Chapter 7 Thalamocortical and corticocortical interactions in the somatosensory system

M.A.L. Nicolelis (Ed.)

Progressin BrainResearch,Vol. 130 © 2001 Elsevier Science B.V. All fights reserved

CHAPTER 7

Thalamocortical and corticocortical interactions in the somatosensory system Miguel A.L. Nicolelis 1,2,3,, and Marshall Shuler 1 1 Department of Neurobiology, Duke University, Durham, NC 27710, USA 2 Department of Biomedical Engineering, Duke University, Durham, NC 27710, USA 3 Department of Psychology: Experimental, Duke University, Durham, NC 27710, USA

Introduction Until recently, neurophysiological theories aimed at accounting for the exquisite tactile perceptual capabilities of mammals have been dominated by the notion that the somatosensory system relies primarily on feedforward computations to generate a broad spectrum of sensations (e.g. fine touch, thermo sensation, pain, etc.) (Mountcastle, 1957, 1974; Dykes, 1983; Johnson et al., 1995). Despite the widespread acceptance of this view, over the last three decades, considerable anatomical, physiological, and behavioral evidence has been put forward to challenge a pure feedforward view of touch. Accordingly, more than ever, the potential contribution of some of the main 'building blocks' of this model of touch, concepts such as the classic receptive field, independent parallel pathways from the periphery to the cortex, cortical columns, and static somatotopic maps, have been the subject of considerable debate (Merzenich et al., 1983; Purves et al., 1992; Ghazanfar and Nicolelis, 1999). The anatomical organization of the somatosensory system provides the first hint that tactile in-

*Corresponding author: Miguel A.L Nicolelis, Department of Neurobiology, Box 3209, Bryan Research Building, Room 333, 101 Research Drive, Durham, NC 27710, USA. Tel. +1-919-684-4580; Fax: +1-919-684-5435; E-mail: [email protected]

formation processing involves more than just feedforward interactions (Fig. 1). On their way to the neocortex, ascending somatosensory pathways converge on neurons located in a series of subcortical nuclei in the spinal cord, brainstem, and thalamus (Kaas and Pons, 1988). In addition to these parallel feedforward pathways that convey information from the peripbery to the cortex, neurons located at cortical and subeortical regions that define the different processing levels of the somatosensory system receive convergent input from multiple descending feedback pathways, originating in several cortical areas (Kaas, 1990). These feedback projections define extensive thalamocortical and corticocortical loops that are largely ignored by the classical feedforward model of touch. Part of the reason for this omission is the inherent experimental difficulty of measuring the potential effects of feedback projections, which has contributed to a scarcity of information and conflicting data regarding the physiological role of recurrent circuits. Research on artificial neural networks suggests that the main computational advantage of a neural system with highly recurrent projections is that such networks do not have to synthesize a new view of the world every time new raw information is sampled (as suggested by a pure FF model). Instead, a recurrent system can use previously learned experiences to generate an 'internal' model of the world (Mumford, 1994), and take advantage of this model to generate expectations and predictions every time an

90

Fig. 1. Schematic diagram of the rat trigeminal somatosensory system. Whiskers on the rat's snout are labeled according to the row and column in which they are located. Whisker columns are labeled from 1 to 5, caudal to rostral, while whisker rows are labeled A to E, dorsal to ventral. Peripheral nerve fibers innervating single whisker follicles have their cell bodies located in the trigeminal ganglion (Vg). Here, only the projections from Vg neurons to two main subdivisions of the trigeminal brainstem complex, the principal trigeminal nucleus (PrV), and the spinal trigeminal nucleus (SpV), are illustrated. Proponents of the feedforward model of touch usually divide these projections into rapidly adapting (RA) and slowly adapting (SA) fibers, according to their physiological responses to tactile stimuli (see text). Each of these categories contains further subdivisions, which are not described here. Neurons located in these two brainstem nuclei give rise to parallel excitatory projections to the ventroposterior medial nucleus of the thalamus (VPM). Neurons in VPM give rise to projections to layer IV of the primary somatosensory cortex (SI). A collateral of these thalamocortical projections reach the reticular nucleus (RT), whose neurons provide the main source of GABAergic inhibition to the VPM. Descending excitatory corticothalamic projections, originating in layer VI of the SI cortex, reach the VPM and the reticular nucleus of the thalamus (RTn). The assumed topographic arrangement of these projections in the VPM and the RT are illustrated in the scheme, Feedback corticofugal projections originated in layer V of the SI cortex also reach the trigeminal brainstem complex, targeting primarily the SpV subdivision.

V0

_

,

exploratory tactile behavior is planned. In the case of the somatosensory system, the vast network o f corticocortical connections and massive corticofugal pathways to the thalamus, brainstem and spinal cord would provide the anatomical substrate for the dissemination o f predictions o f an internal touch model of the world across multiple cortical and subcortical areas (Grossberg, 1999). The interaction between these 'expectations' and raw tactile information provided by feedforward pathways could then define the computations needed to allow animals to accurately perceive the nature o f any given tactile stimulus as well as provide a continuous update o f the internal model. This would be accomplished through dynamic interactions between descending and ascending projections that reciprocally connect populations of cortical and thalamic neurons.

The recent introduction of electrophysiological methods that allow one to simultaneously record the activity of large populations of single neurons, located in multiple cortical areas and subcortical nuclei while carrying out selective and reversible pharmacological inactivation of selective regions o f the cortex has provided new insights on the functional contribution o f corticocortical and thalamocortical loops in tactile information processing. Here, we review some physiological results obtained in our laboratory regarding two o f these recurrent circuits in the rat somatosensory system: the thalamocortical loop between the primary somatosensory cortex (SI) and the ventral posterior medial (VPM) nucleus of the thalamus; and the loop formed by the reciprocal callosal connections between the SI cortices.

The potential role of corticothalamic feedback projections in tactile information processing Like all other m a m m a l i a n sensory systems, the somatosensory system contains massive feedback cor-

91 ticocortical and corticothalamic projections, which define closed loops between cortical areas and between the cortex and the thalamus. For instance, in primates, feedforward somatosensory pathways terminate in four distinct somatosensory areas located in the anterior parietal cortex (areas 3a, 3b, 1 and 2). Projections from the anterior parietal cortex also reach motor cortical areas in the frontal lobe, the secondary somatosensory cortex, and the somatosensory and multi-modal cortical areas of the posterior parietal cortex (Kaas et al., 1983). Somatosensory cortical areas in the parietal cortex are reciprocally connected through feedforward and feedback projections. In addition, cortical neurons located in some somatosensory and motor cortical areas are also connected by massive corticocortical feedback projections. Another source of massive corticofugal projections, which originate in the infragranular layers of the primary and higher order somatosensory cortical areas, project to all intermediary subcortical relays (i.e spinal cord, brainstem, and thalamic nuclei) of the somatosensory system (Kaas et al., 1983). Indeed, once the full domain of these feedback projections is considered, the somatosensory system can only be defined as a highly recurrent network, in which multiple feedback projections are intertwined with several parallel feedforward pathways. The importance of corticothalamic projections can be illustrated by a brief description of the anatomical organization and the physiological effects mediated by these pathways in the somatosensory system of rodents. As in every other mammalian species (Andersen et al., 1972; Adams et al., 1997; Deschenes et al., 1998; Sherman and Guillery, 1996), feedback projections from several somatosensory (e.g. SI, SII, PV, etc) cortical areas (Chmielowska et al., 1989; Bourassa et al., 1995) converge on neurons located in primary and secondary thalamic nuclei (e.g. VPM, POM, and ZI) of the trigeminal system of rodents. Studies in mice (Hoogland et al., 1987, 1991) and rats (Bourassa et al., 1995; Deschenes et al., 1994) have shown that these corticothalamic projections terminate primarily in the distal dendrites of these thalamic neurons (Pinault et al., 1997). In the case of the ventral posterior medial (VPM) nucleus, the primary thalamic relay of the trigeminal system, these corticothalamic

projections seem to be organized in a topographic manner (Hoogland et al., 1987; Bourassa et al., 1995; Deschenes et al., 1998; Zhang and Deschenes, 1998) (see Fig. 1). In this arrangement, corticothalamic projections originating from layer VI neurons, which are located under a particular cortical 'barrel' (Woolsey and Van der Loos, 1970) (e.g. barrel D1 in layer IV), terminate on thalamic neurons located across the thalamic barreloids (Van der Loos, 1976) that define the representation of a whisker arc or column (e.g. barreloids A1, B1, C1, D1, El) in the VPM (see Fig. 1). Corticothalamic projections from layer VI also reach the reticular nucleus (RT) of the thalamus (Pinault et al., 1995, 1997), the main source of GABAergic inhibition in the rat VPM (Pinault and Deschenes, 1998). Anatomical evidence suggests that these cortical-RT projections are also organized in a topographic arrangement, which seem to be orthogonal to that observed in the VPM nucleus (Hoogland et al., 1987, 1988). Thus, axons from layer VI neurons located under a given cortical 'barrel' (e.g. C1) target neurons located across the representation of a whisker row (e.g. C1, C2, C3, C4) in the RT nucleus (Hoogland et al., 1987, 1988). Neurons located in secondary somatosensory thalamic nuclei, such as the posterior medial nucleus (POM), also receive corticothalamic terminals (Hoogland et al., 1987, 1991), albeit these are primarily derived from pyramidal neurons located in layer V of the somatosensory cortex. The morphology of corticothalamic terminals also varies according to whether they terminate in the primary (e.g. VPM) or secondary thalamic relay (e.g. POM) nuclei (Hoogland et al., 1988, 1991). Physiological studies have shown that corticothalamic projections are primarily excitatory and likely employ glutamate as their main neurotransmitter (Turner and Salt, 1998, 1999). The glutamate released from these corticothalamic terminal acts on AMPA, NMDA, and metabotropic receptors located in the distal dendrites of thalamic neurons (McCormick and von Krosigk, 1992; Salt and Eaton, 1996; Turner and Salt, 1999). Activation of metabotropic receptors by in vitro stimulation of corticothalamic axons produces long-lasting, slowrising EPSPs in the thalamus (Salt and Turner, 1998; Turner and Salt, 1998). Based on some of these findings, corticothalamic-mediated activation

92 of metabotropic receptors has been suggested to produce the modulation of neuronal firing in the VPM nucleus (Salt and Turner, 1998; Turner and Salt, 1998). For instance, it is conceivable that the slowly rising depolarization produced by activation of corticothalamic projections could allow thalamic neurons to reach firing threshold in the presence of subthreshold synaptic input. In addition, corticothalamic afferents could also contribute to the slow activation of a low-threshold calcium conductance that underlies the production of bursts of action potentials by thalamic neurons (Sherman and Guillery, 1996). Despite a wealth of anatomical, pharmacological, and in vitro physiological information, the role played by corticothalamic projections in tactile information processing has remained elusive. For instance, penicillin-induced epileptic discharge in the cat somatosensory cortex (Ogden, 1960) and cortical spreading depression in the rat cortex (Albe-Fessard et al., 1983) were found to induce a depression of sensory evoked responses in the thalamus. In another series of experiments, carded out in both anesthetized and awake preparations, Yuan et al. (1985, 1986) reported that lidocaine-induced inactivation of SI cortex resulted in reduced thalamic responses to electrocutaneous stimulation without any effect on the spontaneous activity, stimulus threshold, response latency, or receptive fields of the same thalamic neurons. Other studies, however, have reported a facilitatory influence of SI cortex on evoked thalamic discharges (Andersen et al., 1967, 1972) using cortical spreading depression (Waller and Feldman, 1967) or electrical stimulation (Anderson et al., 1964), It is likely that part of the confusion in the literature arises because corticothalamic pathways can mediate both a monosynaptic excitatory and a dysynaptic inhibitory (via RT nucleus) postsynaptic potential in the thalamus. Thus, depending on how the cortex is stimulated or blocked, a variety of facilitatory and inhibitory responses effects could be induced in the thalamus. This hypothesis is supported by the observation that microstimulation of small territories of the SI cortex can lead to a range of thalamic effects, in addition to an overall suppressive influence of thalamic sensory responses, depending upon the relative topographic location of the stimulus and neurons in the ventral posterior nucleus of the thalamus (Shin and Chapin, 1990c).

In our hands, pharmacological block of SI cortical activity by focal infusion of the GABAA agonist muscimol, and the consequent silencing of pools of cortical neurons that give rise to corticofugal projections to the thalamus and brainstem, produced a series of physiological effects in the rat VPM (Krupa et al., 1999). First, we observed that blocking cortical activity altered both the short and long-latency components of the tactile responses of VPM neurons. The end result of these modifications was the demonstration that corticofugal projections contribute to the definition of the complex spatiotemporal structure (Krupa et al., 1999) of the RFs of VPM neurons. These results were obtained by using traditional single whisker stimuli. When more complex tactile stimuli were employed in our experiments, we observed that the ability of VPM neurons to integrate complex tactile stimuli (e.g. multi-whisker deflections) in a non-linear way was also significantly reduced by a pharmacological block of cortical activity. Both supraand sublinear summation of multi-whisker stimuli (Ghazanfar and Nicolelis, 1997) was reduced in these experiments (Ghazanfar et al., 1997). Overall, these findings not only support the hypothesis that corticothalamic projections may mediate both facilitatory and suppressing effects on thalamic neurons, but they also suggest that the action of these corticofugal projections may also depend on the type of tactile stimulus provided to the somatosensory system. As described below, there is direct evidence that the physiological contribution of these descending pathways to tactile information processing may also depend on the behavioral state of the animal. The functional relevance of corticofugal projections in the rat somatosensory system was investigated in studies carried out in our laboratory t o evaluate the contribution of corticofugal projections to the ability of subcortical neurons to express unmasking of novel tactile responses following a peripheral deafferentation (Krupa et al., 1999). This reorganization process, which we dubbed 'immediate or acute plasticity', is known to trigger a system-wide reorganization of the somatotopic maps located at cortical, thalamic, and brainstem levels (Faggin et al., 1997). The most conspicuous effect of this immediate reorganization is the shifting of receptive fields of individual neurons away from the

93 deafferented region due to the unmasking of neuronal tactile responses that were not present before the peripheral block. Interestingly, such unmasking tends to occur almost simultaneously in the brainstem, thalamus, and cortex (Faggin et al., 1997). In a recent series of experiments, we observed that blocking neuronal activity in the infragranular layers of the SI cortex, a procedure that silences the projecting neurons that give rise to corticobulbar and corticothalamic feedback projections, reduces by almost 50% the number of VPM thalamic neurons that exhibit unmasking of tactile responses following a partial and reversible peripheral deafferentation (Krupa et al., 1999). Although plastic reorganization in the VPM nucleus is still observed after cortical inactivation, its spatial extent is reduced significantly. These findings have been confirmed and extended further by the recent demonstration that the immediate, but not the late phase of plastic reorganization in the ventral posterior lateral nucleus (the thalamic relay for somatosensory fibers from the rest of the body), are reduced or eliminated by removal of corticofugal projections (Parker and Dostrovsky, 1999). Further support for the functional relevance of descending corticofugal projections comes from the observation that these projections have been demonstrated to affect the physiological properties of several other subcortical relays of the somatosensory system. For instance, block of neuronal activity in the SI cortex has been reported to eliminate most of the tactile responses of neurons located in the POM nucleus of the thalamus (Diamond et al., 1992). In addition, corticobulbar projections have also been shown to influence the physiological properties of neurons located in the brainstem nuclei that relay ascending somatosensory information to the thalamus (Jacquin et al., 1990b). For instance, removal of corticofugal projections in rats increases the responsiveness of neurons in spinal trigeminal brainstem complex to whisker stimuli (Jacquin et al., 1990b). Overall, the results reviewed above make a compelling case for the need to incorporate recurrent corticofugal projections as an integral part of a comprehensive and realistic model of touch. Indeed, the recurrent nature of the somatosensory system further strengthens our hypothesis that the mammalian somatosensory system relies on highly distributed neu-

ronal interactions, which emerge from the dynamic interplay of multiple ascending and descending pathways, to represent tactile information (Nicolelis et al., 1993b, 1995, 1996, 1997, 1998a,b; Nicolelis and Chapin, 1994; Nicolelis, 1996). Although the concept of distributed processing is not new, and many investigators have proposed schemes based on population coding (Hebb, 1949; Erickson, 1968, 1986; Georgopoulos et al., 1986; Sejnowski et al., 1988; Mumford, 1992; Deadwyler and Hampson, 1997), this encoding scheme has recently attracted the attention of neuroscientists because of the successful application of artificial neural networks in pattern recognition problems (Grossberg, 1976; Grossberg, 1988; Bishop, 1995. In a distributed coding scheme, divergent neural connections ensure that specific units of information are not held in single or small groups of neurons, but instead are widely distributed, or 'encoded' by large neural ensembles located at multiple cortical and subcortical levels of the system (Hebb, 1949). Consequently, each neuron contributes in some way to processing of most of the information handled by the network. In line with this hypothesis, a series of studies in our and other laboratories (Nicolelis et al., 1993a, 1995, 1998b; Kleinfeld and Delaney, 1996; Masino and Frostig, 1996; Moore and Nelson, 1998; Ghazanfar and Nicolelis, 1999; Polley et al., 1999) have begun to re-examine traditional views of information encoding by the somatosensory system. Anatomical evidence in favor of a distributed model includes the fact that ascending feedforward (FF) somatosensory pathways that carry information from the periphery to the SI cortex exhibit different degrees of divergence (Lu and Lin, 1986; Rhoades et al., 1987; Chmielowska et al., 1989; Jacquin et al., 1990a; Lin et al., 1990; Chiaia et al., 1991; Lu and Lin, 1993; Pinault and Deschenes, 1998; Veinante and Deschenes, 1999), which contribute to the large multi-whisker RFs observed in the VPM and SI (Nicolelis and Chapin, 1994; Ghazanfar and Nicolelis, 1999). Thus, the effects of even small but incremental changes at each processing level of the pathway (e.g. from brainstem to thalamus) would tend to multiply through successive relays and could be markedly amplified by the time they reached the cortex. In addition, wide-field sensory inputs, such as high-threshold mechanical and noxious stimuli,

94 which are transmitted through paralemniscal pathways, could also converge on cortical neurons. These effects could be further amplified by corticocortical connections within the SI and between the SI and other cortical areas (Chapin et al., 1987; Fabri and Burton, 1991; Nicolelis et al., 1991). In this context, the existence of massive divergent corticofugal feedback projections to all subcortical somatosensory relay nuclei provide almost unlimited opportunity for increasing the ultimate radius of influence from a single sensory event (Mumford, 1991). In this model, single neurons would not serve as the functional unit of the system. Instead, neurons would work as part of ensembles that are capable of representing and processing multiple tactile attributes of a given complex stimulus simultaneously. Massive corticofugal projections, that reach somatosensory relay structures located in the thalamus, brainstem, and the spinal cord, could offer the anatomical substrate for the definition of such multitasking networks. Such distributed and recurrent networks could be formed by somatosensory, motor, limbic, and association cortical areas, and influence the activity of neurons located in subcortical centers, even before mechanoreceptors in the skin were activated by a tactile stimulus. According to this view, corticofugal feedback projections could incorporate subcortical nuclei into the computational processes required for the emergence of tactile percepts. Although rarely discussed in the literature, reciprocal loops between cortical and thalamic nuclei could also mediate a different type of corticocortical communication, in which thalamic networks combine convergent signals from one or more cortical areas and then disseminate the resulting signals to vast cortical territories. Such an interactive view of the somatosensory system would predict that top down-influences would be capable of modulating the activity of subcortical neurons during different behavioral states. But is there evidence for the existence of such top-down influences in the somatosensory system? In the next section we describe a well-known phenomenon that may provide the key for unraveling the physiological role played by corticofugal feedback on tactile perception and, hence, serve as the basis for mounting a formidable challenge to the FF model of touch.

A potential physiological role for corticothalamic pathways in tactile processing: sensory gating of neural responses during active tactile exploration A number of studies carried out in many species indicate that during different exploratory behaviors the magnitude and latencies of tactile responses as well as the manner in which the brain responds to complex tactile stimuli, can change considerably. Thus, in rats, reductions in responses to tactile stimuli during motor activity have been observed in SI (Chapin and Woodward, 1981, 1982a,b; Shin and Chapin, 1990b), the ventral posterior lateral thalamus (VPL) (Shin and Chapin, 1990a,b), and the dorsal column nuclei (DCN) (Shin and Chapin, 1989). Similarly, in cats, medial lemniscus sensory responses elicited by stimulation of the radial nerve are reduced in magnitude during limb movement (Ghez and Lenzi, 1971; Coulter, 1974). Primates also show modulations in SI (Nelson, 1984, 1987; Chapman et al., 1988) prior to and during motor movement. Alterations in sensory responses during movement have also been observed in human, evoked potential studies (Coquery, 1971; Lee and White, 1974; Cohen and Starr, 1987). These observations imply that the nervous system is capable of dynamically altering how cortical and subcortical neurons respond to a tactile stimulus, depending on the behavioral context in which such a stimulus is presented to the animal. The crux of this argument, therefore, lies in the hypothesis that the emergence of the broad spectrum of natural tactile sensations experienced by mammals results from a much more intimate association between the somatosensory and motor systems than postulated before by previous neurophysiological theories of touch. But what is the significance of endowing the somatosensory system with the capability of altering the type of tactile information that can reach the cortex during motor activity? First, the alterations in response magnitude may allow tactile and proprioceptive information pertinent to execution or completion of the movement to be selectively enhanced. Thus, during the execution of a planned motor act, reciprocal interactions between the motor and the somatosensory cortices would ensure that certain types of input are gated 'in' while others are gated 'out'. This may be necessary in order to allow

95 the movement to occur as planned without interference from extraneous sensory feedback. In support of this idea, Chapin and Woodward (Chapin and Woodward, 1982a) showed that tactile responses, across the cortical and subcortical relays of the somatosensory system, can be inhibited or enhanced at different epochs of the step cycle in locomoting rats. According to these authors, irrelevant tactile responses, such as those caused by the movement itself, would be selectively gated out at certain times during the movement, while sensory information that would describe certain movement epochs (e.g. foot fall) would be enhanced. There is strong evidence in the literature supporting the hypothesis that this selective modulation of tactile responses is mediated by descending efferent activity from the motor cortex or other central motor nuclei. The most relevant experimental finding supporting the occurrence of centrally mediated gating of tactile information is the observation that sensory responses across the somatosensory pathway can be reduced as much as 100 ms prior to the initiation of a given movement. These findings, which have been obtained in both monkeys (Nelson, 1987; Chapman et al., 1988) and cats (Coulter, 1974; Ghez and Lenzi, 1971) strongly suggest that reductions in tactile responses in the cortex and subcortical relays of the somatosensory system do not result from alterations in tactile or proprioceptive feedback generated by the movement itself. Instead, they may be related to the central motor command that is generated hundreds of milliseconds prior to the movement onset. Further support for this view comes from the observation that microstimulation of the motor cortex can reduce tactile neuronal responses throughout the somatosensory system. For example, Shin and Chapin showed that stimulation of the forepaw region in the MI cortex, prior to electrical stimulation to the forepaw, led to a 43% suppression of tactile responses in the thalamus (Shin and Chapin, 1990b) and 8% reduction in the dorsal column nuclei (Shin and Chapin, 1989). Another set of experiments has demonstrated that the amount of tactile response modulation varies across the different intermediary relays of the ascending somatosensory pathways. For instance, in rats, SI and VPL tactile responses to forepaw stimulation have been shown to decrease 71 and 31%, respectively, when the stimuli are delivered during

the animal's locomotion (Chapin and Woodward, 1981; Shin and Chapin, 1990b). These findings illustrate the general observation that the magnitude and frequency of the somatosensory gating effect tends to increase as one ascends through the somatosensory system. As such, this observation further supports the hypothesis that modulatory signals derive from central neural networks responsible for generating the motor command. Despite robust evidence favoring the existence of a central mechanism for modulating tactile neuronal responses, in some cases one can also demonstrate that feedforward mechanisms contribute to the alteration of cortical and subcortical tactile responses during the execution of exploratory behavior. For example, Schmidt et al. (1990b) have shown that when anesthesia was applied to one or more sensory nerves of the hand, inhibition of tactile stimulation that normally occurred when subjects moved the finger being stimulated, was reduced (i.e. there was less gating of the response) by up to 70%. This led to the conclusion that a significant portion of the gating effect, but not all, was caused by afferent sensory stimulation generated in the periphery by the movement itself. Support for the existence of peripheral mechanisms of gating has also been provided by Chapman et al. (1988), who reported that tactile responses in SI to stimulation of the medial lemniscus or the thalamus were not reduced prior to movement, but only during the execution of the movement. These authors reported that somatosensory gating occurring at the level of the dorsal column nuclei (DCN) was caused by central modulation, since it occurred prior to the movement. However, in their hands any additional gating at higher levels of the somatosensory system was caused by motor-induced peripheral afferent activity. Another series of experiments partially supported this view by showing that while passive movements do not cause tactile gating in the DCN, they can induce a certain degree of gating in the thalamus (VPLc) and SI (Chapman et al., 1988). Thus, even though there is substantial evidence for centrally mediated modulation of tactile responses, one cannot discard the possibility that proprioceptive and tactile afferent signals, generated during the execution of movements, contribute to the gating of tactile responses observed at higher levels of the somatosensory system.

96 It is important to emphasize that noradrenergic, serotoninergic, and cholinergic projections, which originate in different locations of the brainstem and diencephalon and target all intermediary relays of the somatosensory system, could also contribute to the central modulation of tactile neuronal responses during different behavioral states (McCormick and Pape, 1990; Waterhouse et al., 1994) as they do in other sensory systems (McLean and Waterhouse, 1994). Over the last three decades, one of the most elegant examples of multi-disciplinary research in neuroscience has indicated that some of these modulatory systems play a fundamental role in the control of the ascending flow of nociceptive information from the periphery that is used for the perception of pain (Fields and Heinricher, 1985). The demonstration that physiological or pharmacological activation of these descending modulatory projections can block the ascending flow of nociceptive information through the spinothalamic system and produce maintained analgesia has revolutionized our understanding of pain perception. Since pain belongs to the spectrum of tactile sensation that all mammals experience, these observations offer more experimental support for our contention that top-down influences cannot be ignored by any theory aimed at describing the neurophysiological basis of tactile perception. In line with this hypothesis, recent studies in the trigerninal system of awake, freely moving rats have corroborated and extended our conviction that top-down influences, such as those mediating the phenomenon of 'somatosensory gating', play a crucial role in the emergence of tactile perception. In these experiments, simultaneous, multi-site chronic recordings were employed to monitor the activity of large populations of single cortical, thalamic, and brainstem somatosensory neurons, while rats moved freely in a behavioral box. Initially, these experiments allowed us to investigate how the expression of different behaviors (e.g. awake immobility, active whisking, moving without whisker movements) could influence the physiological properties of populations of cortical and subcortical neurons in freely behaving animals (Fanselow and Nicolelis, 1999). Subsequently, the same experimental paradigm was used to measure how similar tactile stimuli are processed under different behavioral conditions across the rat somatosensory system.

We observed that complex and dynamic corticothalamic interactions tend to precede any active tactile discrimination in freely behaving rats. For instance, as awake rats assume an immobile posture (i.e. standing on the four paws without producing any whisker or other major body movements), most of the neurons in the SI cortex and VPM thalamus start producing rhythmic bursts of action potentials, which are translated into 7-12 Hz rhythmic oscillations (Nicolelis et al., 1995; Fanselow and Nicolelis, 1999). In the vast majority of the analyzed events, these 7-12 Hz rhythmic oscillations initiate in the whisker area of the rat SI cortex (the barrel fields) (Fig. 2). After a few tens of milliseconds, these osi

i

i

i i

S, CORTEX S, CORTEX

't

VP~

....

VPM

~A~^A~^,dAik,

1 . . . . ^At,.d~A~,AAAAI.^~A, dAA~,

1 SEC I

i

i[

I t iii i

i

!

CORTEX

SI CORTEX VPM

~

~

THALAMUS

t lilttnm~

t

t

t

,

,

tt|S~

c

t

I

i

Sl CORTEX ~ ! S[ CORTEX

VPM THALAMUS

~ I

THALAMUS

1 SEC Fig. 2. Simultaneous field-potential recordings in VPM and SI during Ix rhythm reveal that 7-12 Hz oscillations tend to appear first in the SI cortex and only later are evident in the sornatosensory thalamus.

97 SpV 1

10 0 10 0

VPM

1

A

20 A

Alk

0 .'-",."T'",'"',"'-,"-'r-",'","r"T'"r'",

2°I

A

A

A

,o

2oo 0"

0 r...r...l-..-r.-.r-.-r-.T--.r-..r-.-,..-T...r.-.~

~.

°

30

t

r

40

i

0 |....r.-.r-.:,-.--,-.-.r.-T.--,----,.-..r--T...,...., I

30

7--r--l---T--T---r--

i

-

-

0 ....,...T...,...-,....,...T...r...r_r..T...r.. , Sl 50 0 5

10"

O-

O.

s~

1o I

Io

0 T'"'r "'r"'Ilt'lr'"~'"'l'"'r'l'll'"'r"T"T "''l

0.1

0

0.1

0

0.1

0

0.1

Tlmo lag (s) Fig. 3. During whisker twitching movements, activity in SI and SpV phase leads activity in VPM. Recordings were made from microelectrodes chronically implanted in SpV, VPM and SI in awake rats. This figure depicts cross-correlograms (CCs) for neurons in these three areas during 7-12 Hz whisker twitching movements, which were accompanied by i~-oscillations. The CCs are centered on the VPM neuron depicted by the arrow, and the numbers above the CCs show the number of milliseconds by which a given SpV or SI neuron phase-led the VPM reference neuron. It can be seen that activity in SI and SpV phase led activity in VPM.

cillations appear in the VPM nucleus and later on they can be observed in the spinal nucleus (but not in the principal) of the trigeminal brainstem complex (Figs. 2 and 3). Importantly, these oscillations were never detected in the rat trigeminal ganglion, suggesting that they are generated centrally. Further analysis revealed that these 7-12 Hz thalamocortical oscillations usually precede, by hundreds of milliseconds, the initiation of small amplitude rhyth-

mic facial whisker twitching (WT) movements in the same frequency range (Nicolelis et al., 1995). We also observed that the initiation of WT movements modulated these oscillations (Nicolelis et al., 1995). Thus, soon after the onset of WT movements, rats invariably started to produce slower (4-6 Hz) rhythmic whisker protractions, which had much larger amplitudes than the WT movements. Previous behavioral studies have indicated that rats use these large rhyth-

98 mic whisker movements to discriminate the tactile attributes of objects (Carvell and Simons, 1990). In fact, 'whisking' is present in most rodents and is considered as an important exploratory behavior of rats. In our experiments, we also documented that as soon as the animal started to produce these slower and larger whisker movements, the 7-12 Hz thalamocortical oscillations disappeared (Nicolelis et al., 1995). Altogether, these observations are very reminiscent of a similar phenomenon originally described by Gastaut in human scalp EEG recordings carded out in the 1950s (Gastaut, 1952). Since its original discovery, both EEG and magnetoencephalographic recordings have been used to demonstrate the occurrence of widespread 10 Hz oscillations, originating in the hand representation of the primary somatosensory cortex of the vast majority of healthy human subjects and non-human primates (Niedermeyer, 1993). In the EEG literature, these oscillations were named motor (Ix) rhythm, since its main characteristic is to appear during awake immobility and disappear as soon as the subject starts any hand movement, a key tactile exploratory behavior of primates. The existence of these similarities led us to postulate that the 7-12 Hz oscillations that precede and are modulated by whisker movements in rats are equivalent to the Ix rhythm of primates. The functional role of Ix oscillations in primates and rodents is still unclear. Although MEG recordings have clearly indicated that this preparatory rhythm is present in most normal human subjects, no clear consensus has been reached regarding the potential physiological role played by these oscillations. Because rats use rhythmic whisker movements as their main tactile exploratory behavior, the presence of the IX rhythm in this species led us to propose that these thalamocortical oscillations could prepare the somatosensory system for the imminent onset of a cycle of tactile exploration. According to this hypothesis, during the occurrence of 7-12 Hz oscillations, tactile information would continue to flow from the VPM to the somatosensory cortex. However, since during these oscillations a significant percentage of VPM neurons are producing bursts, it is conceivable that these neurons would have difficulty in faithfully transmitting complex spariotemporal patterns of tactile information, which are likely to be generated when rats use their whiskers to actively explore an ob-

ject. Instead, we proposed (Nicolelis et al., 1995) that these 7-12 Hz oscillations could be used to enhance or even maximize the ability of the somatosensory system to detect the presence of tactile stimuli, either during awake immobility or during the production of whisker twitching movements. In other words, these 7-12 Hz oscillations could represent an 'expectation' signal, a template that could be produced by the rat somatosensory system in anticipation of whisking. In our view, this 'expectation' signal could be generated as part of central motor program and be disseminated to most of the somatosensory system through corticofugal projections. We speculate that once the presence of a tactile stimulus is detected by an immobile rat (or during WT movements), larger amplitude and slower (46 Hz) rhythmic whisker protractions are initiated so that a more detailed tactile exploration of objects by the arrays of vibrissae can be accomplished. As the animals behavior changes, so does the physiological setting of the thalamocortical loop. During the execution of these rhythmic large amplitude whisker movements, thalamocortical oscillations vanish and VPM neurons switch to a tonic firing mode. In this physiological state, VPM neurons have higher spontaneous firing rates and can faithfully represent and transmit complex spatiotemporal patterns of tactile inputs to the SI cortex. It is important to emphasize, however, that during this active tactile exploration, corticothalamic projections can still mediate important interactive computations at the level of the thalamus. In this interactive view of the somatosensory system, the thalamus is no longer considered as a simple, passive relay of tactile information from the periphery to the cortex. Instead, through its reciprocal interactions with different cortical areas, the somatosensory thalamus, (which includes the tactile portion of the reticular nucleus), could participate in a variety of computations, such as non-linear summarion of tactile stimuli (Ghazanfar and Nicolelis, 1997; Shigemi et al., 1999), signal segmentation through resonant interactions, template matching, and error generation. Similar to the recurrent models of Grossberg (1999) and Mumford (1994), the somatosensory thalamus could function as the site where incoming afferent tactile information is compared with cortically stored templates that resume the previous tactile experience of the animal. As feedfor-

99 the same animals engaged in behaviors that did not include the whisker movements (e.g. movement of the head or body). These findings corroborate work in cats (Coulter, 1974), and in humans (Schmidt et al., 1990a Schmidt et al., 1990b), in which decreases in sensory responsiveness were most robust when the sensory stimulus was applied to the part of the body engaged in a tactile exploration, as compared to adjacent digits or contralateral limbs. Thus, as previously suggested by many authors, the phenomenon of somatosensory gating appears to be fairly topographically specific, since it occurs only during motor activity used for active tactile exploration, rather than following any action involved in increasing the general arousal level of the animal. Previous studies have shown that the presence of one stimulus can alter the ability of cortical and subcortical neurons to respond to a subsequent stimulus for a period of time (Simons, 1985; Simons and Carvell, 1989). Evidence from our experiments and other studies (Castro-Alamancos and Connors, 1996a,b) suggest that the ability of one tactile response to modulate the magnitude of a subsequent one is substantially decreased during motor activity (Fig. 5). For example, when two tactile stimuli were presented with an inter-stimulus interval of 2575 ms in the absence of any whisker movements (i.e. awake immobility), the response to the second

ward and feedback projections may be required for the definition of the complex spatiotemporal structure of receptive fields in the rat VPM (Krupa et al., 1999), ensembles of these neurons could participate in template-matching operations and other computations on afferent tactile signals. In order to test some of these assumptions, another series of experiments was carded out in our laboratory. In these experiments, multi-site chronic recordings were carded out while a nerve cuff electrode was used to provide consistent stimulation to the infraorbital (IO) nerve, the nerve that carries tactile information from the vibrissae to the central nervous system, as rats switched between a series of behavioral states (Fanselow and Nicolelis, 1999). In the first series of experiments, individual electrical stimuli that produced neuronal responses that mimic those obtained by mechanical stimulation of multiple facial whiskers were delivered to the IO nerve while rats were immobile, or when they produced the two different types of whisker movements described above. The cortical (SI) and thalamic (VPM) sensory responses elicited by the electrical stimuli were then compared. As predicted by previous studies, the magnitudes of the neural responses in SI and VPM neurons were substantially reduced during the production of rhythmic whisker movements (Fig. 4). Interestingly, this reduction was not observed when

VPM

SI

20

o~,~-

0

~ 20 ~

o

-50

0

w

i

20 50

100

150

200

-50

0

50

100

150

200

ms post stimulus

Fig. 4. Activity levels in VPM and SI following peripheral stimulation differ depending on the behavioral state of an animal. Individual electrical pulses were presented to the infraorbital nerve in awake, freely moving rats and responses to this stimulation were recorded from chronically implanted microwires in VPM and SI. When the animal was in a state of quiet immobility, the initial excitatory response was followed by a period of suppressed firing, during which activity went below pre-stimulus baseline levels (dotted line). This period of suppressed firing was followed by a late excitatory component at approximately 125 post-stimulus. In contrast, during exploratory whisking behavior, the period of suppressed firing was substantially shorter in VPM and non-existent in SI, and there was no late excitatory component in either area. Error bars represent ±SEM. The initial excitatory peaks have been clipped in order to show the other components of the traces more clearly.

100

VPM Quiet

Whisking

150

100 or)

5O

to

~r

|,iiliii|

0 lstl25 50 75 100 t25 150 175200 I stimulus" time to second stimulus in ms

lstl25 50 75 100 125 150 175200 stimulus time to second stimulus in ms

E T'-

Sl Quiet

Whisking

if}

150 0c-

'OSO0 Oi ,,i/i i lstl25 50 75 100 125 150 175200 stimulus time to second stimulus in ms

I,iiiiil!

lstl25 50 75 100 125 150 175 200 stimulus ' time to second stimulus in me

Fig. 5. Responses to peripheral stimulation differ depending on the behavioral state of an animal. Recordings were made from multiple chronically implanted microwires in awake, freely moving rats. Stimulation was provided to a nerve cuff electrode implanted around the infraorbital nerve. Stimuli were presented in pairs with interstimulus intervals (ISI) ranging from 25-200 ms. This figure demonstrates two effects we observed by looking at the responses to these stimuli during two different behaviors, quiet immobility and exploratory whisking. First, during whisking, responses to the first stimulus in the pairs were smaller than those during the quiet state. This indicates that during the whisking state there is an overall gating of responses to ascending stimuli. The second effect was that when animals were in a state of quiet immobility, responses to the second stimulus in a pair was suppressed if the ISI was 25-75 ms. In contrast, during active, exploratory whisking behavior, the responses were not significantly suppressed for any ISI, compared to the response to the first stimulus in the pair. Error bars represent +SEM.

stimulus was significantly reduced (Fig. 5). However, during periods in which the same rats produced whisker movements, the response of VPM and SI neurons to the second stimulus was not statistically

different in magnitude from the first at any interstimulus interval tested (Fig. 5). Further examination of these results indicated that these effects paralleled a change in the amount of post-excitatory inhibition

101 that follows the first IO stimulus in different behavior states (immobility vs. whisker movements). Thus, in the absence of any movement of the whiskers, we observed the occurrence of a long period of reduced firing following the presentation of the first tactile stimulus (Hellweg et al., 1977; Simons, 1985; Simons and Carvell, 1989). However, this period is substantially shorter (in VPM) or non-existent (in SI) during the presence of exploratory whisker movements (Fig. 5), suggesting that motor-activity related changes in post-stimulus inhibition could account for the differential responses to paired stimuli we observed in different behavioral conditions. Overall, the results of these experiments suggest that during different behavioral states, different types of thalamocortical transmission may occur (in this case, awake immobility versus 'whisking') and that these different modes of transmission may serve different perceptual purposes. Thus, differences in cortical and subcortical tactile response characteristics, from periods of whisker immobility to periods of whisker movements, suggest that the somatosensory system can shift from a state of high-sensitivity for detecting individual punctate stimuli (i.e. during immobility and thalamic bursting), to a state in which the system can process with high-fidelity the complex incoming tactile afferent information that are generated by the active tactile exploratory behavior employed by the animal to probe its surrounding environment. Although there is no definitive proof that corticofugal projections are responsible for either the recruitment of VPM neurons into 7-12 Hz oscillations during awake immobility, or the switch of VPM neurons from bursting to tonic firing mode, several indirect observations can be used to build a strong case in favor of this hypothesis. First, in the vast majority of our recordings, IXoscillations clearly initiate in the SI cortex and only later appear in the VPM thalamus. Likewise, in all other species in which the Ix rhythm has been reported it was found to originate at the cortical level. The hypothesis that corticofugal projections provide the anatomical substrate for recruiting the thalamus into a massive wave of synchronous activity is also supported by the observation that removal of corticothalamic projections significantly reduces or completely abolishes the synchronization of neuronal firing across the thalamus (Contreras et al., 1996).

The potential contribution of corticothalamic projections to the switching in firing mode of thalamic neurons is also supported by several indirect observations. Since corticothalamic axons terminate in the distal dendrites of VPM neurons, and exert their direct excitatory effects through metabotropic receptors, they could provide the type of slow depolarization synaptic events that are required for activating the low-threshold calcium conductance that endows thalamic neurons with the ability to fire in bursts. De-inactivation of this calcium conductance, which requires hyperpolarization of VPM neurons, could also be achieved by corticothalamic projections acting through the reticular nucleus which provides GABAergic innervation to thalamic relay neurons. Though many more experiments are required to fully demonstrate the computations carded out by the interplay of corticofugal and ascending somatosensory pathways, the central assumption of our argument remains valid. The type of dynamic thalamocortical interactions described above cannot be explained by a simple feedforward description of the somatosensory system. As seen above, changes in behavioral state significantly alter the responses to tactile stimuli across cortical and subcortical levels of the somatosensory system. These studies have demonstrated that neuronal response properties can be altered on the order of seconds, as animals switch from one behavioral state to the next. The possibility of altering the manner in which somatosensory neurons respond to the same tactile stimuli under different circumstances confers a high degree of adaptability to the animals since it may allow them to filter information in different ways, as required by the situation in which they are involved. This rapid, behavior-dependent adaptation may also provide the somatosensory system with more flexibility for detecting a wider range of stimuli, or allow preferential detection of certain types of stimulation under different circumstances. Corticocortical loops as the substrate for the integration of bilateral whisker information in the rat barrel cortex The second loop investigated in our studies of the rat somatosensory system is the one defined by reciprocal callosal connections between both rat SI cortices,

102 The experimental evidence reviewed in this section suggests that this loop plays a fundamental role in integrating left and right side whisker information required for the formation of bilateral tactile percepts of the environment. In these studies, the role the cerebral cortex plays in the integration of bilateral tactile information was investigated by inferring from extracellular recordings the temporal and spatial transformations performed by cortical neurons on convergent subcortical and corticocortical input. Study of cortical processes in the barrel cortex is facilitated in part by an ability to exploit an orderly topography of connections found throughout the whisker-barrel axis that reflects the arrangement of contralateral whiskers at the periphery (Woolsey and Van der Loos, 1970; Killackey, 1973). It is upon this anatomical topography that the classical hypothesis regarding barrel cortical function emerged, which postulates that activity within a given barrel cortical column directly relates to the attributes of a stimulus applied to a corresponding contralateral whisker. Incorporating more recent findings, contemporary theories of barrel cortical function have evolved to emphasize the role the barrel cortex plays in integrating information across whiskers to form behaviorally relevant percepts, rather than in extracting information from individual whiskers. Experiments using condition-test paradigms have provided a basic understanding of the temporal and spatial nature of integration between barrel regions corresponding to pairs of whiskers (Simons, 1985; Simons and Carvell, 1989; Brumberg et al., 1996). These experiments have provided evidence that excitation of a region of the barrel cortex is followed by a prolonged period of inhibition, which attenuates over time and acts to diminish the probability of a second, stimulus-evoked response. The magnitude of inhibition appears to be maximal in the region corresponding to the whisker stimulated, with the spatial distribution of inhibition decreasing as a function of distance from this center. However, such studies are limited by the fact that observed cortical responses are not solely functions of cortical processes, but instead, also may reflect the processes of subcortical structures along the whisker-barrel axis. Whatever the nature of subcortical integration along the whiskerbarrel axis, corticocortical integration may still be addressed by exploiting the role the corpus callo-

sum plays in integrating sensory information that is lateralized subcortically. In this context, if the rat is to create a perception of the environment regarding both sides of its face if it is to generate an appropriate behavioral response to bilateral stimuli - - then this information must so too be integrated. That rats are capable of navigating complex terrain even in absolute darkness by virtue of their whiskers seems to dictate that they make comparisons between groups of whiskers. As rats bilaterally and synchronously whisk objects they encounter, to successfully detect the orientation of an obstacle or the width of an aperture they must gather and integrate information regarding the distance of objects from the whisker pads to either sides of the face. The most logical place to identify and characterize how bilateral information is integrated, therefore, is at the level left and right side whisker information first converge (Fig. 6). As ascending whisker-related pathways are fully crossed subcortically, this conver-

-

Barrel Cortex

Barrel Cortex

W h i s k e r Pad

Fig. 6. Schematicof whisker-barrel anatomyillustrating convergence of ascending contralateral, whisker-relatedpathways with ipsilateral, whisker-relatedcallosal pathways. Neurons innervating the whiskerpad terminate in the principle trigeminalnucleus (PrV) and spinal trigeminalnuclei (SpV). Projectionsfrom these nuclei then decussate and terminate in two thalamic nuclei: the ventroposteriormedial nucleus (VPM) and the posteriormedial nucleus (PoM), which in turn project to the barrel cortex. The barrel cortex also receivescallosal input regardingthe ipsilateral whisker pad from a portionof neurons in supra- and infragrandular layers of the oppositebarrel cortex.

103 gence is thought to occur at the level of the SI barrel cortices, as they are interconnected via the corpus callosum in a roughly homotopic manner (White and DeAmicis, 1977; Koralek et al., 1990; Olavarria et al., 1992; Cauller et al., 1998). This presumption is also supported by evidence collected by Pidoux and Verley (1979), who provided local field potential recordings indicating that ipsilaterally evoked responses exist in the barrel cortex and are mediated by the corpus callosum. In support of this view, we have recently provided direct evidence that indicates layer V barrel cortical neurons integrate not only contralateral whisker information, but whisker information from both sides of the face, and further postulate that such interactions underlie the formation of bilateral tactile percepts. By combining methods for creating bilateral, multi-whisker stimuli with multi-electrode recordings we addressed whether single barrel cortical neurons respond to both contra- and ipsilateral whisker stimuli, characterized ipsilaterally evoked response properties, and determined the spatial and temporal aspects of interaction evoked by bilateral whisker stimulation (Shuler et al., 2001). Pharmacological inactivation of the opposite barrel cortex corroborated the proposition that the source of ipsilateral input is the opposite barrel cortex. We further demonstrated that during inactivation, not only were ipsilaterally evoked responses abolished, but the ensuing inhibitory influence such responses exerted on subsequent contralaterally evoked activity was also shown to be removed. Lastly, by designing a behavioral task that requires the cooperativity of the barrel cortices, we determined that rats are, in fact, capable of forming bilateral tactile percepts. What constitutes an effective ipsilateral whisker stimulus was addressed by varying the number and location of whiskers stimulated ipsilateral to cortically implanted electrode arrays. The arrangement of 16, independently drivable whisker deflectors allowed single whiskers, as well as all possible combinations of two, three, and four whiskers to be stimulated in each of four whisker columns (or arcs) tested. The result of such a stimulus regime was the characterization of ipsilaterally evoked responses in 72% of neurons recorded, with an average probability of evoked response of 21.8 ± 13% (mean -tSD). Compared to an 11 ± 3 ms minimal latency

for responses elicited by contralateral stimuli, the average minimal latency for a response elicited by an ipsilateral stimulus was 23 -4- 4 ms. Numerous instances of 'supra-linear' responses were detected as combinations of simultaneously deflected whiskers frequently were capable of eliciting ipsilateral responses when no responses were elicited from the individual whiskers that defined the ipsilateral stimulus. Although response probabilities were also found to increase as the number of whiskers deflected increased, this increase was decidedly sublinear for neurons that responded to the constituent parts of the stimulus when given alone. Therefore, not only is the barrel cortex responsive to ipsilateral stimulation, but the proportions of neurons and their underlying firing probabilities are nonlinearly effected by multi, ipsilateral whisker stimuli. Given the presence of ipsilateral responses in the barrel cortex, we next addressed the impact such responses may have on contralaterally evoked activity, and vice versa. The nature of bilateral whiskerevoked interactions within the barrel cortex was investigated by using a condition-test paradigm that varied the spatiotemporal attributes of left and right side whisker stimuli (Fig. 7). Three parameters of bilateral whisker stimuli were varied: (1) the hemispheric sequence (ipsi- then contralateral, or contrathen ipsilateral stimulation); (2) the inter-stimulus interval (ISI); and (3) the spatial location of condition stimuli. Variation of these three factors tested the null hypotheses that the hemispheric sequence, ISI, and spatial relationship between bilateral whisker stimuli do not change the firing probabilities of barrel cortical neurons as compared to responses evoked when ipsi- or contralateral stimuli are given alone. These attributes of bilateral stimuli were all shown to significantly impact the evoked response probabilities of barrel cortical neurons, demonstrating that the barrel cortices are capable of integrating bilateral whisker information. Furthermore, the change in response probability caused by bilateral interactions could not be explained by postulating that barrel cortical neurons simply were less likely to fire to the test stimulus on trials that neurons had fired to the condition stimulus. This result indicates that ipsilaterally, as well as contralaterally evoked suprathreshold activity is followed by an epoch of inhibition of an even greater spatial extent. This

104 conclusion is further supported by noting that even for neurons without identifiable ipsilateral responses, contralaterally evoked responses to test stimuli were significantly impacted by prior ipsilateral stimulation. Therefore, bilateral interactions give rise to hemispheric and spatial differences in recovery of subsequent responses, potentially reflecting a differential activation of inhibitory networks. To determine the source of ipsilateral input, the opposite barrel cortex was pharmacologically inactivated by infusion of the GABAA agonist, muscimol, prior to bilateral stimulation. Not only did inactivation remove ipsilaterally evoked responses in the intact hemisphere, but the observed effects of prior ipsilateral stimulation on contralaterally evoked responses were also negated. Such results indicate that the barrel cortices provide one another with ipsilateral whisker information. Considering that callosal connections are thought to be excitatory, these results suggest that ipsilaterally evoked suprathreshold activity subsequently activates local inhibitory networks, rather than indicating that such inhibition derives from yet another source. A central question raised by these physiological results is whether rats can make use of bilateral tactile cues to discriminate objects. Surprisingly, though a number of behavioral studies have address how rats use their whiskers to discriminate tactile features of the environment (Hutson and Masterton, 1986; GuicRobles et al., 1989; Barneoud et al., 1991; Pazos et al., 1995; Brecht et al., 1997), no study to date has directly addressed the ability of rats to compare bilateral tactile features. To test the hypothesis that rats can form bilateral percepts, we developed a discrimination task in which rats learn to compare the relative distance of two walls, one to each side of the face, using only the facial whiskers. Of eight rats trained on this task, all learned to associate equidistant or non-equidistant bilateral stimuli with

water reward made available at one of two reward windows, respectively. These results provide the first evidence that rats can indeed combine information from both whisker pads (Shuler et al., 2000). The anatomical constraints of the whisker-barrel system makes it ideal for studying cortical integration of independent sources of sensory input. We exploited this anatomy by investigating cortical integration of contralateral whisker-evoked activity ascending via thalamocortical pathways with that of callosally converging ipsilateral activity. Ipsilaterally evoked responses, as well as interactions due to the hemispheric, temporal, and spatial attributes of bilateral stimuli strongly contradict the classical notion that the rat barrel cortex solely represents stimuli delivered to the contralateral whisker pad. Such interactions evoked by multi-whisker stimuli do not support the hypothesis that cortical profiles of activity result from the topographic, linear superposition of individual responses. Furthermore, such interactions cannot be explained by postulating the existence of superimposed contra- and ipsilateral topographic maps. To propose such a coding scheme, a countless number of topographies would be required to uniquely describe all possible permutations of bilateral whisker stimuli. Rather we propose that ascending thalamic as well as converging transhemispheric input differentially excite the barrel cortex, subsequently initiating a wake of spreading inhibition. Spatial and temporal asymmetries in activating excitatory and inhibitory elements of the network result in spatiotemporally unique profiles of cortical activity, allowing each hemisphere to render unambiguously the attributes of a bilateral stimulus. In conclusion, we propose that such bilateral interactions are fundamental to forming bilateral tactile percepts, allowing rats to discriminate ethologically meaningful stimuli, such as the orientation and diameter of apertures.

Fig. 7. Evoked responses of two, layer V barrel cortical neurons to bilateral whisker stimulation. Simultaneous, single unit recordings from neurons in the left (neuron 1) and fight (neuron 2) hemispheres were obtained while stimulating left and fight whisker arcs (whiskers b3, c3, d3, e3). Solid vertical lines centered at time 0 denote delivery of test stimuli, while dashed vertical lines denote delivery of condition stimuli. Six stimulus conditions are shown for neurons 1 and 2. The top two rows depict responses to the test stimuli (left whisker arc, L; fight whisker arc, R) when given alone. The bottom four rows depict responses under condition-test stimulation; left then right whisker arcs with an ISI of 35 ms (L-R 35 ms), left then fight whisker arcs with an ISI of 175 ms (L-R 175 ms), fight then left whisker arcs with an ISI of 35 ms (R-L 35 ms), and finally,fight then left whisker arcs with an ISI of 175 ms (R-L 175 ms). The y-axis is in spike counts per 1 ms bin of time (300 presentations of each stimulus configurationwere given).

t05

401

Neuron 1

Neuron 2

2o30 lps L

Contra

3O 20 10

l

10

o .i,J,.,,hi,,-,,~,.L~£.

0

L'L"JL "kL "L ' LLaA' 'u" t'r'"'L~ ,,JllJlvl,l,,,,,,,

''''''''1'''''''''"''''''''1

20301

Contra

'''''''''1

4°1

Ipsi

30 20

10

10

0 ~'dlL""'~l"J"~'" t'"' ,,,,,,,,I,,,,,,*,,

0

I,,,,,,,,,

I

JI1,

"11"11'u I'LIIII t ""1 *ill / L -- "i'l ',,,,,,ll,Jll,l,,, ,,,,,,,T,,,

I

Ipsi

203040110 C°ntra I

L'-R 20 35ms 10

t-

.m

0 J,.,,,ll.,J-..~.hJ.,

0

,,,

,~,lllllllllllllJl

301 ~lpsi L'-R 20 175ms 10 0

Contra

20

35ms

lO 0

Contra

4°1

30 Ipsi 20 10 0 .,, L.,Ld,I,J,,,I,~

401]

L~k~,~.l..~..~,L ~•

I'lllIJlltJlllJll

R,-L 20 175ms 10

Contra

Ipsi

30 20 10

Ipsi

,.,

0

/ i, I"'JllJllJll,,ll,,,

°ll t ,

-0.1 0 Time (sec)

0.1

i

Ipsi

,

h'''''llillllllll

~si

IlrllllliTi

4°~1 ° ~ 30 20 0

Contra R,~L

f--~

& d., JL.,,Jlk~.. ,..,,.,,!1

0

o

00

- -

I c;~:r: lif,,i,,,,ij

i Contra

'l'''''''''l

-0.1 0 Time (sec)

0.1

106

Conclusions Although the classic feedforward model of touch has provided a fundamental blueprint for the development of somatosensory research in the last five decades, a variety of experimental findings and theoretical arguments demonstrate that this model no longer offers an accurate description on how tactile perception emerges in the mammalian brain. Instead, anatomical, physiological, and computational arguments favor the hypothesis that tactile perception emerges through interactive and recurrent interactions between multiple cortical and subcortical levels that define the mammalian somatosensory system. Central to this recurrent model of touch is the experimental demonstration that the massive corticofugal projections, that originate in the neocortex and reach most of the subcortical structures that form the somatosensory system, may play as relevant a role in tactile information processing as the parallel feedforward pathways of this system.

Acknowledgements This chapter describes research supported by grants from DARPA-ONR (N00014-98-1-0676), NSF IBN99-80043, and NIH DE-11121-01 to M.A.L.N. and an NRSA (1 F31 MH12570-01A1)to M.S.

References Adams, N.C., Lozsadi, D.A. and Guillery, R.W. (1997) Complexities in the thalamocortical and corticothalamic pathways. Eur. J. Neurosci., 9: 204-209. Albe-Fessard, D., Condes-Lara, M., Kesar, S. and Sanderson, P. (1983) Tonic cortical controls acting on spontaneous and evoked thalamic activity. In: G. Macchi, A. Rustioni and R. Spreatico (Eds.), Somatosensory Integration in the Thalamus: a Reevaluation Based on New Methodological Approaches. Elsevier, Amsterdam. Anderson, P., Eccles, J.C. and Sears, T.A. (1964) Cortically evoked depolarization of primary afferent fibers in the spinal cord. J. NeurophysioL, 27: 63-77. Andersen, P., Junge, K. and Sveen, O. (1967) Cortico-thalamic facilitation of somatosensory impulses. Nature, 214(92): 1011-1012. Andersen, P., Junge, K. and Sveen, O. (1972) Cortifugal facilitation of thalamic transmission. Brain Behav. Evol., 6: 170184. Barneoud, P., Gyger, M., Andres, E and Van der Lots, H. (1991)

Vibrissa-related behavior in mice: transient effect of ablation of the barrel cortex. Behav. Brain Res., 44: 87-99. Bishop, C.M. (1995) Neural Networks for Pattern Recognition. Clarendon Press, Oxford. Bourassa, J., Pinault, D. and Deschenes, M. (1995) Corticothalamic projections from the cortical barrel field to the somatosensory thalamus in rats: a single-fibre study using biocytin as an anterograde tracer. Eur. J. Neurosci., 7(1): 19-30. Brecht, M., Preilowski, B. and Merzenich, M.M. (1997) Functional architecture of the mystacial vibrissae. Behav. Brain Res., 84: 81-97. Brumberg, J.C., Pinto, D.J. and Simons, D.J. (1996) Spatial gradients and inhibitory summation in the rat whisker barrel system. J. Neurophysiol., 76(1): 130-140. Carvell, G.E. and Simons, D.J. (1990) Biometric analyses of vibrissal tactile discrimination in the rat. J. Neurosci., 10: 2638-2648. Castro-Alamancos, M.A. and Connors, B.W. (1996a) Short-term synaptic enhancement and long-term potentiation in neocortex. Proc. Natl. Acad. Sci. USA, 93(February): 1335-1339. Castro-Alamancos, M.A. and Connors, B.W. (1996b) Spatiotemporal properties of short-term plasticity in sensorimotor thalamocortical pathway of the rat. J. Neurosci., 16: 2767-2779. Cauller, L.J., Clancy, B. and Connors, B.W. (1998) Backward cortical projections to primary somatosensory cortex in rats extend long horizontal axons in layer I. J. Comp. Neurol., 390: 297-310. Chapin, J.K. and Woodward, D.J. (1981) Modulation of sensory responsiveness of single somatosensory cortical cells during movement and arousal behaviors. Exp. Neurol., 72(1): 164178. Chapin, J.K. and Woodward, D.J. (1982a) Somatic sensory transmission to the cortex during movement: gating of single cell responses to touch. Exp. NeuroL, 78(3): 654-669. Chapin, J.K. and Woodward, D.J. (1982b) Somatic sensory transmission to the cortex during movement: phasic modulation over the locomotor step cycle. Exp. Neurol., 78(3): 670-684. Chapin, J.K., Sadeq, M. and Guise, J.L.U. (1987) Corticocortical connections within the primary somatosensory cortex of the rat. J. Comp. Neurol., 263: 326-346. Chapman, C.E., Jiang, W. and Lamarre, Y. (1988) Modulation of lemniscal input during conditioned arm movements in the monkey. Exp. Brain Res., 72(2): 316-334. Chiaia, N.L., Rhoades, R.W., Bennett-Clarke, C.A., Fish, S.E. and Killackey, H.P. (1991) Thalamic processing of vibrissal information in the rat. I. Afferent input to the medial ventral posterior and posterior nuclei. J. Comp. Neurol., 314(2): 201216. Chmielowska, J., Carvell, G.E. and Simons, D.J. (1989) Spatial organization of thalamocortical and corticothalamic projection systems in the rat SmI barrel cortex. J. Comp. Neurol., 285: 325-338. Cohen, L.G. and Start, A. (1987) Localization, timing and specificity of gating of somatosensory evoked potentials during active movement in man. Brain, 110(Pt 2): 451-467. Contreras, D., Destexhe, A., Sejnowski, T.J. and Steriade, M. (1996) Control of spatiotemporal coherence of a thalamic

107

oscillation by corticothalamic feedback. Science, 274(5288): 771-774. Coquery, J.M. (1971) Changes in somaesthetic evoked potentials during movement. Brain Res., 31(2): 375. Coulter, J.D. (1974) Sensory transmission through lemniscal pathway during voluntary movement in the cat. J. Neurophysiol., 37(5): 831-845. Deadwyler, S.A. and Hampson, R.E. (1997) The significance of neural ensemble codes during behavior and cognition. Annu. Rev. Neurosci., 20: 217-244. Deschenes, M., Bourassa, J. and Pinault, D. (1994) Corticothalamic projections from layer V ceils in rat are collaterals of long-range corticofugal axons. Brain Res., 664(1-2): 215-219. Deschenes, M., Veinante, P. and Zhang, Z.W. (1998) The organization of corticothalamic projections: reciprocity versus parity. Brain Res. Rev., 28(3): 286-308. Diamond, M.E., Armstrong-James, M., Budway, M.J. and Ebner, F.F. (1992) Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus: dependence on the barrel field cortex. £ Comp. Neurol., 319: 66-84. Dykes, R.W. (1983) Parallel processing of somatosensory information: a theory. Brain Res. Rev., 6: 47-115. Erickson, R.P. (1968) Stimulus coding in topographic and nontopographic afferent modalities: on the significance of the activity of individual sensory neurons. Psychol. Rev., 75(6): 447-465. Erickson, R.P. (1986) A neural metric. Neurosci. Biobehav. Rev., 10: 377-386. Fabri, M. and Burton, H. (1991) Ipsilateral cortical connections of primary somatic sensory cortex in rats. J. Comp. Neurol., 311: 405-424. Faggin, B.M., Nguyen, K.T. and Nicolelis, M.A. (1997) Immediate and simultaneous sensory reorganization at cortical and subcortical levels of the somatosensory system. Proc. Natl. Acad. Sci. USA, 94(17): 9428-9433. Fanselow, E. and Nicolelis, M. (1999) Behavioral modulation of tactile responses in the rat somatosensory system. J. Neurosci., 19: 7603-7616. Fields, H.L. and Heinricher, M.M. (1985) Anatomy and physiology of a nociceptive modulatory system. Phil. Trans. R. Soc. Lond. B, 308: 361-374. Gastaut, H. (1952) Etude electrocorticographique de la reativite des rhytmes rolandiques. Rev. Neurol. (Paris), 87: 176-182. Georgopoulos, A.P., Swartz, A.B. and Ketter, R.E. (1986) Neuronal population coding of movement direction. Science, 233: 1416-1419. Ghazanfar, A.A. and Nicolelis, M.A.L. (1997) Nonlinear processing of tactile information in the thalamocortical loop. J. Neurophysiol., 78(1): 506-510. Ghazanfar, A.A. and Nicolelis, M.A.L. (1999) Spatiotemporal properties of layer V neurons of the rat primary somatosensory cortex. Cereb. Cortex, 9: 348-361. Ghazanfar, A.A., Krupa, D.J. and Nicolelis, M.A.L. (1997) Tactile processing by thalamic neural ensembles: the role of cortical feedback. Soc. Neurosci. Abstr., 1797. Ghez, C. and Lenzi, G.L. (1971) Modulation of sensory trans-

mission in cat lemniscal system during voluntary movement. Pflugers Arch. Eur. £ Physiol., 323(3): 273-278. Grossberg, S. (1976) Adaptive pattern classification and universal recording: II Feedback, expectation, olfaction, illusions. Biol. Cybern., 23: 187-202. Grossberg, S. (1988) Nonlinear neural networks: principles, mechanisms and architectures. Neural Networks, 1: 17-61. Grossberg, S. (1999) The link between brain, learning, attention, and consciouness. Consciousness Cognit., 8: 1-44. Guic-Robles, E., Valdivieso, C. and Guajardo, G. (1989) Rats can learn a roughness discrimination using only their vibrissal system. Behav. Brain Res., 31(3): 285-289. Hebb, D.O. (1949) The Organization of Behavior: a Neuropsychological Theory. John Wiley and Sons, New York. Hellweg, EC., Schultz, W. and Creutzfeldt, O.D. (1977) Extracellular and intracellular recordings from cat's cortical whisker projection area: thalamocortical response transformation. J. Neurophysiol., 40(3): 463-479. Hoogland, P.V., Welker, E. and Van der Loos, H. (1987) Organization of the projections from barrel cortex to thalamus in mice studied with Phaseohis vulgaris-leucoagglutinin and HRP. Exp. Brain Res., 68: 73-87. Hoogland, P.V., Welker, E., Van der Loos, H. and Wouterlood, EG. (1988) The organization and structure of the thalamic afferents from the barrel cortex in the mouse; a PHA-L study. In: M. Bentivoglio and R. Spreafico (Eds.), Cellular Thalamic Mechanisms. Elsevier Science, Amsterdam, pp. 151-161. Hoogland, P.V., Wouterlood, EG., Welker, E. and Van der Loos, H. (1991) Ultrastructure of giant and small thalamic terminals of cortical origin: a study of the projections from the barrel cortex in mice using Phaseolus vulgaris leuco-agglutinin (PHA-L). Exp. Brain Res., 87: 159-172. Hutson, K.A. and Masterton, R;B. (1986) The sensory contribution of a single vibrissa's cortical barrel. J. Neurophysiol., 56(4): 1196-1223. Jacquin, M.E, Chiaia, N.L., Haring, J.H. and Rhoades, R.W, (1990a) Intersubnucleus connections within the rat trigeminal brainstem complex. Somatosens. Motor Res., 7(4): 399-420. Jacquin, M.E, Wiegand, M.R. and Renehan, W.E. (1990b) Structure-function relationships in rat brain stem subnucleus interpolaris. VIII. Cortical inputs. J. Neurophysiol., 64(1): 327. Johnson, K.O., Hsiao, S.S. and Twombly, I.A. (1995) Neural mechanisms of tactile form recognition. In: M.S. Gazzaniga (Ed.), The Cognitive Neurosciences. MIT Press, Cambridge, MA, pp. 253-267. Kaas, J.H. (1990) Somatosensory system. In: G. Paxinos (Ed.), The Human Nervous System. Academic Press, San Diego, pp. 813-844. Kaas, J.H. and Pons, T.P. (1988) The somatosensory system of primates. In: H.D. Steklis and J. Erwin (Eds.), Comparative Primate Biology, Vol. 4. Alan R. Liss, New York, pp. 421468. Kaas, J.H., Merzenich, M.M. and Killackey, H.P. (1983) The organization of the somatosensory cortex following peripheral nerve damage in adult and developing mammals. Annu. Rev. Neurosci., 6: 325-356.

108 Killackey, H.P. (1973) Anatomical evidence for cortical subdivisions based on vertically discrete thalamic projections from the ventral posterior nucleus to cortical barrels in the rat. Brain Res., 51: 326-331. Kleinfeld, D. and Delaney, K.R. (1996) Distributed representation of vibrissa movement in the upper layers of somatosensory cortex revealed with voltage-sensitive dyes [published erratum appears in J. Comp. NeuroL (1997) 378(4):594]. J. Comp. Neurol. , 375(1): 89-108. Koralek, K.A., Olavarria, J. and Killackey, H.P. (1990) Areal and laminar organization of corticocortical projections in the rat somatosensory cortex. J. Comp. Neurol., 299: 135-150. Krnpa, D.J., Ghazanfar, A.A. and Nicolelis, M.A.L. (1999) Immediate thalamic sensory plasticity depends on corticothalamic feedback. Proc. Natl. Acad. Sci. USA, 96: 8200-8205. Lee, R.G. and White, D.G. (1974) Modification of the human somatosensory evoked response during voluntary movement. Electroencephalogr. Clin. NeurophysioL, 36(1): 53-62. Lin, C.S., Nicolelis, M.A., Schneider, J.S. and Chapin, J.K. (1990) A major direct GABAergic pathway from zona incerta to neocortex [see comments]. Science, 248(4962): 1553-1556. Lu, S.-H. and Lin, R.C.S. (1993) Thalamic afferents of the rat barrel cortex: a light- and electron-microscopic study using Phaseolus vulgaris Leucoagglutinin as an anterograde tracer. Somatosens. Motor Res., 10: 1-16. Lu, S.M. and Lin, C.S. (1986) Cortical projection patterns of the medial division of the nucleus posterior thalami in the rat. Soc. Neurosci. Abstr., 12: 1434. Masino, S.A. and Frostig, R.D. (1996) Quantitative long-term imaging of the functional representation of a whisker in rat barrel cortex. Proc. Natl. Acad. Sci. USA, 93: 4942-4947. McCormick, D.A. and Pape, H.-C. (1990) Noradrenergic and serotonergic modulation of a hyperpolarization-activated cation current in thalamic relay neurones, J. Physiol., 431: 319-342. McCormick, D.A. and von Krosigk, M. (1992) Corticothalamic activation modulates thalamic firing through glutamate metabotropic receptors. Proc. Natl. Acad. Sci. USA, 89: 2774. McLean, J. and Waterhouse, B.D. (1994) Noradrenergic modulation of cat area 17 neuronal responses to moving visual stimuli. Brain Res., 667: 83-97. Merzenich, M.M., Kaas, J.H., Wall, J.T., Nelson, R.J., Sur, M. and Felleman, D.J. (1983) Topographic reorganization of somatosensory cortical areas 3b and 1 in adult monkeys following restricted deafferentation. Neuroscience, 8(1): 33-55. Moore, C.I. and Nelson, S.B. (1998) Spatio-temporal subthreshold receptive fields in the vibrissa representation of rat primary somatosensory cortex. J. Neurophysiol., 80(6): 2882-2892. Mountcastle, V. (1957) Modality and topographic properties of single neurons of cats' somatic sensory cortex. J. Neurophysiol., 20: 408-434. Mountcastle, V. (1974). Neural mechanisms in somesthesia. In: V. Mountcastle (Ed.), Medical Physiology, Vol. I. C.V. Mosby, St. Louis, pp. 307-347. Mumford, D. (1991) On the computational architecture of the neocortex. I. The role of the thalamo-cortical loop. Biol. Cybern., 65(2): 135-145.

Mumford, D. (1992) On the computational architecture of the neocortex. II. the role of corticocortical loops. Biol. Cybern., 66: 241-251. Mumford, D. (1994). Neuronal Architectures for pattern-theoretic problems. In: D. Koch (Ed.), Large-Scale Neuronal Theories of the Brain. MIT Press, Cambridge, MA, pp. 125152. Nelson, R.J. (1984) Responsiveness of monkey primary somatosensory cortical neurons to peripheral stimulation depends on 'motor-set'. Brain Res., 304(1): 143-148. Nelson, R.J. (1987) Activity of monkey primary somatosensory cortical neurons changes prior to active movement. Brain Res., 406(1-2): 402--407. Nicolelis, M.A.L. (1996) Beyond maps: a dynamic view of the somatosensory system. Braz. J. Med. BioL Res., 29: 401--412. Nicolelis, M.A. and Chapin, J.K. (1994) Spatiotemporal structure of somatosensory responses of many-neuron ensembles in the rat ventral posterior medial nucleus of the thalamus. J. Neurosci., 14(6): 3511-3532. Nicolelis, M.A.L., Chapin, J.K. and Lin, R.C.S. (1991) Ontogeny of corticocortical projections of the rat somatosensory cortex. Somatosens. Motor Res., 8(3): 193-200. Nicolelis, M.A., Lin, R.C., Woodward, D.J. and Chapin, J.K. (1993a) Dynamic and distributed properties of many-neuron ensembles in the ventral posterior medial thalamus of awake rats. Proc. Natl. Acad. Sci. USA, 90(6): 2212-2216. Nicolelis, M.A.L., Lin, R.C.S., Woodward, D.J. and Chapin, J.K. (1993b) Induction of immediate spatiotemporal changes in thalamic networks by peripheral block of ascending cutaneous information. Nature, 361: 533-536. Nicolelis, M.A., Baccala, L.A., Lin, R.C. and Chapin, J.K. (1995) Sensorimotor encoding by synchronous neural ensemble activity at multiple levels of the somatosensory system. Science, 268(5215): 1353-1358. Nicolelis, M.A.L., Oliveira, L.M.O., Lin, R.C.S. and Chapin, J.K. (1996) Active tactile exploration influences the functional maturation of the somatosensory system. J. Neurophysiol., 75: 2192-2196. Nicolelis, M.A., Fanselow, E.E. and Ghazanfar, A.A. (1997) Hebb's dream: the resurgence of cell assemblies. Neuron, 19(2): 219-221. Nicolelis, M,A., Katz, D. and Krupa, D.J. (1998a) Potential circuit mechanisms underlying concurrent thalamic and cortical plasticity. Rev. Neurosci., 9(3): 213-224. Nicolelis, M.A.L., Ghazanfar, A.A., Stambaugh, C.R., Oliveira, L.M.O., Laubach, M., Chapin, J.K., Nelson, R.J. and Kaas, J.H. (1998b) Simultaneous encoding of tactile information by three primate cortical areas. Nature Neurosci., 1(7): 621-630. Niedermeyer, E. (1993) The normal EEG of the waking adult. In: E. Niedermeyed and E Lopez da Silva (Eds.), Electroencephalography. Basic Principles, Clinical Applications and Related Fields. Williams and Wilkins, Baltimore, MD, pp. 131-152. Ogden, T.E. (1960) Cortical control of thalamic somato-sensory relay nuclei. Electroenceph. Clin. Neurophysiol., 12: 621-634. Olavarria, J.F., DeYoe, E.A., Knierim, J.J., Fox, J.M. and Van Essen, D.C. (1992) Neuronal responses to visual texture pat-

109

terns in middle temporal area of the macaque monkey. J. Neurophysiol., 68(1): 164-181. Parker, J. and Dostrovsky, J. (1999) Cortical involvement in the induction, but not expression, of thalamic plasticity. J. Neurosci., 19: 8623-8629. Pazos, A.J., Orezzoli, S.L., McCabe, P.M., Dietrich, W.D. and Green, E.J. (1995) Recovery of vibrissae-dependent behavioral responses following barrelfield damage is not dependent upon the remaining somatosensory cortical tissue. Brain Res., 689(2): 224-232. Pidoux, B. and Verley, R. (1979) Projections on the cortical somatic I barrel subfield from ipsilateral virbrissae in adult rodents. Electroencephalogr. Clin. Neurophysiol., 46: 715726. Pinault, D. and Deschenes, M. (1998) Projection and innervation patterns of individual thalamic reticular axons in the thalamus of the adult rat: a three-dimensional, graphic and morphometric analysis. J. Comp. NeuroL, 391(2): 180-203. Pinanlt, D., Bourassa, J. and Deschenes, M. (1995) The axonal arborization of single thalamic reticular neurons in the somatosensory thalamus of the rat. Eur. J. Neurosci., 7(1): 3140. Pinault, D., Smith, Y. and Deschenes, M. (1997) Dendrodendritic and axoaxonic synapses in the thalamic reticular nucleus of the adult rat. J. Neurosci., 17(9): 3215-3233. Polley, D.B., Chen-Bee, C.H. and Frostig, R.D. (1999) Varying the degree of single-whisker stimulation differentially affects phases of intrinsic signals in rat barrel cortex. J. Neurophysiol., 81(2): 692-701. Purves, D., Riddle, D.R. and LaMantia, A.-S. (1992) Iterated patterns of brain circuitry (or how the cortex gets its spots). Trends Neurosci., 15(10): 362-368. Rhoades, R.W., Belford, G.R. and Killackey, H.P. (1987) Receptive-field properties of rat ventral posterior medial neurons before and after selective kainic acid lesions of the trigeminal brain stem complex. J. Neurophysiol., 57(5): 1577-1600. Salt, T.E. and Eaton, S.A. (1996) Functions of ionotropic and metabotropic glutamate receptors in sensory transmission in the mammalian thalamus. Prog. NeurobioL, 48: 55-72. Salt, T.E. and Turner, J.P. (1998) Modulation of sensory inhibition in the ventrobasal thalamus via activation of group II metabotropic glutamate receptors by 2R,4R-aminopyrrolidine-2,4-dicarboxylate. Exp. Brain Res., 121(2): 181-185. Schmidt, R.E, Schady, W.J. and Torebjork, H.E. (1990a) Gating of tactile input from the hand. I. Effects of finger movement. Exp. Brain Res., 79(1): 97-102. Schmidt, R.E, Torebjork, H.E. and Schady, W.J. (1990b) Gating of tactile input from the hand. II. Effects of remote movements and anaesthesia. Exp. Brain Res., 79(1): 103-108. Sejnowski, T.J., Koch, C. and Churchland, P.S. (1988). Comput. Neurosci. Sci., 241: 1299-1306. Sherman, S.M. and Guillery, R.W. (1996) Functional organization of thalamocortical relays. J. Neurophysiol., 76(3): 13671395. Shigemi, S., Ichikawa, T., Akasaki, T. and Sato, H. (1999) Temporal characteristics of response integration evoked by

multiple whisker stimulations in the barrel cortex of rats. J. Neurosci., 19: 10164-10175. Shin, H.C. and Chapin, J.K. (1989) Mapping the effects of motor cortex stimulation on single neurons in the dorsal column nuclei in the rat: direct responses and afferent modulation. Brain Res. Bull., 22(2): 245-252. Shin, H.C. and Chapin, J.K. (1990a) Modulation of afferent transmission to single neurons in the ventroposterior thalamus during movement in rats. Neurosci. Lett., 108(1-2): 116-120. Shin, H.C. and Chapin, J.K. (1990b) Movement induced modulation of afferent transmission to single neurons in the ventroposterior thalamus and somatosensory cortex in rat. Exp. Brain Res., 81(3): 515-522. Shin, H.-C. and Chapin, J.K. (1990c) Mapping the effects of SI cortex stimulation on somatosensory relay neurons in the rat thalamus: direct responses and afferent modulation. Somatosens. Motor Res., 7: 421--434. Shuler, M.G., Krupa, D.J. and Nicolelis, M,A.L. (2000) Discrimination of bilateral whisker stimuli in the freely behaving rat.

Soc. Neurosci. Abstr. Shuler, M.G., Krupa, D.J. and Nicolelis, M.A.L. (2001) Bilateral integration of whisker information in the primary somatosensory cortex of rats, submitted for publication. Simons, D.J. (1985) Temporal and spatial integration in the rat SI vibrissa cortex. J. Neurophysiol., 54(3): 615-635. Simons, D.J. and Carvell, G.E. (1989) Thalamocortical response transformation in the rat vibrissa/barrel system. J. Neurophysiol., 61(2): 311-330. Turner, J.E and Salt, T.E. (1998) Characterization of sensory and corticothalamic excitatory inputs to rat thalamocortical neurones in vitro. J. Physiol. (Lond.), 510(Pt 3): 829-843. Turner, J.E and Salt, T.E. (1999). Group III metabotropic glutamate receptors control corticothalamic synaptic transmission in the rat thalamus in vitro [In Process Citation]. J. Physiol. (Lond.), 519 Pt 2, 481-491. Van dcr Loos, H. (1976) Barreloids in mouse somatosensory thalamus. Neurosci. Lett., 2: 1-6. Veinante, E and Deschenes, M. (1999) Single- and multi-whisker channels in the ascending projections from the principal trigeminal nucleus in the rat. J. Neurosci., 19(12): 50855095. Waller, H.J. and Feldman, S.M. (1967) Somatosensory thalamic neurons: effects of cortical depression. Science, 157: 10741077. Waterhouse, B.D., Border, B., Wahl, L. and Mihalloff, G.A. (1994) Topographic organization of rat locus coeruleus and dorsal raphe nuclei: distribution of cells projecting to visual system structures. J. Comp. Neurol., 336: 345-361. White, E. and DeAmicis, R. (1977) Afferent and efferent projections of the region in mouse SmI cortex which contains the posteromedial barrel subfield. J. Comp. NeuroL, 175: 455482. Woolsey, T.A. and Van der Loos, H. (1970) The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex: the description of a cortical field composed of discrete cytoarchitectonic units. Brain Res., 17: 205-242.

110

Yuan, B., Morrow, T.J. and Casey, K.L. (1985) Responsiveness of ventrobasal thalamic neurons after suppression of SI cortex in the anesthetized rat. J. Neurosci., 5: 2971-2978. Yuan, B., Morrow, T.J. and Casey, K.L. (1986) Cortifugal influences of S 1 cortex on ventrobasal thalamic neurons in the

awake rat. J. Neurosci., 6:3611-3617. Zhang, Z.-W. and Deschenes, M. (1998) Projections to layer VI of the postermedial barrel field in the rat: a reappraisal of the role of corticothalamic pathways. Cereb. Cortex, 8: 428-436.