| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Psychology and Volen National Center for Complex Systems, Brandeis University, Waltham, Massachusetts
Submitted 30 September 2004; accepted in final form 28 November 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Destexhe and Sejnowski 2001; Freeman 2000; Niedermeyer 1993; Steriade 2000), most obvious among them a neocortical rhythm focused around 7–12 Hz. Although similar in frequency range, this oscillatory pattern is distinct from the olfactory and hippocampal 
rhythm; instead, it is similar to the human µ/
rhythms, which are characterized by high-voltage 7 to 12 Hz oscillations synchronous across neocortical space, associated with quiet immobility, and interrupted by somatomotor activation (Arroyo et al. 1993; Gastaut 1952; Nikouline et al. 2000; Semba et al. 1980; Tiihonen et al. 1989).
Although such rhythms are ubiquitous across mammalian species and behavioral preparations, their function is hotly debated: various researchers have suggested that they may reflect either a physiological (Pfurtscheller et al. 1997; Tiihonen et al. 1989) or pathological (Buzsaki et al. 1990; Johanson et al. 2003; Robinson and Gilmore 1980; Shaw 2004; Vergnes et al. 1987) disengagement from environmental contingencies—"idling" and "absence seizure" activity, respectively (the latter being supported by the characteristic spike-and-wave shape of this oscillation). Other data, however, have led some researchers to suggest that the somatosensory µ rhythm might represent a task-oriented state (Fanselow and Nicolelis 1999; Nikouline et al. 2000), much as olfactory cortex and bulb rhythms seem related to stimulus acquisition (Bressler 1987).
It has been difficult to provide an unimpeachable argument for or against a relationship between rodent 7 to 12 Hz neocortical rhythms and perception, in part because both main hypotheses claim primary support in interpretation of a single behavioral observation—the fact that rhythmic episodes tend to happen when the animal is immobile, only to fade just before the animal moves. Specifically, the disengagement hypothesis postulates that postinactivity movement rouses the animal out of the oscillatory state, whereas the anticipation/processing hypothesis postulates that the rhythms specifically predict movement. Furthermore, somatosensory cortex may not be the ideal location in which to seek a functional role for such rhythms because rats often do not rhythmically sample somatosensory stimuli in task contexts, even in the case of whisker stimuli (see Krupa et al. 2004). It might be more reasonable to examine potential relationships between neural rhythms and perception in a chemosensory system, in which stimuli are manipulated according to the 7 to 12 Hz rhythm of either sniffing or licking.
Here, we present a task in which a restrained rat's task-oriented anticipation of stimuli can be cleanly distinguished from disengaged (idling/seizure) behavior, and show that 7 to 12 Hz rhythms have to do with the latter behavior. Furthermore, we show this effect using fluid taste stimuli, which are processed by rhythmic stimulus manipulation (licking). Local field potentials (LFPs) were recorded in gustatory cortex (GC) as rats were trained to perform a simple timed fluid self-administration similar to previously described timing/delayed-reinforcement tasks (Church et al. 1994). In these sessions, rats went through periods in which they displayed goal-oriented efforts to obtain fluid reward, and periods in which they continued to drink but put forth no organized effort to obtain the fluid. The 7 to 12 Hz rhythm, which by several criteria were identical to human and rodent µ, appeared almost exclusively during the period of task disengagement, and was clearly dissociable from physical and gustatory processing of liquid stimuli.
These results are the first description of oscillatory activity in the gustatory system, and the first to specifically show that the oscillatory state is related to disengagement from task orientation, rather than to anticipation of chemosensory stimuli; furthermore they provide evidence that, at least in GC, µ rhythm is not related to sensory input. We suggest that 7 to 12 Hz cortical rhythms in rats reflect a state of withdrawal of directed focus on the environment, and suggest that the "withdrawal" is consistent with nonpathological states (i.e., not full-blown epileptic seizure) studied in normal human subjects.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Four female Long Evans rats (250–300 g at the time of surgery) served as the subjects in this study. Animals were maintained on a 12-h light/12 h-dark schedule, with experiments carried out in the light portion of the cycle. Unless otherwise specified, animals in home cages had unrestricted access to chow and water.
Surgery
Animals were anesthetized using an intraperitoneal injection of a ketamine, xylazine, and acepromazine cocktail (100, 5, and 1 mg/kg, respectively). Small (20% of induction dose) injections of ketamine were delivered whenever breathing rate and hindpaw pinch response suggested that anesthesia level was becoming light. The anesthetized rat was placed on a standard stereotaxic frame, after which the scalp was slit and reflected. Holes were bored in the skull for four to six 0–80 ground screws and for two microelectrode assemblies. The dura in these latter holes was reflected, and 16-microwire bundles attached to a "mini-microdrive " (Katz et al. 2001b) were lowered slowly (100 µm/min) to 0.6 mm above layer 5 of GC bilaterally, guided by stereotaxic measurements (AP 1.4, ML ±5 from bregma, DV –4.0 from dura). Once in position, each electrode bundle was cemented to the skull with dental acrylic, as was a bolt for restraining head movements in the later experiment.
After electrode implantation, the rat was removed from the stereotax, and intraoral cannulae (IOC) were inserted bilaterally. Thin polyethylene tubes were passed from the space between the first maxillary molar and the lip, through the masseter muscle and inside the zygomatic arch, and out through the opening in the scalp, where they were cemented to the dental acrylic cap (Phillips and Norgren 1970). The scalp was then sutured or stapled around the implant, and antibiotic ointment was applied liberally to the wound. The rat received a subcutaneous injection of penicillin (0.1 ml) immediately after surgery, and again 48 h later.
Training
Rats underwent adaptation to handling starting 2 days postsurgery, wherein the experimenter cleaned the IOCs and turned the 0–80 microdrive screw turn (lowering the electrodes about 81 µm); soon the electrodes were buried in a taste-responsive part of dysgranular insular cortex (Katz et al. 2001a, 2002).
Seven days after surgery, rats were begun on a regimen of mild water restriction: 30 min of water ad lib in the late afternoon. After 3 days of water restriction, they were introduced to the experimental restraint chamber, a Plexiglas box (Bermejo et al. 1998; Katz et al. 2001a; Welsh et al. 1995) placed in a sound-attenuating chamber that is itself inside an isolation room. The backs of the rats' head caps were bolted to an eyelet in the front panel of the box, the height of which could then be adjusted to minimize the animals' discomfort. The rats' front paws were unrestrained. A mirror under the floor of the chamber, placed at a 45° angle to the horizontal, allowed a small low-light video camera an unobstructed view of the orofacial region of the rat; digital video recordings were made of training sessions for later analysis of occurrence of consumptive behaviors (licking and chewing).
In the first session of restraint, lever training began. A lever was placed directly under the right forepaw, and pressing of this lever caused a 40-µl aliquot of water to be delivered (under nitrogen pressure) through the IOC into the mouth, where it was consumed (a process that involves visible licking and chewing by the rat). Rats quickly became eager to press for water. Rats were free to press at will, but over the next several sessions they were trained that they would receive water reward only after 30 s (the "foreperiod") had elapsed since the previous delivery: lever presses made before the end of a foreperiod were unrewarded, and the first press after the end of a foreperiod (at any latency >30 s) was rewarded and restarted the clock. Figure 1 describes this training process graphically.
|
Recordings from all microwires were initially amplified and filtered, and fed into a parallel processor capable of simultaneously digitizing 32 signals at 40 kHz (Plexon, Dallas, TX). Sixteen of these signals were split off to a separate amplifier with filtering set for LFPs recording (band-pass 3–90 Hz), and from there to a computer, where they were digitized at 1,000 Hz. LFPs, unit, behavioral, and video records were synchronized to a single clock.
Behavioral analysis
Lever presses were recorded throughout the session, which were divided in trials. Trials ended (and new trials began) with a rewarded lever press. Parameters extracted from each trial included: 1) total number of lever presses; 2) number of presses in the 10 s preceding a rewarded press; 3) latency of the rewarded lever press, measured from the start of the trial; and 4) SD of all lever press latencies across a series of 5 consecutive trials. This last variable estimates how well clustered in time the responses tend to be in a manner that is relatively insensitive to occasional early presses; the closer the rat comes to pressing only at the time of available reward, the lower this number will be. For ease of visualization and to reduce the impact of single-trial outliers on the statistics, values of each parameter (except for SD) were smoothed using a sliding window of 5 trials and an overlap of 4 trials. Combinations of these parameters were used to describe each trial (see RESULTS) in an abstract 3-dimensional (3D) space, and a k-means cluster analysis method was used to divide them in two groups (Jain and Dubes 1988). This clustering method allowed the whole session to be divided in the two periods (task oriented and disengaged) as described in detail in the RESULTS, with a resolution of 5 trials, determined by the smoothing filter.
A video coder, blind to the purpose of the experiment and the behavioral state of individual rats, examined the sessions, noting periods in which the rats engaged in intensive licking/chewing behavior. The times of these bouts, as noted from the time stamps on the video record, mostly corresponded to the times of stimulus delivery. The coder noted start and end times of bouts, and calculated average bout length from these data.
Electrophysiological analysis
The marked difference in amplitude and frequency between periods of oscillatory and nonoscillatory activity allowed rhythmic episodes to be detected using a threshold-crossing method, but this method was validated by consideration of power spectral densities (see RESULTS). LFPs were first band-passed filtered, using a fast Fourier transformation (FFT) algorithm, between 7 and 12 Hz and then 7 to 12 Hz episodes were identified if the power crossed a determined threshold. Single high-voltage events were detected using a threshold-crossing method: sets of cycles separated by <1,000 ms were grouped together and considered a single episode; in this way the beginning of an episode was defined as the first high-voltage spike after a period of 
1 s of no high-voltage 7 to 12 Hz activity, and the end of the episode was determined by the last spike before a 1-s period of low-voltage activity. We revalidated our detection method by visual inspection and by comparing the field potentials threshold-crossing method with a spectrogram threshold crossing. Once rhythmic episodes were detected, their total duration per trial was measured for both the task-oriented and disengagement periods.
Power spectral density was computed using the Welch method (Welch 1967) during periods of 7 to 12 Hz oscillations and desynchronization. Cross-correlations were estimated on unfiltered data, by computing the cross-covariance between field potentials recorded in different hemispheres during the occurrence of 7 to 12 Hz episodes and during periods of visually identified high-frequency, low-amplitude activity. To validate the correlation results obtained with the cross-covariance analysis, spectral coherence in the 7 to 12 Hz range was computed on the same data using a 0.512-s Hanning window with an overlap of 0.256 s.
To track the normalized time course of the oscillatory pattern in relation to lever presses, we directly computed the time course of the 7 to 12 Hz power in the overall spectrum during the interval surrounding each lever press (from 2 s before until 5 s after) using a sliding spectrogram window of 1.024-s duration (i.e., 1,024 points). Results were averaged to produce a smooth (also normalized) record of the impact of lever presses on 7 to 12 Hz components of the overall LFPs.
Histology
After the experimental sessions, subjects were deeply anesthetized with ketamine (150 mg/kg) and perfused through the heart with saline followed by 5% formalin in saline. Seven seconds of DC current (7 µA) were passed through selected microwires in preparation for staining. The brains were removed and immersed in a sucrose formalin mixture, where they remained, refrigerated, until fixed. Sections (80 µm) cut through the implanted areas were stained with Prussian blue for ferrous deposits blasted off the electrode tips, and counterstained with cresyl violet for cell bodies.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Four implanted, water-restricted rats were habituated to head restraint, with their head cap bolted to the front panel of the box, and were trained to use their right forepaws to press a lever for reward (usually water) at fixed intervals. Each session began with the rat positioned in front of a lever and was terminated when the rat had been in a disengaged state (see following text) for at least about 30 min. Because rats' ability to maintain focus varied from day to day, session length varied within and between rats; the average duration of each session was 118.32 ± 24.65 min (mean ± SD), or 202 ± 43.3 trials, n = 19.
Experimental sessions could consistently be divided into two phases, reflecting a switch between behavioral states (Fig. 2). One phase (which we refer to as the task-oriented phase) was typified by organized lever-pressing activity: rats waited quietly through the initial portion of each trial, and then initiated an intense pressing regime as the end of the foreperiod approached (these lever presses are represented by the blue dots in Fig. 2A). This intense lever pressing ultimately resulted in the rat receiving fluid reward for the press just after the end of the foreperiod, at which time the trial ended and the behavioral cycle repeated. Task-oriented behavior, which lasted in this case for the first 7/10 of the session shown in Fig. 2A, resembles that observed in other simple inhibition/timing tasks (Church et al. 1994). During this task-oriented phase, each rat's behavior was very clearly oriented toward the goal of water acquisition, in that lever pressing increased as the end of the foreperiod (and, thus the time of available fluid reward) approached.
|
Several behavioral measures related to the lever-press regimes distinguished the task-oriented from the disengaged phase, as shown in Fig. 2, B1–B3. In each, the red section of the graph represents the disengaged phase. Note that the latency at which the rewarded lever press occurred (Fig. 2B1) was consistently stable during the task-oriented phase, as was the SD of all (unrewarded and rewarded) lever-press latencies (this measure differed from the latency of reward measure in that it broadly reflected the tendency for even the unrewarded lever presses to cluster toward the end of the foreperiod; Fig. 2B2); the number of lever presses in the last 10 s preceding a reward (Fig. 2B3), meanwhile, was reliably >1 during the task-oriented phase. This pattern was consistent across sessions and animals, lasting on average 75.67 ± 20.33 min, or 114.89 ± 39.64 trials, n = 19. In contrast, rewarded lever presses often occurred with longer latency during the disengaged phase than during task orientation (46.17 ± 10.19 s vs. 31.37 ± 0.73 s, n = 19; top 3/10 of Fig. 2B1), and the SD of lever-press times in the disengaged phase was high (16.86 ± 8.16 vs. 5.80 ± 1.02 during task oriented periods, n = 19; top 3/10 of Fig. 2B2). Also, fewer lever presses occurred in the 10 s before the reward during the disengaged phase than during the task-oriented period (0.71 ± 0.55 vs. 12.03 ± 5.25, n = 19; top 3/10 of Fig. 2B3). The projection of trials onto 3D combinations of these parameters consistently revealed a clean dissociation between periods of task-orientation and disengagement (Fig. 2C); this dissociation was further validated by k-means cluster analysis, which divided the trials into two groups (blue vs. red trials in Fig. 2C). Although we cannot say for sure that either task-oriented or disengaged trials represent single, unified clusters, the dissociation of task-oriented from disengaged trials is clearly demonstrated in this analysis.
Because shifts of rat behavior from task-oriented to disengaged typically happened with similar suddenness, it was usually possible to estimate the change point with fair accuracy. The 5-trial smoothing algorithm used in data preparation introduced a slight lack of precision in the cluster solution, but this was considered acceptable because: 1) the shift in state itself probably did not occur instantaneously, and thus there was little point in attempting to estimate it with zero error; and 2) the slight lack of precision did not impede the subsequent analysis, or the conclusions based on our suggestion that rats progressed from an obviously task-oriented mode to a disengaged mode.
7 to 12 Hz oscillations in GC
LFPs recorded in the GC simultaneously with performance of the fixed-interval reinforcement task showed periods of high-frequency, low-amplitude activity (Fig. 3, A and B1), occasionally interrupted by bursts of spike-and-wave high-amplitude activity (Fig. 3, A and B2). During these episodes, both the amplitude and the frequency content of the LFPs were distinctive; the smaller-amplitude signal was relatively flat between 7 and 12 Hz, but the high-amplitude signal included a distinctive peak of power in the 7 to 12 Hz range (Fig. 4A). Such episodes had variable duration, with an average length of 1.64 ± 1.09 s, n = 18; the calculated duration was to some degree dependent on search algorithm parameterization. Both in form and in frequency, these episodes clearly resemble classically described human and rodent µ rhythm. For this reason, and because the resemblance extends to similarity of interhemispheric coherence and similarity of relationship to sensory and motor events (see following text), we provisionally refer to the observed rhythm as "GC µ".
|
|
The observed oscillatory pattern demonstrated a high degree of intra- and interhemispheric synchrony: Figure 4B shows the cross-correlograms between bilateral local field potentials recorded during periods of rhythmic and low-amplitude activity. Across sessions and animals, the average cross-correlogram peak is 0.76 ± 0.12, n = 19 for high-amplitude events recorded bilaterally, and the shape of the cross-correlogram shows ripples characteristic of coherent oscillations of nearly 10 Hz. Cross-correlograms made during periods of low-amplitude activity showed a significantly smaller peak (0.35 ± 0.08, n = 19) and no ripples of nearly 10 Hz. The difference between the central peak heights was highly significant [t(18) = 10.6353, P < 0.00001]. Correlation results obtained from the cross-covariance analysis were confirmed by estimating the mean coherence between signals in the 7 to 12 Hz band (Fig. 4C); the mean coherence value was 0.84 ± 0.08, n = 19 during periods of GC µ, significantly different from the one obtained during period of low-amplitude activity (0.17 ± 0.06, P < 0.00001).
Correlation between GC µ and behavior within sessions
On the basis of the above data, we investigated the functional implications of GC µ activity by relating the occurrence of oscillatory episodes to time within a session. Figure 5A shows a session-long trace of LFPs recorded from a single GC electrode (about 2 h); the incidence of high-amplitude rhythmic activity is much higher during the last half of this session than that during the first half. When estimated times of GC µ episodes occurrence were overlaid on the trial structure, as shown in Fig. 5B (a representative session similar to that presented in Fig. 2A), a striking finding concerning the relationship between GC µ and task performance emerged. Red lines denoting the occurrence of GC µ events clearly appear only after the transition between the task-oriented and disengaged phases (k-means clustering) estimated by the horizontal arrow; black dots show unrewarded lever presses, leading to the rewarded lever presses in green. Inspection of this plot reveals that the occurrence of oscillations was largely restricted to periods of disengagement.
|
This analysis makes it clear that GC µ can be dissociated from the desire to acquire fluid stimuli, but leaves unexamined the possibility that it might be related to lever pressing or stimulus processing during the disengaged phase. We investigated these issues by examining the prevalence of GC µ around the time of task-oriented and disengaged lever presses. Lever presses made during the task-oriented phase were not typically preceded by substantial incidence of GC µ—as already explained, little oscillatory activity was produced during proper performance of the task. During the disengaged phase, however, prepress LFPs contained a great deal of power in the 7 to 12 Hz range (Fig. 6A). The relatively high amplitude of GC µ vanished just before the lever presses themselves. Thus in the disengaged phase GC µ did not predict acquisition of fluid, but rather was inhibited by movement, just as previously reported in studies of similar somatosensory rhythms (Pfurtscheller et al. 1997).
|
As one final test of this hypothesis, 3 rats received one session in which sucrose, a highly effective tastant, replaced water as the fluid reward. In these sessions, disengagement occurred as suddenly as during water sessions, and again GC µ appeared almost exclusively during the disengaged phase (Fig. 7A). Also as in water sessions, impending lever presses quenched GC µ, and again this inhibition lasted for the 5 s of processing/consumption of administered fluid (Fig. 7B). Thus GC µ clearly has nothing to do with taste processing in this task, regardless of whether the taste is water or sucrose.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Oscillatory activity in gustatory cortex
Local field potentials recorded in the gustatory cortex reveal alternating periods of low-amplitude, high-frequency activity with periods of high-voltage oscillations. High-voltage events are characterized by a spike-and-wave shape, with a sharp negative deflection followed by a positive slow wave, and have a frequency of 7–12 Hz and a high degree of interhemispheric synchrony. All of these properties are in agreement with previous reports of the 7 to 12 Hz rhythm in other cortical areas of rodents (Buzsaki et al. 1990; Shaw 2004; Wiest and Nicolelis 2003) and humans (Arroyo et al. 1993; Nikouline et al. 2000; Tiihonen et al. 1989).
Gustatory cortex is adjacent to somatosensory cortex; this raises the theoretical possibility that the rhythm recorded in GC could actually be observed as an effect of volume transmission. The correlation between single-unit activity and local field potentials at single electrodes, however, supports the conclusion that the observed activity was local in nature. It is possible that GC oscillations could be synchronous with those generated in the somatosensory thalamocortical axis; this possibility will be addressed in future studies.
Oscillations and non-task-oriented behavior
7 to 12 Hz oscillations in rodents are associated with awake immobility and extinguished by sensory stimuli or movement (Buzsaki et al. 1990; Hartmann and Bower 1998; Nicolelis et al. 1995). This one basic result has been interpreted by some as demonstrating that 7 to 12 Hz oscillations reflect a state of "idling " and by others as evidence that they represent preparation to process incoming sensory stimuli. The behavioral test used in this report, a fixed-time fluid self-administration task, allows us to unambiguously dissociate the two behavioral states possibly associated with GC µ. Rats involved in this task were either engaged, with intense lever pressing preceding the end of the foreperiod and fluid reward collected shortly after, or disengaged from the task, with fewer lever presses often occurring long after the end of the foreperiod. In either case, the rats spent much time immobile, and in both cases the immobility precedes movement and stimulus acquisition, but only in the disengaged phase was the immobility associated with GC µ.
Rats' behavior in the late phase could hypothetically reflect any or all of several internal factors, including most obviously lack of thirst (others include frustration and fatigue), but several observations lead us to doubt that satiation alone provides a complete and compelling explanation of disengagement. First of all, the amount of water delivered across the task-oriented portion of the session, typically in the order of 8 ml (40 µl x 202 trials), is far less than that required to satiate a water-restricted rat. In fact, the rats continued to consume fluid deliveries during the last phase—something that they would not have done if completely satiated. They also routinely consumed about 5 ml of water in their home cages after the session. For this reason, we simply refer to the last phase of the session as the disengaged phase.
Relationship with licking and chewing
The absence of the GC µ during chewing/licking in response to intraoral delivery of fluids excludes an implication of this rhythm in gustatory processing. If the function of this rhythm were related to licking, one would expect to observe this oscillation during periods after fluid administration, similar to the way that olfactory occurs in tight correlation to sniffing. Although high-amplitude neural rhythms in the olfactory system seem intrinsically related to sensory inputs (see, for instance, Fontanini et al. 2003; Haberly 1997; Kay et al. 1996), the occurrence of the GC µ is simply not related to the physical or chemosensory processing of tastants.
Orofacial rhythms are not without neural correlates in gustatory cortex, however: oscillations related to tongue/mouth movements are reflected in the activity of single units in gustatory cortex of awake, tasting rats (Katz et al. 2001a). Such activity has little impact on the LFPs, probably because it comes from only a small subsample of GC neurons. As a group, GC neurons show little resonance in the 7 to 12 Hz range (see Fig. 7A of Katz et al. 2001a) during performance of a tasting task, further suggesting that the observed rhythms are unrelated to licking.
Functional interpretation
What then is the function, if any, performed by the GC µ? By using a task in which a clear behavioral dissociation can be observed as rats first apply themselves to, and then withdraw from, a timed lever-press task, we are able to provide a preparation in which the two theories make testably distinct predictions, and our results are robust: in our hands GC µ reflects a coherent, bilateral cortical state related to disengagement from task orientation.
Although it is difficult to dissociate "idling" from epilepsy, several facts argue against the strong conclusion that our rats were experiencing full-blown seizures. First, late lever presses were frequently preceded by trials with little rhythmic activity, and thus it was not the oscillations that made rats "unable" to respond (see Fig. 5); it did not appear that GC µ directly delayed responses. Second, our rats licked and swallowed tastants delivered during oscillations, rather than letting fluid drip out of their mouths; this result, analogous to earlier work in somatosensory cortex (Wiest and Nicolelis 2003), distinguishes the behavior of normal rats amid a GC µ episode from that typically observed during epileptic seizures (Cooper and Upton 1978; Snead 1995). Most tellingly, the relationship between GC µ and behavior observed here is precisely that observed in research on normal, nonepileptic humans, who demonstrate poor task orientation during episodes of 7 to 12 Hz somatosensory (Slobounov et al. 2000) or visual (Molle et al. 2002) rhythms; this contrasts with the successful task performance that accompanies cortical and hippocampal rhythms (Sederberg et al. 2003; Semba and Komisaruk 1984). The occurrence of 7 to 12 Hz rhythms is related to loss of focus or attention, but is not necessarily pathological.
Nonetheless, we would suggest that the hard demarcation of a boundary between such mammalian rhythms and seizures is uncalled for. It may be possible to conceptualize absence seizure as a condition of "poorly controlled" normal rhythmicity. Full-blown absence seizures might therefore be described as extreme cases of disengagement, abnormal in duration and in its relationship with contingencies, and thus to be consistent with the rhythms observed here.
In conclusion, by bilaterally recording gustatory cortical activity in both hemispheres while simultaneously observing easily defined patterns of fluid task-related (and unrelated) behavior in rats, our data provide novel insight into the nature of rhythmic activity in the brain. We have demonstrated the ubiquity of rhythmic phenomena in the 7 to 12 Hz range, and have extended our understanding of their function in a manner consistent with the extant literature. It appears that 7 to 12 Hz rhythms consistently arise—during sleep (i.e., spindles) (Kandel and Buzsaki 1997; Steriade and Contreras 1998), quiet resting (Nicolelis et al. 1995), or absence seizures (Snead 1995)—when, for one reason or another, the animal is subject to relatively low intensities of sensory input. We argue that the 7 to 12 Hz oscillation might therefore reflect the intrinsic dynamics of cortical loops, a systemwide process involving the emergence of large populations of neurons into a dynamic state (Fanselow and Nicolelis 1999; Hartmann and Bower 1998). Sensory activation results in the inhibition of the 7 to 12 Hz rhythm, temporarily promoting a shift of the thalamocortical system toward a more stimulus oriented, activated state.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: A. Fontanini, Volen National Center for Complex Systems, MS 013, Brandeis University, 415 South St., Waltham, MA 02454 (E-mail: alfredof{at}brandeis.edu)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Bermejo R, Houben D, and Zeigler HP. Optoelectronic monitoring of individual whisker movements in rats. J Neurosci Methods 83: 89–96, 1998.[CrossRef][ISI][Medline]
Bressler SL. Relation of olfactory bulb and cortex. I. Spatial variation of bulbocortical interdependence. Brain Res 409: 285–293, 1987.[CrossRef][ISI][Medline]
Buzsaki G and Draguhn A. Neuronal oscillations in cortical networks. Science 304: 1926–1929, 2004.
Buzsaki G, Smith A, Berger S, Fisher LJ, and Gage FH. Petit mal epilepsy and parkinsonian tremor: hypothesis of a common pacemaker. Neuroscience 36: 1–14, 1990.[CrossRef][ISI][Medline]
Church RM, Meck WH, and Gibbon J. Application of scalar timing theory to individual trials. J Exp Psychol Anim Behav Process 20: 135–155, 1994.[CrossRef][ISI][Medline]
Cooper IS and Upton AR. Effects of cerebellar stimulation on epilepsy, the EEG and cerebral palsy in man. Electroencephalogr Clin Neurophysiol Suppl 349–354, 1978.
Destexhe A and Sejnowski TJ. Thalamocortical Assemblies. Oxford, UK: Oxford Univ. Press, 2001.
Fanselow EE and Nicolelis MA. Behavioral modulation of tactile responses in the rat somatosensory system. J Neurosci 19: 7603–7616, 1999.
Fontanini A, Spano P, and Bower JM. Ketamine-xylazine-induced slow (<1.5 Hz) oscillations in the rat piriform (olfactory) cortex are functionally correlated with respiration. J Neurosci 23: 7993–8001, 2003.
Freeman WJ. Neurodynamics: An Exploration in Mesoscopic Brain Dynamics. London: Springer-Verlag, 2000.
Gastaut H. Electrocorticographic study of the reactivity of rolandic rhythm. Rev Neurol (Paris) 87: 176–182, 1952.[Medline]
Haberly LB. Olfactory cortex. In: The Synaptic Organization of the Brain, edited by Shepherd GM. New York: Oxford Univ. Press, 1997, p. 377–416.
Hartmann MJ and Bower JM. Oscillatory activity in the cerebellar hemispheres of unrestrained rats. J Neurophysiol 80: 1598–1604, 1998.
Jain AK and Dubes RC. Algorithms for Clustering Data. Upper Saddle River, NJ: Prentice Hall, 1988.
Johanson M, Revonsuo A, Chaplin J, and Wedlund JE. Level and contents of consciousness in connection with partial epileptic seizures. Epilepsy Behav 4: 279–285, 2003.[CrossRef][ISI][Medline]
Kandel A and Buzsaki G. Cellular-synaptic generation of sleep spindles, spike-and-wave discharges, and evoked thalamocortical responses in the neocortex of the rat. J Neurosci 17: 6783–6797, 1997.
Katz DB, Simon SA, and Nicolelis MA. Dynamic and multimodal responses of gustatory cortical neurons in awake rats. J Neurosci 21: 4478–4489, 2001a.
Katz DB, Simon SA, and Nicolelis MA. Taste-specific neuronal ensembles in the gustatory cortex of awake rats. J Neurosci 22: 1850–1857, 2002.
Katz DB, Simon SA, and Nicolelis MAL. Electrophysiological studies of gustation in awake rats. In: Methods and Frontiers in the Chemical Senses, edited by Simon SA and Nicolelis MAL. Boca Raton, FL: CRC Press, 2001b, p. 339–357.
Kay LM, Lancaster LR, and Freeman WJ. Reafference and attractors in the olfactory system during odor recognition. Int J Neural Syst 7: 489–495, 1996.[CrossRef][ISI][Medline]
Krupa DJ, Wiest MC, Shuler MG, Laubach M, and Nicolelis MA. Layer-specific somatosensory cortical activation during active tactile discrimination. Science 304: 1989–1992, 2004.
Molle M, Marshall L, Fehm HL, and Born J. EEG theta synchronization conjoined with alpha desynchronization indicate intentional encoding. Eur J Neurosci 15: 923–928, 2002.[CrossRef][ISI][Medline]
Nicolelis MAL, Baccala LA, Lin RCS, and Chapin JK. Sensorimotor encoding by synchronous neural ensemble activity at multiple levels of the somatosensory system. Science 268: 1353–1358, 1995.
Niedermeyer E. The normal EEG of the waking adult. In: Electroencephalography: Basic Principles, Clinical Applications, and Related Fields (3rd ed.), edited by Niedermeyer E and Lopes da Silva F. Baltimore, MD: Williams & Wilkins, 1993, p. 131–152.
Nikouline VV, Wikstrom H, Linkenkaer-Hansen K, Kesaniemi M, Ilmoniemi RJ, and Huttunen J. Somatosensory evoked magnetic fields: relation to pre-stimulus mu rhythm. Clin Neurophysiol 111: 1227–1233, 2000.[CrossRef][ISI][Medline]
Pfurtscheller G, Neuper C, Andrew C, and Edlinger G. Foot and hand area mu rhythms. Int J Psychophysiol 26: 121–135, 1997.[CrossRef][ISI][Medline]
Phillips MI and Norgren R. A rapid method for permanent implantation of an intraoral fistula in rats. Behav Res Method Instr 2: 124, 1970.
Robinson PF and Gilmore SA. Spontaneous generalized spike-wave discharges in the electrocorticograms of albino rats. Brain Res 201: 452–458, 1980.[CrossRef][ISI][Medline]
Sederberg PB, Kahana MJ, Howard MW, Donner EJ, and Madsen JR. Theta and gamma oscillations during encoding predict subsequent recall. J Neurosci 23: 10809–10814, 2003.
Semba K and Komisaruk BR. Neural substrates of two different rhythmical vibrissal movements in the rat. Neuroscience 12: 761–774, 1984.[CrossRef][ISI][Medline]
Semba K, Szechtman H, and Komisaruk BR. Synchrony among rhythmical facial tremor, neocortical "alpha " waves, and thalamic non-sensory neuronal bursts in intact awake rats. Brain Res 195: 281–298, 1980.[CrossRef][ISI][Medline]
Shaw FZ. Is spontaneous high-voltage rhythmic spike discharge in Long Evans rats an absence-like seizure activity? J Neurophysiol 91: 63–77, 2004.
Slobounov SM, Fukada K, Simon R, Rearick M, and Ray W. Neurophysiological and behavioral indices of time pressure effects on visuomotor task performance. Brain Res Cogn Brain Res 9: 287–298, 2000.[CrossRef][Medline]
Snead OCI. Basic mechanisms of generalized absence seizures. Ann Neurol 37: 146–157, 1995.[CrossRef][ISI][Medline]
Steriade M. Corticothalamic resonance, states of vigilance and mentation. Neuroscience 101: 243–276, 2000.[CrossRef][ISI][Medline]
Steriade M and Contreras D. Spike-wave complexes and fast components of cortically generated seizures. I. Role of neocortex and thalamus. J Neurophysiol 80: 1439–1455, 1998.
Tiihonen J, Kajola M, and Hari R. Magnetic mu rhythm in man. Neuroscience 32: 793–800, 1989.[CrossRef][ISI][Medline]
Vergnes M, Marescaux C, Depaulis A, Micheletti G, and Warter JM. Spontaneous spike and wave discharges in thalamus and cortex in a rat model of genetic petit mal-like seizures. Exp Neurol 96: 127–136, 1987.[CrossRef][ISI][Medline]
Welch P. The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short, modified periodograms. IEEE Trans Audio Electroacoust 15: 70–73, 1967.[CrossRef]
Welsh JP, Lang EJ, Sugihara I, and Llinás R. Dynamic organization of motor control within the olivocerebellar system. Nature 374: 453–457, 1995.[CrossRef][Medline]
Wiest MC and Nicolelis MA. Behavioral detection of tactile stimuli during 7–12 Hz cortical oscillations in awake rats. Nat Neurosci 6: 913–914, 2003.[CrossRef][ISI][Medline]
This article has been cited by other articles:
|
|
 |
|
  F.-Z. Shaw 7-12 Hz High-Voltage Rhythmic Spike Discharges in Rats Evaluated by Antiepileptic Drugs and Flicker Stimulation J Neurophysiol, January 1, 2007; 97(1): 238 - 247. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Fontanini and D. B. Katz State-Dependent Modulation of Time-Varying Gustatory Responses J Neurophysiol, December 1, 2006; 96(6): 3183 - 3193. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Wang, A. Fontanini, and D. B. Katz Temporary basolateral amygdala lesions disrupt acquisition of socially transmitted food preferences in rats Learn. Mem., November 1, 2006; 13(6): 794 - 800. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-O. Polack and S. Charpier Intracellular activity of cortical and thalamic neurones during high-voltage rhythmic spike discharge in Long-Evans rats in vivo J. Physiol., March 1, 2006; 571(2): 461 - 476. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Alonso Idling With the {micro} Rhythm? Focus on "7 to 12 Hz Activity in Rat Gustatory Cortex Reflects Disengagement From a Fluid Self-Administration Task" J Neurophysiol, May 1, 2005; 93(5): 2372 - 2373. [Full Text] [PDF]
|
|
|
|
 |
|
 Highlights from the Literature Physiology, April 1, 2005; 20(2): 81 - 85. [Full Text] [PDF]
|
|
| ||||||
5 s after lever press in the 4-s data point.