TMP BF41
TMP BF41
TMP BF41
Behavioral/Systems/Cognitive
Neurons can engage in synchronized oscillatory activity in the gamma-frequency range when responding to sensory stimuli. Both the
oscillatory patterning and the synchronization of responses are enhanced with arousal and attention or when the electroencephalogram
is activated by electrical stimulation of the mesencephalic reticular formation. Here we show with intracortical application of cholinergic
antagonists that the enhancement of gamma oscillations and response synchronization is mediated by acetylcholine and muscarinic
receptors. We demonstrate further that coapplication of cholinergic agonists with synchrony-inducing light stimuli causes a lasting
increase in the probability that the stimulated cells engage in gamma oscillations and response synchronization. These changes develop
slowly over tens of minutes and then persist for many hours. Thus, cholinergic modulation plays a crucial role both in the fast, state-
dependent facilitation of gamma oscillations and response synchronization and in use-dependent long-term modifications of cortical
dynamics that favor synchronization of responses in the gamma-frequency range.
Key words: gamma oscillations; response synchronization; cat visual cortex; cholinergic antagonists; cholinergic agonists; mesencephalic
reticular formation
from nonspecific thalamic nuclei (for review, see Steriade et al., sustained part of the visual response (200 – 4500 msec). Autocorrelation
1997). and cross-correlation histograms were calculated from individual trials
Here we investigate the hypothesis that cholinergic mecha- with a resolution of 1 msec and time shifts of ⫾80 msec. Shuffled cross-
nisms play a crucial role in facilitating the following: (1) visually correlograms (shift predictors) were routinely computed to test for cor-
relations resulting from stimulus locking. These shift predictors were flat,
induced oscillations in the gamma-frequency range, (2) the asso-
and hence no correction of the cross-correlograms was required. For
ciated precise synchronization of cortical responses, and (3) the quantitative analysis, correlations were fitted by damped cosine (Gabor)
related use-dependent long-term modifications of cortical re- functions (König, 1994). This allowed determination of the strength of
sponse properties. oscillatory modulation and synchronization from the ratios of the am-
plitudes of the first satellite peak and the central peak, respectively, over
Materials and Methods the offset of the fitted function: MAS (modulation amplitude of the first
Anesthesia and recording. Anesthesia was induced with ketamine (Ketan- satellite peak) and MA (modulation amplitude of the central peak). A
est; 10 mg/kg, i.m.; Parke-Davis, Courbevoie, France) and xylacine correlogram was considered to reveal significant oscillatory (or synchro-
(Rompun; 2 mg/kg, i.m.; Bayer, Leverkusen, Germany) and maintained nized) activity if the first satellite (or central) peak had a Z score ⬎2 and
with a mixture of 70% N2O and 30% O2 supplemented by 0.4 – 0.8% if MAS or MA was ⱖ0.1. The incidence of oscillation (OI) and synchro-
halothane. Heart rate, end-tidal CO2, and temperature were monitored nization (SI) was expressed as the percentage of responses exhibiting
continuously. Paralysis was obtained with pancuronium bromide (Pan- significant oscillations and synchronization, respectively, during baseline
curonium; 0.15 mg 䡠 kg ⫺1 䡠 hr ⫺1; Organon, Oberschleissheim, Ger- and test epochs. Power spectra of LFPs and EEGs were computed with a
many). Multi-unit (MU) and local field potential (LFP) responses were fast Fourier transform algorithm, normalized for the signal energy be-
recorded in primary visual cortex (A17) with up to six varnish-coated tween 1 and 120 Hz, and analyzed in six different frequency bands: 1– 4
tungsten electrodes placed at different distances (200 m to 3 mm) from (delta), 4 – 8 (theta), 8 –14 (alpha), 14 –21 (beta), 21– 48 (low gamma),
the drug application. EEGs were recorded from small silver ball elec- and 52–70 (high gamma) Hz. Results presented here refer to the theta
trodes placed on the visual cortex. MU activity was bandpass filtered and lower gamma bands, which showed the most significant changes. For
from 1 to 3 kHz, spikes were detected with a Schmitt trigger whose group analysis of MU and LFP responses, mean values from the different
threshold was set to twice the noise level, and time stamps were acquired postejection epochs were compared with the mean values from the base-
with a precision of 2 sec. LFP and EEG signals were bandpass filtered line. As onset latency of the delayed effects, we considered the postejec-
from 5–100 and 1–100 Hz, respectively, and digitized at 1 kHz (1401Plus; tion time window in which a significant change was first observed. All
Cambridge Electronics Design, Cambridge, UK). To analyze the effects comparisons were made between baseline and test epochs of identical
of scopolamine, MU and LFP recordings were obtained from 20 and 23 duration. Group data were compared by the Wilcoxon signed rank test
sites, respectively. To examine the effects of cholinergic agonists, MU and (Statview version 5.0.1; SAS Institute, Cary, NC).
LFP activity was recorded from 22 sites and 14 sites, respectively. These
studies were performed in six animals: four animals with agonists and
four with scopolamine (two of them with agonist and antagonist appli-
Results
cation). In two of these animals, the hemisphere contralateral to scopol- In nine lightly anesthetized, paralyzed cats, we simultaneously
amine application was also studied (MU activity from five sites; LFPs recorded MU activity and LFPs with arrays of microelectrodes
from six sites). Electrodes were positioned at the same anteroposterior and the EEG with silver ball electrodes from primary visual cortex
stereotaxic coordinates as in the ipsilateral hemisphere (4 – 6 mm from (A17). Visual stimulation was performed with moving gratings
the application site). In two animals, MRF stimulation was combined covering the receptive fields of all recording sites, and the orien-
with scopolamine application (MUs and LFPs from 10 sites). Control tation of stimuli was adjusted to coactivate as many of the re-
experiments were performed in three additional animals in which ago- corded neurons as possible. In addition, the MRF was electrically
nists were applied in the absence (MUs from 14 sites; LFPs from 28 sites)
stimulated before the presentation of light stimuli to enhance
or in the presence (MUs from 10 sites; LFP from 16 sites) of visual
stimulation.
gamma oscillations and response synchronization and to deter-
MRF stimulation. Bipolar concentric stimulation electrodes were bi- mine whether this enhancement is mediated by cholinergic re-
laterally placed in the MRF at Horsley Clark coordinates A2, H8, L ⫾ 2. ceptors. The muscarinic cholinergic antagonist scopolamine or
Stimuli consisted of trains (60 –100 msec) of five constant current pulses the cholinergic agonists acetylcholine and carbachol were applied
(100 –165 sec at 75 Hz, 0.25–2 mA) that were applied 150 –300 msec intracortically with iontophoresis through multibarrel pipettes in
before the onset of visual responses. The location of the electrodes was the vicinity of the recording electrodes. We assessed the firing rate
adjusted to produce maximal facilitation of cortical evoked potentials with peristimulus time histograms (PSTHs), the oscillatory pat-
elicited by electrical stimulation (50 sec, 1–2 mA) of the optic chiasm terning of responses from the MAS of autocorrelograms, and the
(A14.5, H5, L ⫾ 3). synchronization from the MA in cross-correlograms. In addition,
Iontophoresis. Drugs were diluted in distilled water and were applied
we computed the power spectra of LFP signals (see Materials and
with seven-barrel pipettes: acetylcholine chloride (2 M, pH 4.5), carba-
chol (1 M, pH 4), and scopolamine (0.2 M, pH 4). The entire tip diameter
Methods).
was 8 –15 m. Continuous ejection was achieved for 20 –50 min at a
current of 100 –250 nA. Retention currents (15 nA) with opposite sign Scopolamine application
were used to control drug leakage. Two different pharmacological pro- As described previously (Gray et al., 1989), light stimuli evoked
tocols were used: the first consisted of baseline measurements without MU and LFP responses that exhibited an oscillatory patterning in
and with MRF stimulation and scopolamine application without and the gamma-frequency range and were synchronized with zero
with MRF stimulation (every epoch, 300 trials, 50 min) (see Fig. 1); the phase lag (Fig. 1 A). As shown previously (Munk et al., 1996;
second consisted of a baseline, followed by two epochs of agonists appli- Herculano-Houzel et al., 1999), MRF stimulation enhanced this
cation (carbachol and/or acetylcholine) and multiple postejection ep- oscillatory patterning and the synchronization ( p ⬍ 0.05; Wil-
ochs (every epoch, 300 trials, 50 min) (see Fig. 2 A).
coxon signed rank test) of light responses (Fig. 1, compare A, B).
Visual stimulation and data analysis. Moving gratings were used as
visual stimulation and were presented on a 21 inch computer monitor at This is reflected by an increased regularity and persistence of
100 Hz refresh rate. Stimuli were presented for 4500 msec and repeated oscillatory discharges in cross-correlograms computed from MU
every 10 sec. For quantitative assessment of correlation functions from responses, by greater regularity and amplitude of oscillations in
MU responses and of power spectra from LFP and EEG recordings, data the LFP, and by sharpening of the gamma-band peak in the
from 10 consecutive trials were pooled. Analysis was confined to the power spectrum of the LFP. Scopolamine application strongly
Rodriguez et al. • Cholinergic Control of Gamma Oscillations J. Neurosci., November 17, 2004 • 24(46):10369 –10378 • 10371
Conditioning-dependent changes in
firing rate
Analysis of firing rates revealed that the
conditioning-dependent increases in oscil-
latory modulation and synchronization
were unrelated to changes in response am-
plitude (Fig. 8 B, D). On average, the dis-
charge rates of preconditioning and post-
conditioning responses were similar ( p ⬎
0.6; n ⫽ 22) (Fig. 8 A). This confirms pre-
vious results that have also shown that
there is no systematic relationship between
the probability of neurons to engage in
synchronous and/or oscillatory firing and
the actual discharge rate (Herculano-
Houzel et al., 1999).
Previous experiments have shown that
the variance of light responses to succes-
sive, identical stimuli gets reduced when
neurons engage in synchronized gamma
oscillations (Herculano-Houzel et al.,
1999). To test whether the enhancement of
gamma oscillations induced by pairing
cholinergic and visual stimulation also
lead to a reduction of response variability, Figure 4. Control experiments. Effects of carbachol (Cch) application without and with visual stimulation (VS) on gamma
we computed the SD of the firing rates (in oscillations in MU responses. A, Autocorrelograms of light-evoked MU responses computed before (left) and after (right) carba-
chol had been applied twice without visual stimulation. The displayed autocorrelograms are averaged from 10 trials at time points
spikes per second) of successively evoked indicated by the asterisks in the central plot, which shows the time course of the oscillatory modulation (MAS). Note the absence
MU responses and of the gamma power of of any significant oscillatory modulation before and after cholinergic stimulation. B, Data from the same recording site, obtained
associated LFP responses before and after before and after carbachol had been applied in conjunction with visual stimulation. Note the massive increase of gamma oscil-
pairing. As the examples in Figure 8 B–E lations (MAS) in the autocorrelograms, starting 75 min after the second conditioning. C, Mean changes in oscillation incidence and
show, pairing reduced the variance of strength for a single case for which three subsequent protocols could be obtained. Note that, in both protocols with unpaired
gamma power (C, E) and discharge rates visual and cholinergic stimulation, oscillations were weakened, whereas the interleaved protocol with paired stimulation led to a
(B, D) in successively recorded light re- massive increase.
sponses. Both effects were associated with
an overall increase in gamma power in LFPs (C, E) but showed no as revealed by intracellular recordings from parietal cortex (Steriade
correlation with overall changes in discharge rate (B, D). At the et al., 1993). However, the latter study was performed with systemic
group level (Fig. 8 F, G), the SD of gamma power and of firing application of muscarinic and nicotinic antagonists. One possibility
rates was reduced by 48% ( p ⬍ 0.005) and 45% ( p ⬍ 0.0003), is that blockade of muscarinic receptors disabled cortical gamma
respectively. oscillations and thereby prevented the manifestations of MRF effects
that might themselves be mediated by noncholinergic transmitter
Discussion systems. Candidates are the monoaminergic projections and the glu-
The present results demonstrate two distinct cholinergically me- tamatergic activation of noncholinergic cells in the basal forebrain.
diated effects on gamma oscillations and response synchroniza- The other possibility is that the MRF effects are actually mediated by
tion that act with different time constants but are closely related enhanced release of acetylcholine in the cortex. We favor the latter
to one another. The effects of scopolamine application indicate hypothesis for two reasons. First, recent in vitro experiments revealed
that a steady-state activation of muscarinic receptors is a neces- that carbachol facilitates gamma oscillations in slices of the hip-
sary prerequisite for the oscillatory patterning and synchroniza- pocampus (Fisahn et al., 1998) and cerebral cortex (Buhl et al.,
tion of visual responses in the gamma frequency range. Further- 1998). Second, in vivo data show that activation of cholinergic cells in
more, the data show that the facilitatory effects that MRF the basal forebrain with neurotensin leads to enhanced gamma ac-
stimulation has on gamma oscillations and response synchroni- tivity in the cortical EEG (Cape et al., 2000). These cells are also
zation are abolished by blocking muscarinic receptors. A compat- activated by MRF stimulation via a glutamatergic pathway (Khateb
ible effect was described for power shifts in EEGs and burst firing et al., 1997).
Rodriguez et al. • Cholinergic Control of Gamma Oscillations J. Neurosci., November 17, 2004 • 24(46):10369 –10378 • 10375
Figure 6. Conditioning effects on spike oscillations after normalization against EEG power.
Oscillation strength (MAS) was normalized for the gamma power in corresponding EEG record-
ings, and effects were determined after conditioning (Post-ejection) for the three control ex-
periments with carbachol (Cch) application either with (black; n ⫽ 6) or without (gray; n ⫽ 7)
contingent visual stimulation (VS). Error bars indicate 1 SEM. The MAS/gamma ratio of 1 corre-
sponds to the mean of the baseline measurements before conditioning.
Figure 7. Effect of distance between drug application and recording site. A, B, Scopolamine
(Scop) action on oscillation strength (MAS) in autocorrelograms of MU responses at a nearby (A;
Figure 5. LFP frequency spectra from the same recording site and experimental sequence as ⱕ200 m) and a remote (B; ⬃2000 m) recording site. MAS reduction was faster and stron-
in Figure 4. A, Effects of applying carbachol (Cch) without visual stimulation (VS). Note that ger in sites recorded closer to the application site ( A) than in sites placed more distant to the
agonist application induces gamma oscillations in the absence of visual stimulation, the fre- scopolamine application ( B). * indicates epochs without scopolamine and MRF stimulation. C,
quency being lower than that of light-induced gamma oscillations. Note also that agonists did D, Enhancement of oscillation strength (MAS) by conditioning at a nearby (C; ⱕ200 m) and a
not cause a lasting facilitation of gamma oscillations. B, Averaged frequency spectra of LFP remote (D; ⬃1500 m) recording site. Note that latency and strength of gamma oscillations
oscillations induced by light (baseline) and induced by carbachol application without visual enhancement are similar at both sites. Cch, Carbachol; VS, visual stimulation.
stimulation. Box plots represent median values (line within box), quartiles (top and bottom
lines of box), and the 10th and 90th percentiles. C, Effect of applying agonist in conjunction with
light stimulation on LFP visual responses. Note the delayed increase of gamma oscillations after ergic agonists cannot mimic the specific action of the mixed
conditioning, also correlated with the MU from the same electrode displayed in Figure 4 B. GABAergic and cholinergic projections from the basal forebrain
to the cortex because of unselective activation of both synaptic
and extrasynaptic muscarinic and nicotinic receptors at glutama-
The effects of agonist application differed for spontaneous tergic and GABAergic neurons. Another nonexclusive possibility
activity and light-evoked responses. The analysis of spontaneous is that the occurrence of gamma oscillations and precise response
activity in the LFP showed that agonist application without visual synchronization require cooperativity among spatially distrib-
stimulation enhanced spontaneous oscillations, which is in uted cortical columns and that topically restricted application of
agreement with the findings cited above. However, agonist appli- cholinergic agonists had not affected a sufficiently large array of
cation did not enhance the oscillatory patterning and the syn- neurons. Furthermore, the enhancement of the oscillatory pat-
chronization of light-evoked responses as was to be expected terning and the synchronization of light-evoked responses may
from the effects of MRF stimulation. This may have several rea- require the concerted action of additional, noncholinergic cofac-
sons. First, the widespread and sustained application of cholin- tors. Candidates are the numerous other modulatory systems
10376 • J. Neurosci., November 17, 2004 • 24(46):10369 –10378 Rodriguez et al. • Cholinergic Control of Gamma Oscillations
Evidence indicates that cholinergic modulation of cortical ac- Gruber T, Keil A, Muller MM (2001) Modulation of induced gamma band
tivity is essential for a large variety of cognitive processes (Everitt responses and phase synchrony in a paired associate learning task in the
human EEG. Neurosci Lett 316:29 –32.
and Robbins, 1997; Sarter and Bruno, 1997). However, it is un-
Gruber T, Muller MM, Keil A (2002) Modulation of induced gamma band
clear how exactly the numerous and very heterogeneous effects responses in a perceptual learning task in the human EEG. J Cognit Neu-
that acetylcholine has on the different classes of cortical neurons rosci 14:732–744.
support these cognitive processes. The present data suggest as Herculano-Houzel S, Munk MH, Neuenschwander S, Singer W (1999) Pre-
unifying mechanism the short- and long-term facilitation of os- cisely synchronized oscillatory firing patterns require electroencephalo-
cillations and response synchronization in the gamma-frequency graphic activation. J Neurosci 19:3992– 4010.
range. Cognitive functions that depend on cholinergic control Karni A, Sagi D (1993) The time course of learning a visual skill. Nature
365:250 –252.
comprise the ability to focus attention and to retain stimulus
Khateb A, Fort P, Williams S, Serafin M, Jones BE, Mühlethaler M (1997)
features in both short- and long-term memory (Everitt and Rob- Modulation of cholinergic nucleus basalis neurons by acetylcholine and
bins, 1997; Sarter and Bruno, 1997). Interestingly, all of these N-methyl-D-aspartate. Neuroscience 81:47–55.
functions have been shown to be associated with enhanced syn- König P (1994) A method for the quantification of synchrony and oscilla-
chronization of responses in the gamma-frequency range (Tiiti- tory properties of neuronal activity. J Neurosci Methods 54:31–37.
nen et al., 1993; Tallon-Baudry et al., 1998; Miltner et al., 1999; Kopell N, Ermentrout GB, Whittington MA, Traub RD (2000) Gamma
Müller et al., 2000; Fries et al., 2001; Gruber et al., 2001, 2002), rhythms and beta rhythms have different synchronization properties.
Proc Natl Acad Sci USA 97:1867–1872.
suggesting that these functions depend on response synchroniza-
Lisman J (1989) A mechanism for the Hebb and the anti-Hebb processes
tion. Thus, the present data suggest, as one unifying mechanism underlying learning and memory. Proc Natl Acad Sci USA 86:9574 –9578.
for the heterogeneous effects of cholinergic hypofunction, a dis- Markram H, Lübke J, Frotscher M, Sakmann B (1997) Regulation of synap-
turbance of both short- and long-term mechanisms that are re- tic efficacy by coincidence of postsynaptic APs and EPSPs. Science
quired for the manifestation and stabilization of gamma oscilla- 275:213–215.
tions and response synchronization. Miltner WH, Braun C, Arnold M, Witte H, Taub E (1999) Coherence of
Our results predict that pathologies associated with cholin- gamma-band EEG activity as a basis for associative learning. Nature
397:434 – 436.
ergic hypofunction should be associated with a shift in EEG
Mioche L, Singer W (1989) Chronic recordings from single sites of kitten
power spectra toward lower frequencies, with reduced and less striate cortex during experience-dependent modifications of receptive-
precise response synchronization and with impairments of use- field properties. J Neurophysiol 62:185–197.
dependent plasticity. This should motivate a search for a correla- Mitzdorf U, Singer W (1978) Prominent excitatory pathways in the cat vi-
tion between EEG abnormalities indicative of impaired gamma sual cortex (A 17 and A 18): a current source density analysis of electrically
oscillations and reduced response synchronization on the one evoked potentials. Exp Brain Res 33:371–394.
hand and cognitive disturbances such as occur in Alzheimer’s Moruzzi G, Magoun HW (1949) Brain stem reticular formation and activa-
tion of the EEG. Electroencephalogr Clin Neurophysiol 1:455– 473.
disease and are suspected to be associated with cholinergic hypo-
Müller MM, Gruber T, Keil A (2000) Modulation of induced gamma band
function on the other hand. activity in the human EEG by attention and visual information process-
ing. Int J Psychophysiol 38:283–299.
Munk MH, Roelfsema PR, König P, Engel AK, Singer W (1996) Role of
References reticular activation in the modulation of intracortical synchronization.
Ahissar M, Hochstein S (1993) Attentional control of early perceptual
learning. Proc Natl Acad Sci USA 90:5718 –5722. Science 272:271–274.
Ahissar M, Hochstein S (1997) Task difficulty and the specificity of percep- Nowak LG, Sanchez-Vives MV, McCormick DA (1997) Influence of low
tual learning. Nature 387:401– 406. and high frequency inputs on spike timing in visual cortical neurons.
Azouz R, Gray CM (2000) Dynamic spike threshold reveals a mechanism Cereb Cortex 7:487–501.
for synaptic coincidence detection in cortical neurons in vivo. Proc Natl Pare D, Smith Y, Parent A, Steriade M (1988) Projections of brainstem core
Acad Sci USA 97:8110 – 8115. cholinergic and non-cholinergic neurons of cat to intralaminar and retic-
Bröcher S, Artola A, Singer W (1992) Agonists of cholinergic and noradren- ular thalamic nuclei. Neuroscience 25:69 – 86.
ergic receptors facilitate synergistically the induction of long-term poten- Perez-Orive J, Mazor O, Turner GC, Cassenaer S, Wilson RI, Laurent G
tiation in slices of rat visual cortex. Brain Res 573:27–36. (2002) Oscillations and sparsening of odor representations in the mush-
Buhl EH, Tamas G, Fisahn A (1998) Cholinergic activation and tonic exci- room body. Science 297:359 –365.
tation induce persistent gamma oscillations in mouse somatosensory cor- Rasmusson DD (2000) The role of acetylcholine in cortical synaptic plastic-
tex in vitro. J Physiol (Lond) 513:117–126. ity. Behav Brain Res 115:205–218.
Cape EG, Manns ID, Alonso A, Beaudet A, Jones BE (2000) Neurotensin- Rauschecker JP, Hahn S (1987) Ketamine-xylazine anaesthesia blocks con-
induced bursting of cholinergic basal forebrain neurons promotes solidation of ocular dominance changes in kitten visual cortex. Nature
gamma and theta cortical activity together with waking and paradoxical 326:183–185.
sleep. J Neurosci 20:8452– 8461. Rodriguez E, George N, Lachaux JP, Martinerie J, Renault B, Varela FJ (1999)
Dringenberg HC, Vanderwolf CH (1998) Involvement of direct and indi- Perception’s shadow: long-distance synchronization of human brain ac-
rect pathways in electrocorticographic activation. Neurosci Biobehav Rev tivity. Nature 397:430 – 433.
22:243–257. Roelfsema PR, Engel AK, König P, Singer W (1997) Visuomotor integration
Everitt BJ, Robbins TW (1997) Central cholinergic systems and cognition. is associated with zero time-lag synchronization among cortical areas.
Annu Rev Psychol 48:649 – 684. Nature 385:157–161.
Fisahn A, Pike FG, Buhl EH, Paulsen O (1998) Cholinergic induction of Sarter M, Bruno JP (1997) Cognitive functions of cortical acetylcholine: to-
network oscillations at 40 Hz in the hippocampus in vitro. Nature ward a unifying hypothesis. Brain Res Brain Res Rev 23:28 – 46.
394:186 –189. Schoups A, Vogels R, Qian N, Orban G (2001) Practising orientation iden-
Frank MG, Issa NP, Stryker MP (2001) Sleep enhances plasticity in the de- tification improves orientation coding in V1 neurons. Nature
veloping visual cortex. Neuron 30:275–287. 412:549 –553.
Fries P, Reynolds JH, Rorie AE, Desimone R (2001) Modulation of oscilla- Semba K, Fibiger HC (1989) Organization of central cholinergic systems.
tory neuronal synchronization by selective visual attention. Science Prog Brain Res 79:37– 63.
291:1560 –1563. Shulz DE, Sosnik R, Ego V, Haidarliu S, Ahissar E (2000) A neuronal ana-
Gray CM, König P, Engel AK, Singer W (1989) Oscillatory responses in cat logue of state-dependent learning. Nature 403:549 –553.
visual cortex exhibit inter-columnar synchronization which reflects Singer W (1999) Neuronal synchrony: a versatile code for the definition of
global stimulus properties. Nature 338:334 –337. relations? Neuron 24:49 – 65, 111–125.
10378 • J. Neurosci., November 17, 2004 • 24(46):10369 –10378 Rodriguez et al. • Cholinergic Control of Gamma Oscillations
Smith Y, Pare D, Deschenes M, Parent A, Steriade M (1988) Cholinergic and Tiitinen H, Sinkkonen J, Reinikainen K, Alho K, Lavikainen J, Naatanen R
non-cholinergic projections from the upper brainstem core to the visual (1993) Selective attention enhances the auditory 40-Hz transient re-
thalamus in the cat. Exp Brain Res 70:166 –180. sponse in humans. Nature 364:59 – 60.
Steriade M, Pare D, Parent A, Smith Y (1988) Projections of cholinergic Traub RD, Jefferys JG, Whittington MA (1997) Simulation of gamma
and non-cholinergic neurons of the brainstem core to relay and asso- rhythms in networks of interneurons and pyramidal cells. J Comput Neu-
ciational thalamic nuclei in the cat and macaque monkey. Neuro- rosci 4:141–150.
science 25:47– 67. Wainer BH, Mesulam MM (1990) Ascending cholinergic pathways in the
Steriade M, Amzica F, Nunez A (1993) Cholinergic and noradrenergic mod- rat brain. In: Brain cholinergic systems (Steriade M, Biesold D, eds), pp
ulation of the slow (⬃0.3 Hz) oscillation in neocortical cells. J Neuro- 65–119. Oxford: Oxford UP.
physiol 70:1385–1400. Weinberger NM, Bakin JS (1998) Learning-induced physiological memory
Steriade M, Amzica F, Contreras D (1996) Synchronization of fast (30 – 40 in adult primary auditory cortex: receptive fields plasticity, model, and
Hz) spontaneous cortical rhythms during brain activation. J Neurosci mechanisms. Audiol Neurootol 3:145–167.
16:392– 417. Wespatat V, Tennigkeit F, Singer W (2004) Phase sensitivity of synaptic
Steriade M, Jones BE, McCormick DA (1997) Thalamus. Organization and modifications in oscillating cells of rat visual cortex. J Neurosci 24:
function, Vol 1. Oxford: Elsevier. 9067–9075.
Tallon-Baudry C, Bertrand O, Peronnet F, Pernier J (1998) Induced Whittington MA, Traub RD, Jefferys JG (1995) Synchronized oscillations in
gamma-band activity during the delay of a visual short-term memory task interneuron networks driven by metabotropic glutamate receptor activa-
in humans. J Neurosci 18:4244 – 4254. tion. Nature 373:612– 615.