency, just apparent at 6 μm and clear at 13 μm. Monkey detection threshold at 30 Hz is 608 μm. Renewal density histogram is essentially a repetition of expectation density analysis after random shuffling of interval sequences to destroy temporal sequence. Results show importance of temporal order in which impulses occur for this particular neural code. Bin size: 2 ms at left; 375 μs at right.

Set of receiver operating‐characteristic curves for a human observer discriminating 40–Hz mechanical sinusoids from those of other frequencies, delivered to distal pad of right middle finger. Discriminations were made at about 15 dB above threshold; 200 trials for each curve. Inset, relation of index of detectability to differences in cycle lengths between comparison frequencies and 40 Hz.

A: psychophysical (large dots) and neural (small dots) SD values at different stimulus frequencies. Large dots with solid line, SD values of distribution of sensory events at different standard frequencies for one human observer as mechanical stimuli 20 dB above threshold were delivered to distal pad of right forefinger. Dotted line, calculated from SD values, shows the differences in cycle lengths between two stimuli that this observer would be able to discriminate in a forced‐choice experiment at level of P = 0.95 correct answer. Small dots, SD values for cycle histograms from 13 postcentral rapidly adapting neurons, at different frequencies, and at those intensities at which the values were smallest. B: relation of psychophysical and neural uncertainties, to stimulus cycle lengths at different frequencies, for mechanical sinusoids delivered to the glabrous skin of hand. Heavy line, relation of discriminable difference in cycle length (left) to cycle length of standard frequency, a Weber function. Small dots, relation of standard deviations of cycle histograms of postcentral neuronal discharges (left) to stimulus cycle length, which is analogous to coefficient of variation.

Results obtained in combined psychophysical and electrophysiological experiment carried out in waking, behaving Macaca mulatta. Solid line, psychometric function, plotting percentage of correct detections the monkey made of mechanical sinusoids at 30 Hz, of different amplitudes, delivered to glabrous skin of a contralateral finger. Stimulus delivered to center of peripheral receptive field of rapidly adapting neuron in monkey's postcentral gyrus. Dashed lines, relative energy in neuronal signal at stimulus frequency (1st Har.) and at the IInd and IIIrd harmonics, obtained by Fourier analysis. Strength of cortical neuronal signal at stimulus frequency rose at equal rate with increase in likelihood of detection.

Illustrations of maps obtained in experiments of type illustrated in Fig. 25. They show variation in details of representations in areas 3b and 1 for 2 owl monkeys, left, and 2 squirrel monkeys, right.

Reorganizations of hand representations in areas 3b and 1 of owl monkeys after section of medial nerve. A: location and normal maps for areas 3b and 1; two representations are nearly mirror images. Digits and palmar pads numbered in order; insular (I), hypothenar (H), thenar (T) pads; distal (d), middle (m), and proximal (p) phalanges. B: normal hand representations in detail. C: cortex normally related to median nerve shown by dots, depried of afferent input after nerve section. D: completed reorganization of maps several months after nerve section. Much of formerly deprived cortex is now activated by stimulation of dorsal surfaces of hand and digits (black). Palmar pads innervated by ulnar nerve have an increased cortical representation.