Content area
Time is a fundamental and critical factor in daily life. Millisecond timing, which is the underlying temporal processing for speaking, dancing, and other activities, is reported to rely on the cerebellum. In this review, we discuss the cerebellar spike-coding mechanisms for temporal processing. Although the contribution of the cerebellum to both classical conditioning and voluntary movements is well known, the difference of the mechanisms for temporal processing between classical conditioning and voluntary movements is not clear. Therefore, we review the evidence of cerebellar temporal processing in studies of classical conditioning and voluntary movements and report the similarities and differences between them. From some studies, which used tasks that can change some of the temporal properties (e.g., the duration of interstimulus intervals) with keeping identical movements, we concluded that classical conditioning and voluntary movements may share a common spike-coding mechanism because simple spikes in Purkinje cells decrease at predicted times for responses regardless of the intervals between responses or stimulation.[PUBLICATION ABSTRACT]
Cerebellum (2014) 13:651658
DOI 10.1007/s12311-014-0580-5
REVIEW
Spike-Coding Mechanisms of Cerebellar Temporal Processing in Classical Conditioning and Voluntary Movements
Kenji Yamaguchi & Yoshio Sakurai
Published online: 2 July 2014# Springer Science+Business Media New York 2014
Abstract Time is a fundamental and critical factor in daily life. Millisecond timing, which is the underlying temporal processing for speaking, dancing, and other activities, is reported to rely on the cerebellum. In this review, we discuss the cerebellar spike-coding mechanisms for temporal processing. Although the contribution of the cerebellum to both classical conditioning and voluntary movements is well known, the difference of the mechanisms for temporal processing between classical conditioning and voluntary movements is not clear. Therefore, we review the evidence of cerebellar temporal processing in studies of classical conditioning and voluntary movements and report the similarities and differences between them. From some studies, which used tasks that can change some of the temporal properties (e.g., the duration of interstimulus intervals) with keeping identical movements, we concluded that classical conditioning and voluntary movements may share a common spike-coding mechanism because simple spikes in Purkinje cells decrease at predicted times for responses regardless of the intervals between responses or stimulation.
Keywords Temporal processing . Cerebellum . Eyeblink conditioning . Voluntary movement . Purkinje cell . Simple spike
Introduction
Time is fundamental and critical for various activities ranging from the sleepwakefulness cycle to walking, speaking, and playing music and sports. Humans and other organisms including lower-order species can engage in such timed activities because of the processing of temporal information across a wide range of timings [1]. Previous studies [2, 3] reported that mammalian temporal processing has been categorized into five timescales: circadian, ultradian, interval, millisecond, and microsecond timings (Fig. 1). These time ranges have been suggested to be governed by different neuronal mechanisms [1, 2, 5]. Among the different timing functions, millisecond temporal processing has been reported to rely on the cerebellum; although, the detailed neuronal mechanisms remain largely unclear [1].
Most studies of cerebellar function have focused on motor learning [6] and compensation for movement [7]. However, some studies have instead suggested that the cerebellum performs purely cognitive functions [811]. For example, the cerebellar cortex has been reported to display activation during engagement in timing perception tasks that required no motor control [12].
In particular, temporal processing is probably more relevant to the cerebellum than any other cognitive function. Performance of a task with temporal constraints was disrupted by impairing Purkinje cell plasticity, which is critical for cerebellar cortical computation [13], whereas several cognitive functions (e.g., sociability, spatial navigation, fear conditioning, and general anxiety) were not affected by the impairment of various types of cerebellar plasticity [14]. For this reason, the cerebellar cortex may be particularly critical for cognitive tasks that require precise temporal accuracy [13]. In addition, answering which event, continuous or discrete, is dominant in the cerebellar timing function is also important [15]. Concerning this problem, impairment and stimulation
K. Yamaguchi (*) : Y. SakuraiThe Department of Psychology, the Graduate School of Letters, Kyoto University, Yoshida-honmachi, Sakyou-ku, Kyoto 606-8501, Japane-mail: [email protected]
K. YamaguchiJapan Society for the Promotion of Science, Tokyo, Japan
652 Cerebellum (2014) 13:651658
Fig. 1 Schematic illustrations of the mammalian timescales of different types of timing. Circadian, ultradian, interval, millisecond, and microsecond timings are approximately 24 h, a few hours, a few seconds to hours, 101,000 ms, and sub-milliseconds in duration, respectively.
Each box contains an explanation about the related behaviors, brain structures, and mechanisms. The abscissa is aligned on a log scale. Data from Mauk and Buonomano (2004) [2], Buhusi et al. (2002) [1], and Miyamoto et al. (2012) [4] were cited in this figure
studies [16, 17] reported that the cerebellum was only involved in discrete events. However, another impairment [18] and recent electrophysiological [19] studies suggested that the cerebellum is important for both continuous and discrete events. Therefore, the question remains open for discussion.
The evidence that millisecond timing is associated with the cerebellum was often obtained using functional magnetic resonance imaging (fMRI) [20, 21] and brain stimulation [17, 22] in human subjects. These studies have described the detailed profiles of the millisecond timing related to the cerebellum. Fewer studies (e.g., [19, 23]) have concurrently reported the relationship between temporal processing of several hundred milliseconds and the cerebellar neuronal spikes that may directly organize behavioral timing. These studies used tasks that could control temporal properties with identical motor movements to assess the temporal substrates of the cerebellum without intervention by movement factors. In this review, we propose that the timing of conditioned responses in classical conditioning and voluntary movements may be based on a common mechanism of neural coding for cerebellar temporal processing. We review the evidence illustrating such neural coding in the cerebellum and discuss the mechanism underlying the coding.
Basic Circuits in the Cerebellum for Temporal Processing
Purkinje cells are the largest neurons in the cerebellar cortex, and they mediate the sole outputs of the cortex [24]. Purkinje cells have two types of excitatory synapses, one from parallel fibers (via mossy fibers and granule cells) and the other from climbing fibers (via inferior olive neurons) [25]. Only one climbing fiber innervates each Purkinje cell, and over 100,000 parallel fibers connect with a single Purkinje cell [24]. When a Purkinje cell receives inputs from a climbing fiber, it generates complex spikes with a frequency of 12 Hz [26]. On another front, when a Purkinje cell receives inputs from parallel fibers, it generates simple spikes [27]. Most background activity of the simple spikes is generated intrinsically by Purkinje cells with frequencies of up to 100 Hz [28]. Wise et al. [29] suggested that temporal patterns of simple spikes may be responsible for cerebellar functions for temporal computation because more than 90 % of cerebellar cortical spikes are simple spikes.
In David Marrs model [30], a climbing-fiber firing reflects cerebral instruction for an elemental movement, and mossy fibers send signals regarding context, i.e., the climbing-fiber firing, to the receiving Purkinje cell. Repetitive firings of the
Cerebellum (2014) 13:651658 653
climbing fiber and mossy fibers let the Purkinje cell learn such contexts. Eventually, the Purkinje cell fires by the occurrence of the context alone, which then causes the next element movement. This means that skilled motor learning is realized by potentiation of Purkinje cell synapses after a successful movement. On the other hand, James Albus model [31] indicated that skilled motor learning is accomplished by depression of Purkinje cell synapses after an unsuccessful movement to reduce such movement. The studies of Ito [25, 32] developed this, demonstrating that repetitive climbing-fiber inputs to Purkinje cell synapses in synchronization with parallel fiber inputs lead to a long-lasting depression of transmission from the parallel fibers to the Purkinje cell (long-term depression (LTD)). A study by Koekkoek et al. [33] suggested that learning-dependent timing is mediated by this parallel fiberPurkinje cell synaptic LTD because they observed that the timing of conditioned responses was disrupted when the parallel fiberPurkinje cell synaptic LTD was blocked. This is in agreement with some theoretical studies of cerebellar learning [3436] in that the parallel fiberPurkinje cell synaptic LTD can be considered a basic mechanism of cerebellar temporal processing.
Cerebellar Spike Coding for Temporal Processing: Classical Conditioning
After Brogden and Gantt reported the essential role of the cerebellum in learning and memory [37], the relation between the cerebellum and classical conditioning has often been demonstrated particularly using eyeblink reflexes (lesion/inactivation [33, 3845]; electrophysiological recording [23, 4657]). Regarding the generation of the timing of classical eyeblink conditioning, a series of studies demonstrated that the oscillatory properties of behavioral and neuronal activity tune a type of clock for proper timing in alert cats [5861]. The study of Domingo et al. (1997) [58] suggested that a <20-Hz oscillator underlies the generation of a conditioned eyeblink response and reflex. The subsequent study of Trigo et al. (1999) [59] demonstrated that orbicularis oculi motoneurons activity oscillated at 20 Hz during the conditioned eyeblink response and reflex movements, and the study of Snchez-Campusano et al. (2007) [60] demonstrated that deep cerebellar nucleus neuronal activity showed antiphasic oscillation with eyeblink movements during the acquisition of learning. Finally, the study of Snchez-Campusano et al. (2009) [61] used a nonlinear association analysis and time-dependent causality method to assess causal directionalities in the cerebellarmotoneuron network associations during classical eyeblink conditioning. They concluded that default states are determined by the oscillation of orbicularis oculi motoneurons and that the cerebellum plays a role in modulating/reinforcing motor learning. As evidences of the notable contribution of the cerebellar
cortex to eyeblink conditioning, several studies using electro-physiological recording demonstrated that the simple spikes of Purkinje cells exhibited excitatory or inhibitory responses before the presentation of unconditioned stimuli in well-trained animals [23, 47, 5157]. In this section, we introduce and discuss the findings of the spike-coding mechanism of cerebellar temporal processing for classical conditioning.
Some studies [23, 33, 40, 41, 51, 52, 5557] assessed the changes of the behavioral and/or cerebellar responses while controlling the interstimulus intervals (ISIs) between conditioned and unconditioned stimuli. In those studies, Hesslows group described Purkinje cell activity that relates to generation of the timing [23, 52, 53, 5557]. A study published in 2007 [52] tracked Purkinje cell activity during the acquisition, extinction, and reacquisition of classical eyeblink conditioning using direct stimulation of mossy fibers as the conditioned stimulus and stimulation of climbing fibers, inferior olive neurons, or periocular skin as the unconditioned stimulus. The study demonstrated that the responses of Purkinje cells to the conditioned stimulus markedly decreased toward the end of ISIs between conditioned and unconditioned stimuli in the animals that had acquired conditioning. This pattern was also confirmed in subsequent studies [23, 53, 5557]. Moreover, Purkinje cells responded to the conditioned stimulus earlier during conditioning with shorter ISIs (200 ms) than during that with longer ISIs (300 ms).
A study in 2011 [23] focused on temporal properties and examined the changes in the responses of Purkinje cells following the control of ISIs and conditioned stimulus duration. This study reported that the offsets of most responses of Purkinje cells were time-locked with ISI offsets regardless of the length of conditioned stimulus duration. It was concluded that the cells conditioned with longer ISIs had a longer latency to response offsets. Furthermore, the latencies to response onset and peak as well as their offset increased following increases in ISIs (Fig. 2). The latencies to onset, offset, and peak were significantly correlated with the length of ISIs, and the response duration was weakly correlated with the ISI length. In addition, the effects of the shift from shorter to longer ISIs were assessed, and the results indicated that Purkinje cells could change their response timing to adapt to new ISIs. Finally, the study concluded that the offset of Purkinje cell responses elicited the overt ocular response offset.
In addition to these findings, a study published in 2014 [57] evaluated the often suggested underlying mechanism, parallel fiberPurkinje cell synaptic LTD, for the classical eyeblink conditioning. The study indicated the discrepancy between the fact that overt conditioned responses were not elicited with short ISIs and that the parallel fiberPurkinje cell synaptic LTD, a synaptic learning mechanism in the conditioning [33], would be expected to occur at short ISIs. Therefore, the study assessed the responses of Purkinje cells during conditioning
654 Cerebellum (2014) 13:651658
Fig. 2 Average conditioned Purkinje cell responses for different inter-stimulus intervals (ISIs). Each line shows the average Purkinje cell response without unconditioned stimulus presentation after conditioning with 200, 300, and 500 ms ISIs (extra-thick, thin, and thick lines, respectively). Intervals between conditioned and unconditioned stimuli are represented in gray shading (darker, 200 ms; middle, 300 ms; brighter, 500 ms). The conditioned stimulus duration (bar below graph) is shown all together. The vertical axis presents the percentages of average background activity over 40 trials. Republished with permission of Springer Science+Business Media, from Fig. 2 in Jirenhed and Hesslow (2011) [23]; permission conveyed through Copyright Clearance Center, Inc
with ISIs shorter or longer than 100 ms. The results illustrated that Purkinje cell activity decreased during conditioning with longer than 150-ms ISIs as demonstrated in previous studies [23, 52, 53, 55, 56] but increased during conditioning with 50-ms ISIs. The study concluded that the synaptic mechanism underlying conditioning remained unclear because the parallel fiberPurkinje cell synaptic LTD could not explain the behavioral phenomenon. Thus, the authors expressed agreement with the other probable mechanisms: long-term potentiation (LTP) driven by molecular layer interneurons, combination of plasticity [62], or the effects of more subtle aspects of the stimulus that have not yet been recognized.
Cerebellar Spike Coding for Temporal Processing: Voluntary Movement
The contribution of the cerebellum to voluntary movements is well documented (reviewed in [6, 63]), and some previous studies have reported the relationship between the execution of well-timed voluntary movements and cerebellar neuronal spikes (e.g., [6466]). However, these studies observed only the timing of animals spontaneous movements, and the temporal properties of the movements were not controlled strictly. To clarify the involvement of more cognitive aspects of
temporal processing in the cerebellar computation, behavioral tasks that control temporal properties should be used. Although classical conditioning studies [23, 52] indicated that Purkinje cells can adapt to changes of temporal properties during learning, it is unclear whether the features of such neural responses are completely identical also during voluntary movements that may include some types of intention. In this section, we introduce the recent findings of experiments that controlled some of the temporal properties in voluntary movement tasks.
A study by Ohmae et al. [19] demonstrated that the cerebellar dentate nucleus plays a major role in the prediction of sensory events with voluntary eye movements. To make monkeys encode temporal information for prediction, the authors trained them to detect a single omission of isochronous repetitive stimuli with saccade responses. The repetitive stimulus was simultaneous blinking of the visual and auditory objects, and ISIs of blinking were 100, 200, 300, 400, or 600 ms. As a result of the single-neuron recording from the dentate nucleus during this task, dentate nucleus neurons exhibited oscillatory responses in synchronization with the blinking stimulus (Fig. 3a). The response intensity was gradually increased during the time course (Fig. 3c) regardless of the direction of saccade targets. The degrees of the intensity were positively correlated with the length of ISIs (Fig. 3b, d). These results indicate that the cerebellar dentate nuclei can encode temporal prediction, a type of cognitive function, for voluntary movements, and ISIs used in this study are consistent with the range of millisecond timing, which may be deeply associated with the cerebellum (Fig. 1). The authors concluded that these timing signals were generated by the cerebellar cortex, which has been identified as the structure producing timing-specific signals in some simulation studies [2, 35]. The spatiotemporal patterns of the simple spikes of Purkinje cells deeply affect cerebellar nucleus neuron discharges in the way of inhibition [6668].
As an attempt to assess the temporal processing for voluntary movements, we introduce our recent study [69]. This study developed two novel tasks to explore a function that counts the absolute time between discrete events and at the same time another function that measures continuous interevent intervals that have a rhythmic sequence. These two functions, referred to as duration-based and beat-based timings, respectively, in the study of Grube et al. [17], used almost identical movements in our study. Our tasks required the rats to touch a switch continuously at regular intervals of a few hundred milliseconds with their paw. When the rats failed to pause for a fixed interval between behavioral responses, the current trial was canceled and restarted. Hence, the rats needed to perceive the intervals between the touch responses to successfully complete the trials. Behavioral timing in a task that requires only one interval to complete one trial was defined as duration-based
Cerebellum (2014) 13:651658 655
Fig. 3 Dentate nucleus neuron responses during an oddball detection task for different interstimulus intervals (ISIs). a Examples of the sequential responses of one cell for different ISIs. The small dots below each line indicate the time of stimulus presentation. 1st stim and Oddball vertical lines denote the timing of the first stimulus and discriminative stimulus for saccade (i.e., oddball) in each trial, respectively. b The response amplitude differences between ISIs. The population neural activity (n=86) during intervals between the stimulus immediately before oddball (Last stim in graph) and the appearance of oddball is shown by different ISIs. Dotted traces of each line show the response for 100 ms after the appearance of oddball. The baseline indicates average firing rate of 500 ms before the first stimulus. c The average degrees of firing modulation during trials for different ISIs. The plots to the extreme right
indicate the responses to the stimuli immediately before oddball. d The relationship between the degrees of firing modulation and ISIs at the timing of the first and last stimuli. 1st stim (black circles) and Last stim (red circles) denote the mean and 95 % confidence intervals of the response amplitude at the timing of the first stimulus and a stimulus immediately before oddball according to the ISI, respectively. The circles connected by solid and dotted lines indicate data from 86 cells with a transient decrease and 23 cells with a transient increase in activity during intervals between each stimulus, respectively. ad Republished with permission of SOCIETY FOR NEUROSCIENCE, from Fig. 3 in Ohmae et al. (2013) [19]; permission conveyed through Copyright Clearance Center, Inc
timing, and that requiring multiple intervals was defined as beat-based timing. This study indicated that those tasks could make rats expect the controlled fixed intervals precisely and change their behavior between the duration- and beat-based timing tasks. Our preliminary study [70] reported the result of a recording of multiple Purkinje cells during the aforementioned beat-based timing task. Simple spikes paused near the time of the touch only when the rat could pause for fixed intervals (Fig. 4). This suggests that Purkinje cells measure motor timing by pausing simple spikes at the time when the movements are accomplished. The simple spikes are known to display regular activity patterns that include successive firing and short or long pauses, and such regularity is believed to control downstream deep cerebellar nuclei [71]. Considering the fact that the patterns of simple spike activity sometimes synchronize among adjacent Purkinje cells [72], the pattern
presented in Fig. 4 appears to fit into the literature evidence that simple spikes exhibit regular pauses during synchronization among adjacent Purkinje cells. Precise temporal control in the cerebellum may be realized by such regularity of pauses, and as indicated by prior studies [73, 74], that regularity may mediate the rate coding by simple spike, which is directly responsible for sending commands to motor systems. Given that context, the observations by Ohmae et al. [19] and our group [70] about the mechanisms of cerebellar timing coding are consistent. However, our preliminary study had the possibility that the simple spike activity reflected only a pushpull kinetic function rather than a timing function because we analyzed the activity while the rats were moving their paw. On the other hand, the study of Ohmae et al. (2013) demonstrated that monkeys maintained their sight to a central fixed point until the time of the discriminative stimuli presentation,
656 Cerebellum (2014) 13:651658
Fig. 4 Population Purkinje cell activity when the rat waited for 300-ms intervals between behavioral responses. This figure shows raster plots and peristimulus time histograms of simple spikes from Purkinje cells (n=7). In this session, the rat had to touch the target four times while maintaining an interval of at least 300 ms between the touches. The top graph (2nd touch) shows the simple spike firing during intervals 300 ms before and 200 ms after the second touch for the successful trials. Similarly, the activities around the third and fourth touches are indicated as 3rd touch and 4th touch, respectively. The bottom graph (Error touch) presents the activity during intervals 300 ms before and 200 ms after the touch that could not be maintained for fixed intervals (erroneous trials). Data of the first touch of the trials are not shown because the first touch was not needed to maintain the fixed interval. Data were taken from one rat and one session (1 h). Each bin is 3 ms. The statistical test revealed that the numbers of simple spike firings for 8 ms around the time of touch were significantly different between touches that could be maintained the fixed intervals (2nd, 3rd, and 4th touches) and that of Error touch (Wilcoxon rank-sum test; **p<0.01, *p<0.05)
and the changes in deep cerebellar nuclear activity were shown only after the onset of the isochronous repetitive stimuli presentation. An additional experiment that includes control conditions is needed for our study in a future work to rule out such a possibility.
Cerebellar Spike Coding for Temporal Processing: General Mechanism
Taken together, we described some common points between the studies of classical conditioning and voluntary movements. Then, we discussed those points and the remaining problems.
The common point of the temporal processing studies of classical conditioning and voluntary movements is the pause of neuronal firing when the animals are processing the timing. As described above, several classical conditioning studies [23, 47, 5153, 5557] demonstrated that Purkinje cells exhibited a pause or clear suppression of firing during ISIs, the periods
between the conditioned and unconditioned stimuli. A voluntary movement study [70] similarly reported the firing pause of simple spikes at the time of interval offset (Fig. 4). In the study by Ohmae et al. [19], 79 % of recorded dentate nucleus neurons displayed increasing responses to the repetitive stimuli (Fig. 3a). This explains the presence of a pause of the simple spikes of Purkinje cells for temporal processing because dentate nucleus activity may reflect disinhibition from Purkinje cellular inhibitory inputs [67, 68]. Moreover, when ISIs shifted, Purkinje cells could alter the temporal properties of their activity to adapt to other ISIs during both conditioned responses in classical conditioning [23, 52] and voluntary movements [19]. In addition, ISIs used in the aforementioned studies were similar (100600 ms), and both classical conditioning [57] and voluntary movements [19] studies reported weaker neural responses when ISIs shorter than 100 ms were used. All of these results suggest a mechanism of cerebellar temporal processing in which simple spikes decrease their firing toward predicted timing.
However, such mechanisms cannot be identified because the evidence from studies of the voluntary movements is limited. In addition, it has not been clarified why the simple spike response in our study [70] was not oscillatory as ob-served in the study by Ohmae et al. [19]. To solve this problem, additional experiments are needed. Studies of the cerebellar temporal processing for voluntary movements have recently started. Although such studies included serious difficulty in eliminating motor factors, it will be important to clarify the mechanism of temporal processing over short ranges of time for intentional acts. As Wetmore et al. [57] indicated, the synaptic mechanisms remain unclear. Theoretical studies that simulate the phenomenon observed in the experiment of Wetmore et al. are also needed.
Acknowledgments This work was supported by JSPS KAKENHI grant nos. 24243069 and 24223004 (Y.S.).
Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References
1. Buhusi CV, Meck WH. What makes us tick? Functional and neural mechanisms of interval timing. Nat Rev Neurosci. 2005;6:75565.
2. Mauk MD, Buonomano DV. The neural basis of temporal processing. Ann Rev Neurosci. 2004;27:30740.
3. Delgado-Garca JM, Grau C, DeFeudis P, del Pozo F, Jimenez JM, Delgado JM. Ultradian rhythms in the mobility and behavior of rhesus monkeys. Exp Brain Res. 1976;25:7991.
4. Miyamoto H, Nakamaru-Ogiso E, Hamada K, Hensch TK. Serotonergic integration of circadian clock and ultradian sleep-wake cycles. J Neurosci. 2012;32:14794803.
Cerebellum (2014) 13:651658 657
5. Hinton SC, Meck WH. The internal clocks of circadian and interval timing. Endeavour. 1997;21:827.
6. Thach WT, Goodkin HP, Keating JG. The cerebellum and the adaptive coordination of movement. Ann Rev Neurosci. 1992;15:40342.
7. Manni E, Petrosini L. Lucianis work on the cerebellum a century later. Trends Neurosci. 1997;20:1126.
8. Gao JH, Parsons LM, Bower JM, Xiong J, Li J, Fox PT. Cerebellum implicated in sensory acquisition and discrimination rather than motor control. Science. 1996;272:5457.
9. Ito M. Control of mental activities by internal models in the cerebellum. Nat Rev Neurosci. 2008;9:30413.
10. D'Angelo E, Casali S. Seeking a unified framework for cerebellar function and dysfunction: from circuit operations to cognition. Front Neural Circuits. 2013. doi:http://dx.doi.org/10.3389/fncir.2012.00116
Web End =10.3389/fncir.2012.00116 .
11. Tomlinson SP, Davis NJ, Bracewell RM. Brain stimulation studies of non-motor cerebellar function: a systematic review. Neurosci Biobehav Rev. 2013;37:76689.
12. Xu D, Liu T, Ashe J, Bushara KO. Role of the olivocerebellar system in timing. J Neurosci. 2006;26:59905.
13. Rahmati N, Owens CB, Bosman LW, Spanke JK, Lindeman S, Gong W, et al. Cerebellar potentiation and learning a whisker-based object localization task with a time response window. J Neurosci. 2014;34: 194962.
14. Galliano E, Potters JW, Elgersma Y, Wisden W, Kushner SA, De Zeeuw CI, et al. Synaptic transmission and plasticity at inputs to murine cerebellar Purkinje cells are largely dispensable for standard nonmotor tasks. J Neurosci. 2013;33:12599618.
15. Ivry RB, Spencer RM, Zelaznik HN, Diedrichsen J. The cerebellum and event timing. Ann N Y Acad Sci. 2002;978:30217.
16. Spencer RM, Zelaznik HN, Diedrichsen J, Ivry RB. Disrupted timing of discontinuous but not continuous movements by cerebellar lesions. Science. 2003;300:14379.
17. Grube M, Lee KH, Griffiths TD, Barker AT, Woodruff PW. Transcranial magnetic theta-burst stimulation of the human cerebellum distinguishes absolute, duration-based from relative, beat-based perception of subsecond time intervals. Front Psychol. 2010. doi:http://dx.doi.org/10.3389/fpsyg.2010.00171
Web End =10. http://dx.doi.org/10.3389/fpsyg.2010.00171
Web End =3389/fpsyg.2010.00171 .
18. Bo J, Block HJ, Clark JE, Bastian AJ. A cerebellar deficit in sensorimotor prediction explains movement timing variability. J Neurophysiol. 2008;100:282532.
19. Ohmae S, Uematsu A, Tanaka M. Temporally specific sensory signals for the detection of stimulus omission in the primate deep cerebellar nuclei. J Neurosci. 2013;33:1543241.
20. Dreher JC, Grafman J. The roles of the cerebellum and basal ganglia in timing and error prediction. Eur J Neurosci. 2002;16:160919.
21. Lewis PA, Miall RC. Brain activation patterns during measurement of sub- and supra-second intervals. Neuropsychologia. 2003;41: 158392.
22. Koch G, Oliveri M, Torriero S, Salerno S, Lo Gerfo E, Caltagirone C. Repetitive TMS of cerebellum interferes with millisecond time processing. Exp Brain Res. 2007;179:2919.
23. Jirenhed DA, Hesslow G. Learning stimulus intervalsadaptive timing of conditioned Purkinje cell responses. Cerebellum. 2011;10:52335.
24. Ito M. Input and output pathways in the cerebellar cortex. In: Ito M, editor. The cerebellum: brain for an implicit self. Upper Saddle Rive: FT Press Science; 2011. p. 4551.
25. Ito M. Long-term depression. Ann Rev Neurosci. 1989;12:85102.26. Eccles JC, Llins R, Sasaki K. The excitatory synaptic action of climbing fibres on the Purkinje cells of the cerebellum. J Physiol. 1966;182:26896.
27. Thach WT. Somatosensory receptive fields of single units in cat cerebellar cortex. J Neurophysiol. 1967;30:67596.
28. Cerminara NL, Rawson JA. Evidence that climbing fibers control an intrinsic spike generator in cerebellar Purkinje cells. J Neurosci. 2004;24:45107.
29. Wise AK, Cerminara NL, Marple-Horvat DE, Apps R. Mechanisms of synchronous activity in cerebellar Purkinje cells. J Physiol. 2010;588:237390.
30. Marr D. A theory of cerebellar cortex. J Physiol. 1969;202:43770.31. Albus JS. A theory of cerebellar function. Math Biosci. 1971;10:2561.32. Ito M, Sakurai M, Tongroach P. Climbing fibre induced depression of both mossy fibre responsiveness and glutamate sensitivity of cerebellar Purkinje cells. J Physiol. 1982;324:11334.
33. Koekkoek SKE, Hulscher HC, Dortland BR, Hensbroek RA, Elgersma Y, Ruigrok TJH, et al. Cerebellar LTD and learning-dependent timing of conditioned eyelid responses. Science. 2003;301:17369.
34. Buonomano DV, Mauk MD. Neural network model of the cerebellum: temporal discrimination and the timing of motor responses. Neural Comput. 1994;6:3855.
35. Medina JF, Garcia KS, Nores WL, Taylor NM, Mauk MD. Timing mechanisms in the cerebellum: testing predictions of a large-scale computer simulation. J Neurosci. 2000;20:551625.
36. Honda T, Yamazaki T, Tanaka S, Nagao S, Nishino T. Stimulus-dependent state transition between synchronized oscillation and randomly repetitive burst in a model cerebellar granular layer. PLoS Comput Biol. 2011;7:e1002087.
37. Brogden WJ, Horsley GW. Intraneural conditioning: cerebellar conditioned reflexes. Arch Neurol Psychiatr. 1942;48:43755.
38. McCormick DA, Guyer PE, Thompson RF. Superior cerebellar peduncle lesions selectively abolish the ipsilateral classically conditioned nictitating membrane/eyelid response of the rabbit. Brain Res. 1982;244:34750.
39. Clark GA, McCormick DA, Lavond DG, Thompson RF. Effects of lesions of cerebellar nuclei on conditioned behavioral and hippocampal neuronal responses. Brain Res. 1984;291:12536.
40. Welsh JP, Harvey JA. Cerebellar lesions and the nictitating membrane reflex: performance deficits of the conditioned and unconditioned response. J Neurosci. 1989;9:299311.
41. Yeo CH, Lobo DH, Baum A. Acquisition of a new-latency conditioned nictitating membrane responsemajor, but not complete, dependence on the ipsilateral cerebellum. Learn Mem. 1997;3:55777.
42. Rogers RF, Fender AF, Steinmetz JE. The cerebellum is necessary for rabbit classical eyeblink conditioning with a non-somatosensory (photic) unconditioned stimulus. Behav Brain Res. 1999;104:10512.
43. Attwell PJ, Cooke SF, Yeo CH. Cerebellar function in consolidation of a motor memory. Neuron. 2002;34:101120.
44. Bao S, Chen L, Kim JJ, Thompson RF. Cerebellar cortical inhibition and classical eyeblink conditioning. Proc Natl Acad Sci U S A. 2002;99:15927.
45. Robleto K, Thompson RF. Extinction of a classically conditioned response: red nucleus and interpositus. J Neurosci. 2008;28:26518.
46. McCormick DA, Thompson RF. Cerebellum: essential involvement in the classically conditioned eyelid response. Science. 1984;223:2969.
47. Berthier NE, Moore JW. Cerebellar Purkinje cell activity related to the classically conditioned nictitating membrane response. Exp Brain Res. 1986;63:34150.
48. Katz DB, Steinmetz JE. Single-unit evidence for eye-blink conditioning in cerebellar cortex is altered, but not eliminated, by interpositus nucleus lesions. Learn Mem. 1997;4:88104.
49. Tracy JA, Thompson JK, Krupa DJ, Thompson RF. Evidence of plasticity in the pontocerebellar conditioned stimulus pathway during classical conditioning of the eyeblink response in the rabbit. Behav Neurosci. 1998;112:26785.
50. Rogers RF, Britton GB, Steinmetz JE. Learning-related interpositus activity is conserved across species as studied during eyeblink conditioning in the rat. Brain Res. 2001;905:1717.
51. Kotani S, Kawahara S, Kirino Y. Purkinje cell activity during learning a new timing in classical eyeblink conditioning. Brain Res. 2003;994: 193202.
658 Cerebellum (2014) 13:651658
52. Jirenhed DA, Bengtsson F, Hesslow G. Acquisition, extinction, and reacquisition of a cerebellar cortical memory trace. J Neurosci. 2007;27:2493502.
53. Rasmussen A, Jirenhed DA, Hesslow G. Simple and complex spike firing patterns in Purkinje cells during classical conditioning. Cerebellum. 2008;7:5636.
54. Zbarska S, Bloedel JR, Bracha V. Cerebellar dysfunction explains the extinction-like abolition of conditioned eyeblinks after NBQX injections in the inferior olive. J Neurosci. 2008;28:1020.
55. Svensson P, Jirenhed DA, Bengtsson F, Hesslow G. Effect of conditioned stimulus parameters on timing of conditioned Purkinje cell responses. J Neurophysiol. 2010;103:132936.
56. Jirenhed DA, Hesslow G. Time course of classically conditioned Purkinje cell response is determined by initial part of conditioned stimulus. J Neurosci. 2011;31:90704.
57. Wetmore DZ, Jirenhed DA, Rasmussen A, Johansson F, Schnitzer MJ, Hesslow G. Bidirectional plasticity of Purkinje cells matches temporal features of learning. J Neurosci. 2014;34:17317.
58. Domingo JA, Gruart A, Delgado-Garca JM. Quantal organization of reflex and conditioned eyelid responses. J Neurophysiol. 1997;78: 251830.
59. Trigo JA, Gruart A, Delgado-Garca JM. Discharge profiles of abducens, accessory abducens, and orbicularis oculi motoneurons during reflex and conditioned blinks in alert cats. J Neurophysiol. 1999;81:166684.
60. Snchez-Campusano R, Gruart A, Delgado-Garca JM. The cerebellar interpositus nucleus and the dynamic control of learned motor responses. J Neurosci. 2007;27:662032.
61. Snchez-Campusano R, Gruart A, Delgado-Garca JM. Dynamic associations in the cerebellar-motoneuron network during motor learning. J Neurosci. 2009;29:1075063.
62. Gao Z, van Beugen BJ, De Zeeuw CI. Distributed synergistic plasticity and cerebellar learning. Nat Rev Neurosci. 2012;13:61935.
63. Ito M. Voluntary motor control. In: Ito M, editor. The cerebellum: brain for an implicit self. Upper Saddle River: FT Press Science; 2011. p. 1508.
64. Welsh JP, Lang EJ, Suglhara I, Llins R. Dynamic organization of motor control within the olivocerebellar system. Nature. 1995;374: 4537.
65. Kitazawa S, Kimura T, Yin PB. Cerebellar complex spikes encode both destinations and errors in arm movements. Nature. 1998;392: 4947.
66. Witter L, Canto CB, Hoogland TM, de Gruijl JR, De Zeeuw CI. Strength and timing of motor responses mediated by rebound firing in the cerebellar nuclei after Purkinje cell activation. Front Neural Circuits. 2013;7:133.
67. Chaumont J, Guyon N, Valera AM, Dugu GP, Popa D, Marcaggi P, et al. Clusters of cerebellar Purkinje cells control their afferent climbing fiber discharge. Proc Natl Acad Sci U SA. 2013;110:162238.68. De Zeeuw CI, Hoebeek FE, Bosman LW, Schonewille M, Witter L, Koekkoek SK. Spatiotemporal firing patterns in the cerebellum. Nat Rev Neurosci. 2011;12:32744.
69. Yamaguchi K, Sakurai Y. Novel behavioral tasks to explore cerebellar temporal processing in milliseconds in rats. Behav Brain Res. 2014;263:13843.
70. Yamaguchi K, Takahashi S, Sakurai Y. Cerebellar Purkinje-cell spikes interaction during a timing behavior task in the rat. In: 42th Society for Neuroscience Annual Meeting, New Orleans; 2012. Program No. 809.12.
71. Shin SL, Hoebeek FE, Schonewille M, De Zeeuw CI, Aertsen A, De Schutter E. Regular patterns in cerebellar Purkinje cell simple spike trains. PLoS One. 2007;2:e485.
72. Shin SL, De Schutter E. Dynamic synchronization of Purkinje cell simple spikes. J Neurophysiol. 2006;96:348591.
73. De Schutter E, Steuber V. Patterns and pauses in Purkinje cell simple spike trains: experiments, modeling and theory. Neuroscience. 2009;162:81626.
74. Cao Y, Maran SK, Dhamala M, Jaeger D, Heck DH. Behavior-related pauses in simple-spike activity of mouse Purkinje cells are linked to spike rate modulation. J Neurosci. 2012;32:867885.
Springer Science+Business Media New York 2014